Exact match. Not showing close matches.
PICList
Thread
'[EE] Single Phase AC control via PWM (maybe a chal'
2005\01\08@122340
by
Dave Tweed
|
"Bob S." <spam_OUTop1cwkTakeThisOuT
punkass.com> wrote:
> I've been thinking of doing something like that, but with 2 mosfets or
> IGBT's inverted back to back (like 2 SCR's to do full wave AC triac
> control), ...
Beware the body diode! You might end up with essentially a dead short.
It's usually simpler to wrap a single MOSFET/IGBT with a bridge rectifier,
as Roland showed, if you can tolerate the two additional diode drops.
> ... with inductors on either side of the mosfets ...
Probably not necessary -- the leakage inductance of the transformer primary
is probably sufficient.
> ... and maybe a metalized poly cap in parallel with the mosfets to kill
> any fly back voltages.
Make sure you have a path for the stored energy (another diode?) and then
the only spikes you should see would be due to the switching time of the
diode.
> The reason I'd like to control an AC transformer this way are many fold.
All good reasons, but most people these days use an active PFC (power
factor corrector) circuit that dispenses with the big iron altogether.
The isolation is provided by a much smaller transformer that operates
at tens or hundreds of kHz instead of 50/60 Hz. You can get dedicated
PFC controller chips, or roll your own using a dsPIC (see Olin's recent
Circuit Cellar article). You get all the same benefits in a much smaller
and more efficient package.
-- Dave Tweed
2005\01\10@024242
by
Russell McMahon
|
>> I've been thinking of doing something like that, but with 2 mosfets
>> or
>> IGBT's inverted back to back (like 2 SCR's to do full wave AC triac
>> control), ...
A single MOSFET will already conduct in both directions when turned
on. ie for an N channel FET the gate must be positive relative to
source BUT the drain can swing positive or negative, BUT ...
> Beware the body diode! You might end up with essentially a dead
> short.
Indeed. The single FET will not be off in the reverse direction due to
the body diode conducting. If you want to use FETs you can still use
two, but you connect them in SERIES. The arrangement is very
unintuitive. Connect source to soource and gate to gate (!) and
connect AC path to be switched to the 2 drains. To turn on provide an
appropriate positive voltage to the interconnected gates relative to
the interconnected sources. This usually requires a floating supply -
which can be annoying. A 9v PP3 (transistor radio size) alkaline
battery should last 100 on hours here if switched at high frequency or
approaching shelf life if switched occasionally and slow turn off/on
is OK.
> It's usually simpler to wrap a single MOSFET/IGBT with a bridge
> rectifier,
> as Roland showed, if you can tolerate the two additional diode
> drops.
Probably true. Again, the gates float, but you can use a high
resistance high voltage feed. As you could in the arrangement above.
If doing this care needs to be taken to clamp gates so they don't
exceed Vgsmax.
RM
2005\01\10@131807
by
Bob S.
|
|
Thanks for your input, but....
What you describe as an "active PFC", sounds suspiciously like an SMPS.
Been there, tried that, it doesn't work very good for high amperage
(low voltage)variable output power supplies (unless your box is
basically a big aluminum heatsink).
I had a SMPS prototype made, and while it did work, it required a lot of
heatsinking for the output diodes, and even some heatsinks for the
switching mosfets. It wasn't much more compact, (wasn't linear either),
and by the time you got done with all the extra caps, heatsinks, and
what not, it weighed about the same as the 60hz transformer we're using.
Basically, I didn't see anything remotely "efficient" about it.
I'm also wanting to go with a plastic enclosure, as the interface
control I'm wanting to use requires it (no metal allowed).
Dave Tweed wrote:
{Quote hidden}> "Bob S." <
.....op1cwkKILLspam@spam@punkass.com> wrote:
>
>> I've been thinking of doing something like that, but with 2 mosfets or
>> IGBT's inverted back to back (like 2 SCR's to do full wave AC triac
>> control), ...
>
>
> Beware the body diode! You might end up with essentially a dead short.
>
> It's usually simpler to wrap a single MOSFET/IGBT with a bridge rectifier,
> as Roland showed, if you can tolerate the two additional diode drops.
>
>> ... with inductors on either side of the mosfets ...
>
>
> Probably not necessary -- the leakage inductance of the transformer primary
> is probably sufficient.
>
>> ... and maybe a metalized poly cap in parallel with the mosfets to kill
>> any fly back voltages.
>
>
> Make sure you have a path for the stored energy (another diode?) and then
> the only spikes you should see would be due to the switching time of the
> diode.
>
>> The reason I'd like to control an AC transformer this way are many fold.
>
>
> All good reasons, but most people these days use an active PFC (power
> factor corrector) circuit that dispenses with the big iron altogether.
> The isolation is provided by a much smaller transformer that operates
> at tens or hundreds of kHz instead of 50/60 Hz. You can get dedicated
> PFC controller chips, or roll your own using a dsPIC (see Olin's recent
> Circuit Cellar article). You get all the same benefits in a much smaller
> and more efficient package.
>
> -- Dave Tweed
2005\01\10@153053
by
Dave Tweed
|
|
"Bob S." <op1cwkKILLspampunkass.com> wrote:
> What you describe as an "active PFC", sounds suspiciously like an SMPS.
> Been there, tried that, it doesn't work very good for high amperage
> (low voltage)variable output power supplies (unless your box is
> basically a big aluminum heatsink).
>
> I had a SMPS prototype made, and while it did work, it required a lot of
> heatsinking for the output diodes, and even some heatsinks for the
> switching mosfets. It wasn't much more compact, (wasn't linear either),
> and by the time you got done with all the extra caps, heatsinks, and
> what not, it weighed about the same as the 60hz transformer we're using.
> Basically, I didn't see anything remotely "efficient" about it.
Yes, a PFC is an SMPS; specifically a boost regulator with a special
control algorithm. But the circuit you've been describing is also an
SMPS, and will have many of the same issues. It sounds like you got a
really bad design, and unfortunately, it's soured you on the whole
concept. I assure you, those results are not typical.
> I'm also wanting to go with a plastic enclosure, as the interface
> control I'm wanting to use requires it (no metal allowed).
??? Where are you going to hide that big transformer?
-- Dave Tweed
2005\01\10@173707
by
Martin K
Bob S. wrote:
> Thanks for your input, but....
>
> What you describe as an "active PFC", sounds suspiciously like an SMPS.
> Been there, tried that, it doesn't work very good for high amperage
> (low voltage)variable output power supplies (unless your box is
> basically a big aluminum heatsink).
Active PFC is used in low Vout power supplies all the time when followed
by any range of topologies.
Can you give some specs on your design? I bet myself or someone else
(Bob Blick, Olin) could tell you what was wrong with it.
--
Martin K
http://wwia.org/sgroup/biofuel/
2005\01\28@145323
by
Bob S.
|
|
Heh, the transformer isn't that big, as its output is very low voltage
(read 1.8 vac max.), and it typically only puts out 40 watts or so. It
IS capable of putting out upwards of maybe 80 watts under a dead short
fault condition on the outputs (without melting itself).
The output amperage is VERY high though (20+ amps RMS.), which will
excessively heat up any SMPS output diodes (hence the heat sinks). Even
if you were to use synchronous rectification with low R-on mosfets, your
still talking about 20+ amps of current.
Also from what I've read about them, SMPS's are not generally used in
variable output applications.
This method would only be slightly similar to an SMPS (mosfets and PWM),
as what the transformer would see is just a smooth 50/60hz AC wave form
(the small inductors take the high freq. pwm'd wave form and smooth it
back out). This would also run in series between the transformer and
the line, not converted to DC then remodulated to a high frequency AC
signal like an SMPS does. The output would be 50/60hz AC (not DC like
every SMPS I've ever heard of).
Controlling the power on/from the high voltage side means that you don't
have to deal with the high amperages (i.e. heat) that are seen on the
transformer output. The only heat generated, would be from the
switching losses, which at 49 to 52 khz, and a nominal 1/4 amp of
current, shouldn't get very hot at all I would think.
I don't think the designer of that prototype I mentioned earlier, was
either stupid or negligent in his design. I don't even know where in my
computer I would have those design specs, this was a couple of years ago.
At any rate, if you think you can design a variable output SMPS that can
output 300mv to 2v (AC or DC), can put out over 25 amps at the top end
(continuously), and NOT generate gobs of heat (must be able to be
enclosed in a plastic ABS box without melting it (ideally it should be
cool to the touch)), have only a small heatsink, and no fan, I'd have
say that would impress me.
Although our current power supply can draw about 40 watts during normal
operation, it is capable of drawing upwards of 80 watts without failing
(i.e., overbuilt, and for a reason). The current design also does not
generate hardly any heat (no heatsink), unless there is a severe fault
condition (a dead short near the transformer outputs, or asymmetrical
gating). The main problems with my current design, is that there is no
decent way to compensate for line voltage variations using phase
control, and the transformer is a tad too big to fit into a decent
sloped top box.
I had asked around to a few other people/firms, and the best response I
got was that it would cost me $10,000 just for the design & prototype. A
bit too pricey for my wallet (especially if it didn't work as advertised).
Martin K wrote:
>
>
> Bob S. wrote:
>
>> Thanks for your input, but....
>>
>> What you describe as an "active PFC", sounds suspiciously like an SMPS.
>> Been there, tried that, it doesn't work very good for high amperage
(low voltage)variable output power supplies (unless your box is
basically a big aluminum heatsink).
>
>
>
> Active PFC is used in low Vout power supplies all the time when
followed by any range of topologies.
> Can you give some specs on your design? I bet myself or someone else
(Bob Blick, Olin) could tell you what was wrong with it.
>
>Dave Tweed wrote:
>Yes, a PFC is an SMPS; specifically a boost regulator with a special
>control algorithm. But the circuit you've been describing is also an
>SMPS, and will have many of the same issues. It sounds like you got a
>really bad design, and unfortunately, it's soured you on the whole
>concept. I assure you, those results are not typical.
>
>> I'm also wanting to go with a plastic enclosure, as the interface
>> control I'm wanting to use requires it (no metal allowed).
>
>
>??? Where are you going to hide that big transformer?
-- Dave Tweed
2005\01\28@145323
by
Bob S.
|
|
Heh, the transformer isn't that big, as its output is very low voltage
(read 1.8 vac max.), and it typically only puts out 40 watts or so. It
IS capable of putting out upwards of maybe 80 watts under a dead short
fault condition on the outputs (without melting itself).
The output amperage is VERY high though (20+ amps RMS.), which will
excessively heat up any SMPS output diodes (hence the heat sinks). Even
if you were to use synchronous rectification with low R-on mosfets, your
still talking about 20+ amps of current.
Also from what I've read about them, SMPS's are not generally used in
variable output applications.
This method would only be slightly similar to an SMPS (mosfets and PWM),
as what the transformer would see is just a smooth 50/60hz AC wave form
(the small inductors take the high freq. pwm'd wave form and smooth it
back out). This would also run in series between the transformer and
the line, not converted to DC then remodulated to a high frequency AC
signal like an SMPS does. The output would be 50/60hz AC (not DC like
every SMPS I've ever heard of).
Controlling the power on/from the high voltage side means that you don't
have to deal with the high amperages (i.e. heat) that are seen on the
transformer output. The only heat generated, would be from the
switching losses, which at 49 to 52 khz, and a nominal 1/4 amp of
current, shouldn't get very hot at all I would think.
I don't think the designer of that prototype I mentioned earlier, was
either stupid or negligent in his design. I don't even know where in my
computer I would have those design specs, this was a couple of years ago.
At any rate, if you think you can design a variable output SMPS that can
output 300mv to 2v (AC or DC), can put out over 25 amps at the top end
(continuously), and NOT generate gobs of heat (must be able to be
enclosed in a plastic ABS box without melting it (ideally it should be
cool to the touch)), have only a small heatsink, and no fan, I'd have
say that would impress me.
Although our current power supply can draw about 40 watts during normal
operation, it is capable of drawing upwards of 80 watts without failing
(i.e., overbuilt, and for a reason). The current design also does not
generate hardly any heat (no heatsink), unless there is a severe fault
condition (a dead short near the transformer outputs, or asymmetrical
gating). The main problems with my current design, is that there is no
decent way to compensate for line voltage variations using phase
control, and the transformer is a tad too big to fit into a decent
sloped top box.
I had asked around to a few other people/firms, and the best response I
got was that it would cost me $10,000 just for the design & prototype. A
bit too pricey for my wallet (especially if it didn't work as advertised).
Martin K wrote:
>
>
> Bob S. wrote:
>
>> Thanks for your input, but....
>>
>> What you describe as an "active PFC", sounds suspiciously like an SMPS.
>> Been there, tried that, it doesn't work very good for high amperage
(low voltage)variable output power supplies (unless your box is
basically a big aluminum heatsink).
>
>
>
> Active PFC is used in low Vout power supplies all the time when
followed by any range of topologies.
> Can you give some specs on your design? I bet myself or someone else
(Bob Blick, Olin) could tell you what was wrong with it.
>
>Dave Tweed wrote:
>Yes, a PFC is an SMPS; specifically a boost regulator with a special
>control algorithm. But the circuit you've been describing is also an
>SMPS, and will have many of the same issues. It sounds like you got a
>really bad design, and unfortunately, it's soured you on the whole
>concept. I assure you, those results are not typical.
>
>> I'm also wanting to go with a plastic enclosure, as the interface
>> control I'm wanting to use requires it (no metal allowed).
>
>
>??? Where are you going to hide that big transformer?
-- Dave Tweed
2005\01\28@182759
by
olin_piclist
Bob S. wrote:
> IS capable of putting out upwards of maybe 80 watts under a dead short
If you can bend the laws of physics to get 80 watts into a dead short, you
don't need anyone's help here.
> Also from what I've read about them, SMPS's are not generally used in
> variable output applications.
This just proves that you shouldn't believe everything that's written.
*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
2005\01\29@000622
by
Martin Klingensmith
Bob S. wrote:
> At any rate, if you think you can design a variable output SMPS that
> can output 300mv to 2v (AC or DC), can put out over 25 amps at the top
> end (continuously), and NOT generate gobs of heat (must be able to be
> enclosed in a plastic ABS box without melting it (ideally it should be
> cool to the touch)), have only a small heatsink, and no fan, I'd have
> say that would impress me.
Can I ask you what you need .3-2v DC or AC for? ( a resistive soldering
iron? )
--
Martin K
2005\01\29@012105
by
Russell McMahon
>> IS capable of putting out upwards of maybe 80 watts under a dead
>> short
> If you can bend the laws of physics to get 80 watts into a dead
> short, you
> don't need anyone's help here.
Indeed.
But that's not what he said.
He said "putting out upwards of maybe 80 watts *under* a dead short"
NOT "*into* a dead short".
The difference is significant.
In the former case the short dissipates nothing but the leads etc
dissipate 80 watts, as happens even in the slightly less than ideal
case.
In the latter case patents should be filed as quickly as possible.
The distinction may be considered to be nit picking, but it does need
to be made if you are going to pick nits.
RM
2005\01\29@063040
by
p.cousens
{Quote hidden}> -----Original Message-----
> From:
.....piclist-bouncesKILLspam.....mit.edu
> [
EraseMEpiclist-bouncesspam_OUTTakeThisOuTmit.edu] On Behalf Of Bob S.
> Sent: Monday, January 10, 2005 6:18 PM
> To:
piclistspam_OUTmit.edu; pic microcontroller discussion list
> Subject: Re: [EE] Single Phase AC control via PWM (maybe a challenge?)
>
>
> Thanks for your input, but....
>
> What you describe as an "active PFC", sounds suspiciously
> like an SMPS.
> Been there, tried that, it doesn't work very good for high amperage
> (low voltage)variable output power supplies (unless your box is
> basically a big aluminum heatsink).
>
> I had a SMPS prototype made, and while it did work, it
> required a lot of
> heatsinking for the output diodes, and even some heatsinks for the
> switching mosfets. It wasn't much more compact, (wasn't
> linear either),
> and by the time you got done with all the extra caps, heatsinks, and
> what not, it weighed about the same as the 60hz transformer
> we're using.
> Basically, I didn't see anything remotely "efficient" about it.
>
> I'm also wanting to go with a plastic enclosure, as the interface
> control I'm wanting to use requires it (no metal allowed).
If I had to do this I'd hack a dimmable electronic l2v halogen lighting
transformer
Give me $3,000 I'll supply you 10 :-)
PC
> I had asked around to a few other people/firms, and the best response
I
> got was that it would cost me $10,000 just for the design & prototype.
A
> bit too pricey for my wallet (especially if it didn't work as
advertised).
2005\01\29@083300
by
Dave Tweed
|
|
Bob S. <@spam@op1cwkKILLspampunkass.com> wrote:
> Dave Tweed wrote:
> > Bob S. wrote:
> > > I'm also wanting to go with a plastic enclosure, as the interface
> > > control I'm wanting to use requires it (no metal allowed).
> >
> > ??? Where are you going to hide that big transformer?
>
> Heh, the transformer isn't that big, as its output is very low voltage
> (read 1.8 vac max.), and it typically only puts out 40 watts or so. It
> IS capable of putting out upwards of maybe 80 watts under a dead short
> fault condition on the outputs (without melting itself).
Obviously, you mean "no metal allowed in the enclosure", not just
"no metal allowed". I thought you wanted to minimize total metal.
> The output amperage is VERY high though (20+ amps RMS.), which will
> excessively heat up any SMPS output diodes (hence the heat sinks).
> Even if you were to use synchronous rectification with low R-on
> mosfets, your still talking about 20+ amps of current.
>
> Also from what I've read about them, SMPS's are not generally used in
> variable output applications.
Well, assuming you want DC output (but I'm starting to wonder about
that), my first thought is that every modern desktop PC motherboard
has a power supply with all of those capabilities and more (variable
output, 20+ amps) to supply the core power of the CPU chip. They are
small, efficient, and produced in huge quantities, so the parts are
cheap. What more could you want?
-- Dave Tweed
2005\01\29@092252
by
Russell McMahon
> Well, assuming you want DC output (but I'm starting to wonder about
> that), my first thought is that every modern desktop PC motherboard
> has a power supply with all of those capabilities and more (variable
> output, 20+ amps) to supply the core power of the CPU chip. They are
> small, efficient, and produced in huge quantities, so the parts are
> cheap. What more could you want?
He wants AC.
RM
2005\01\29@170303
by
Peter L. Peres
On Sat, 29 Jan 2005, p.cousens wrote:
> If I had to do this I'd hack a dimmable electronic l2v halogen
> lighting transformer Give me $3,000 I'll supply you 10 :-)
I'd charge a little more having been down that avenue (hint: the
dimmables I have in mind are *too* cheap for hacking - it's the case of
not being able to remove or modify any parts without their blowing up -
maybe a more expensive model ...)
Peter
2005\01\29@211036
by
p.cousens
Just rewind the output for lower voltage/higher current
12v to 3v is only 1/4 so four windings paralleled, 1/4 of the number of
existing turns
( most units I've opened have around 12 turns on the output, no feedback
from the output)
A 150 watt unit's output would be 12.5 Amps anyway.
Rewound 4 X 1/4 would be 50 amps.
Plastic case, galvanic isolation.
Standard dimmer on the input for output voltage control.
Seems like it would meet all the requirements.
The soft start could be a problem or an advantage though
Defiantly worth playing with!
PC
> {Original Message removed}
2005\01\30@163823
by
Bob S.
|
|
Thanks Russell. I didn't understand myself why Olin was even talking
about that. Boy, people just love to go off on tangents that have
little or nothing to do with what the original post was about
(explaination below).
Normally the power has to travel through a 4 foot 16 Gauge cord, through
a "pen" body of sorts, and finally to a short piece of 18 gauge
resistance wire like nichrome. Basically a "controlled" dead short of
sorts. AC or DC will work. What I was referring to here was this.
Sometimes, the user might screw up the cord (pulls on cord rather than
jack cap), which can eventually lead to a dead short that is relatively
close to the transformer causing it to draw upwards of 80 watts.
What I was "trying" to impress upon people who were reading this, was
that although my current power supply normally draws 40 or 50 watts of
power, it IS "capable" of drawing upwards of 80 watts (without frying
itself). Meaning that is it is designed to operate with a load that is
100% higher than normal conditions without damaging the transformer or
the triac circuit. Barring a bad cord, a lighting strike/spike, or
similar "fault" condition; it should last forever.... I.e: Any design
that would replace it, would have to be just as "bullet proof" as this
one, or better.
Although a modern PC motherboard is capable of putting out several
different CPU voltages, you have 16, maybe upwards of 32 different
voltages that it is capable of generating? I need something more on the
scale of 256 different voltages ranging from 0.3 and 2 volts (again,
AC/DC output doesn't matter). BTW, those DC/DC converters found on
motherboards typically heatsink the mosfets to a big copper trace, and
it is somewhat air cooled via a case and/or nearby CPU fan, and some MB
Mfg.'s are even starting to design them for better cooling (have big
heatsink(s) on the mosfets, and locate them on the MB for better fan
cooling). Now, if you stick that inside a plastic box with minimal
heatsinking, and no fan cooling, your probably going to end up with a
fried circuit, and/or possibly a melted plastic box. Yes, I even looked
into using something like that, except all the chips out there only go
down to 1 vdc, not 0.3 VDC, and they have a somewhat limited range of
output voltages.
In fact, from what I've read in tech briefs from various sources (Intel,
IBM, etc), they are running into some serious power density/heat
barriers at 1 vdc and below because of the high amperage required by
those low voltage CPU's, to the point where they are trying to design
new and different topologies (and hardware components) that are capable
of doing that without generating gobs & gobs of heat (read wasted
energy). If your able to come up with something that could put out very
low voltage DC at very high amperages without generating gobs of heat
(using off the shelf parts), you might wanna get it patented, and then
sell it to a few MB manufacturers. There, that was my "tangent" for
this thread.....
This is why I was wanting to investigate this new type of AC phase
control (see my original post), as I do not need DC output, and I figure
that controlling it from the high voltage side would generate little or
no heat. Now I could do the same with basic triac phase control running
off of a pic, and all I would need to learn is how to monitor the line
voltage going to the transformer (hardware wise), so that the pic could
compensate for line voltage variances (110vac versus 135vac, but have
the same output). But then there are certain other things you then have
to take into account (correcting the non-linearity of the output range,
creating and tweaking a PID control algorithm, etc....) that a PWM'd AC
phase control design would inherently NOT have to deal with.
To put it "simply", here are the basic specs.
1) Able to generate voltages from 0.3 to 2 volts RMS (AC or DC, I don't
Care) in at least 256 distinct steps in the output range.
2) Able to generate over 25 amperes on the output at the top end of the
range.
3) Line voltage compensation (output is same for a given setting, no
matter what the line voltage is reading (90 to 140vac)).
4) Linear output range (hard to do with AC triac phase control).
5) It CANNOT generate gobs of heat (again: plastic box, limited sized
heatsink available, no cooling fans).
6) It CANNOT generate gobs of RF.
7) Doesn't have prohibitively high price tag, parts wise.
Some "features" that would be nice to have, but are not requirements
per-se....
A) Can detect an output fault condition (too many amps being drawn?
Shut down, maybe throw out an error code to an LED display).
B) Can automatically work with either 110 or 220 VAC input. (PWM'd AC
phase control could easily do this w/o an extra transformer winding).
C) Could have the output range cut in half (0.15 to 1 V RMS) via an
adjustment to PIC program (user could toggle range by pressing a button)
D) Minimal parts design! Less parts = less things that can go wrong,
easier to build, easier to fix. I.e., the "KISS" rule....
I would appreciate it if someone here would help me investigate PWM AC
Phase control (AKA Sinewave Phase control), and not try to steer me like
some idiot sheep towards building a SMPS (already tried it, doesn't
work, end of story!). SMPS's are not the "end all - be all" design for
every application! This is a new kind of AC phase control method (not
forward or reverse PC), so at the very least it would be a learning
experience for anyone looking into it.
Russell McMahon wrote:
>>> IS capable of putting out upwards of maybe 80 watts under a dead short
>
>
>
>> If you can bend the laws of physics to get 80 watts into a dead
short, you
>> don't need anyone's help here.
>
>
>
> Indeed.
> But that's not what he said.
> He said "putting out upwards of maybe 80 watts *under* a dead short"
NOT "*into* a dead short".
> The difference is significant.
> In the former case the short dissipates nothing but the leads etc
dissipate 80 watts, as happens even in the slightly less than ideal case.
> In the latter case patents should be filed as quickly as possible.
>
> The distinction may be considered to be nit picking, but it does need
to be made if you are going to pick nits.
>
>
> RM
>
2005\01\30@163826
by
Bob S.
|
|
Thanks Russell. I didn't understand myself why Olin was even talking
about that. Boy, people just love to go off on tangents that have
little or nothing to do with what the original post was about
(explaination below).
Normally the power has to travel through a 4 foot 16 Gauge cord, through
a "pen" body of sorts, and finally to a short piece of 18 gauge
resistance wire like nichrome. Basically a "controlled" dead short of
sorts. AC or DC will work. What I was referring to here was this.
Sometimes, the user might screw up the cord (pulls on cord rather than
jack cap), which can eventually lead to a dead short that is relatively
close to the transformer causing it to draw upwards of 80 watts.
What I was "trying" to impress upon people who were reading this, was
that although my current power supply normally draws 40 or 50 watts of
power, it IS "capable" of drawing upwards of 80 watts (without frying
itself). Meaning that is it is designed to operate with a load that is
100% higher than normal conditions without damaging the transformer or
the triac circuit. Barring a bad cord, a lighting strike/spike, or
similar "fault" condition; it should last forever.... I.e: Any design
that would replace it, would have to be just as "bullet proof" as this
one, or better.
Although a modern PC motherboard is capable of putting out several
different CPU voltages, you have 16, maybe upwards of 32 different
voltages that it is capable of generating? I need something more on the
scale of 256 different voltages ranging from 0.3 and 2 volts (again,
AC/DC output doesn't matter). BTW, those DC/DC converters found on
motherboards typically heatsink the mosfets to a big copper trace, and
it is somewhat air cooled via a case and/or nearby CPU fan, and some MB
Mfg.'s are even starting to design them for better cooling (have big
heatsink(s) on the mosfets, and locate them on the MB for better fan
cooling). Now, if you stick that inside a plastic box with minimal
heatsinking, and no fan cooling, your probably going to end up with a
fried circuit, and/or possibly a melted plastic box. Yes, I even looked
into using something like that, except all the chips out there only go
down to 1 vdc, not 0.3 VDC, and they have a somewhat limited range of
output voltages.
In fact, from what I've read in tech briefs from various sources (Intel,
IBM, etc), they are running into some serious power density/heat
barriers at 1 vdc and below because of the high amperage required by
those low voltage CPU's, to the point where they are trying to design
new and different topologies (and hardware components) that are capable
of doing that without generating gobs & gobs of heat (read wasted
energy). If your able to come up with something that could put out very
low voltage DC at very high amperages without generating gobs of heat
(using off the shelf parts), you might wanna get it patented, and then
sell it to a few MB manufacturers. There, that was my "tangent" for
this thread.....
This is why I was wanting to investigate this new type of AC phase
control (see my original post), as I do not need DC output, and I figure
that controlling it from the high voltage side would generate little or
no heat. Now I could do the same with basic triac phase control running
off of a pic, and all I would need to learn is how to monitor the line
voltage going to the transformer (hardware wise), so that the pic could
compensate for line voltage variances (110vac versus 135vac, but have
the same output). But then there are certain other things you then have
to take into account (correcting the non-linearity of the output range,
creating and tweaking a PID control algorithm, etc....) that a PWM'd AC
phase control design would inherently NOT have to deal with.
To put it "simply", here are the basic specs.
1) Able to generate voltages from 0.3 to 2 volts RMS (AC or DC, I don't
Care) in at least 256 distinct steps in the output range.
2) Able to generate over 25 amperes on the output at the top end of the
range.
3) Line voltage compensation (output is same for a given setting, no
matter what the line voltage is reading (90 to 140vac)).
4) Linear output range (hard to do with AC triac phase control).
5) It CANNOT generate gobs of heat (again: plastic box, limited sized
heatsink available, no cooling fans).
6) It CANNOT generate gobs of RF.
7) Doesn't have prohibitively high price tag, parts wise.
Some "features" that would be nice to have, but are not requirements
per-se....
A) Can detect an output fault condition (too many amps being drawn?
Shut down, maybe throw out an error code to an LED display).
B) Can automatically work with either 110 or 220 VAC input. (PWM'd AC
phase control could easily do this w/o an extra transformer winding).
C) Could have the output range cut in half (0.15 to 1 V RMS) via an
adjustment to PIC program (user could toggle range by pressing a button)
D) Minimal parts design! Less parts = less things that can go wrong,
easier to build, easier to fix. I.e., the "KISS" rule....
I would appreciate it if someone here would help me investigate PWM AC
Phase control (AKA Sinewave Phase control), and not try to steer me like
some idiot sheep towards building a SMPS (already tried it, doesn't
work, end of story!). SMPS's are not the "end all - be all" design for
every application! This is a new kind of AC phase control method (not
forward or reverse PC), so at the very least it would be a learning
experience for anyone looking into it.
Russell McMahon wrote:
>>> IS capable of putting out upwards of maybe 80 watts under a dead short
>
>
>
>> If you can bend the laws of physics to get 80 watts into a dead
short, you
>> don't need anyone's help here.
>
>
>
> Indeed.
> But that's not what he said.
> He said "putting out upwards of maybe 80 watts *under* a dead short"
NOT "*into* a dead short".
> The difference is significant.
> In the former case the short dissipates nothing but the leads etc
dissipate 80 watts, as happens even in the slightly less than ideal case.
> In the latter case patents should be filed as quickly as possible.
>
> The distinction may be considered to be nit picking, but it does need
to be made if you are going to pick nits.
>
>
> RM
>
2005\01\30@173733
by
olin_piclist
Bob S. wrote:
> To put it "simply", here are the basic specs.
>
> 1) Able to generate voltages from 0.3 to 2 volts RMS (AC or DC, I
> don't Care) in at least 256 distinct steps in the output range.
>
> 2) Able to generate over 25 amperes on the output at the top end of
> the range.
>
> 3) Line voltage compensation (output is same for a given setting, no
> matter what the line voltage is reading (90 to 140vac)).
>
> 4) Linear output range (hard to do with AC triac phase control).
>
> 5) It CANNOT generate gobs of heat (again: plastic box, limited sized
> heatsink available, no cooling fans).
>
> 6) It CANNOT generate gobs of RF.
>
> 7) Doesn't have prohibitively high price tag, parts wise.
So why not just full wave rectify the AC line, then PWM that into the
transformer primary? A feedback circuit measures the actual power output,
and the PWM duty cycle at some fixed frequency of maybe 100Hz is adjusted
accordingly. Since you are apparently just trying to heat the final load,
you don't need a nice sine wave coming out of the secondary.
> I would appreciate it if someone here would help me investigate PWM AC
> Phase control (AKA Sinewave Phase control), and not try to steer me
> like some idiot sheep towards building a SMPS (already tried it,
> doesn't work, end of story!).
If it didn't work for you, that only proves your implementation was bad, not
the SMPS is inherently bad. Don't look now, but "PWM AC Phase control" is
really a SMPS dressed up with a different name.
*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
2005\01\30@174507
by
Peter L. Peres
On Sun, 30 Jan 2005, p.cousens wrote:
> Just rewind the output for lower voltage/higher current 12v to 3v is
> only 1/4 so four windings paralleled, 1/4 of the number of existing
> turns ( most units I've opened have around 12 turns on the output, no
> feedback from the output) A 150 watt unit's output would be 12.5 Amps
> anyway. Rewound 4 X 1/4 would be 50 amps. Plastic case, galvanic
> isolation. Standard dimmer on the input for output voltage control.
> Seems like it would meet all the requirements. The soft start could be
> a problem or an advantage though Defiantly worth playing with!
You clearly have a different type in mind. The ones I netted used
current feedback from the load so they had no output with no load. The
device simply 'regulated' by having a power limit (transformer
saturation). Only two high voltage transistors, a few diodes and a diac
inside. The diac was for safe starting iirc.
Peter
2005\01\30@190927
by
Dave Tweed
|
|
"Bob S." <KILLspamop1cwkKILLspampunkass.com> wrote:
> Thanks Russell. I didn't understand myself why Olin was even talking
> about that. Boy, people just love to go off on tangents that have
> little or nothing to do with what the original post was about
> (explaination below).
Indeed. Such are the consequences of asking a bad question. It's only
taken us 31 days (your original post was on Dec. 30) to get the *real*
requirements of your application, and a clue about the things you've
already tried. If you had presented this information up front, it would
have saved everyone a lot of time and anguish.
> Normally the power has to travel through a 4 foot 16 Gauge cord
That's about 32 milliohms right there, almost half of your 80-milliohm
working load. And that's not even considering the connectors in the
path. But I guess you've worked all of that out and are satisfied with
the results.
Anyway, we're back where we started, which was you asking:
> Only big problem I've seen for doing this circuit so far, is being
> able to generate a 48khz+ PWM using a PIC with fine enough resolution
> (I'd prefer 10 bit resolution, especially if you want to do the
> universal input). Any simple/easy/cheap way to do that trick with a
> PIC? Is there a SMPS control chip circuit that could be modified or
> tricked to do this (and could be controlled via a PIC)?
And now you're asking:
> I would appreciate it if someone here would help me investigate
> PWM AC Phase control (AKA Sinewave Phase control), and not try to
> steer me like some idiot sheep towards building a SMPS (already
> tried it, doesn't work, end of story!). SMPS's are not the "end
> all - be all" design for every application! This is a new kind
> of AC phase control method (not forward or reverse PC), so at the
> very least it would be a learning experience for anyone looking
> into it.
Spell it out for us: Have you tried generating the 48 kHz 10-bit PWM
by any means at all (ignoring a cheap PIC-based imnplementation for
the moment) and were you satisfied with the results? It now appears
that you're asking someone else to do that investigation for you.
-- Dave Tweed
2005\01\31@041000
by
Bob S.
|
|
Hmm, unless I'm totally missing something here, or the basics of AC
power transformers has suddenly changed through some quirk of quantum
physics... Last time I checked, 60 hz power transformers kinda liked
AC, not DC, or pulsed DC. I think it would probably have the same
effects as asymmetrically gating a triac (core saturation, followed by
the smell of burnt wire varnish). Hmm, 100hz PWM'd 60hz wave form, that
would sound really interesting coming from a transformer (AC or DC).
Sigh.... No, I do not necessarily need a nice sine wave to the load. I
could just use a triac hooked up to a PIC. Then again, I could just
scrap the whole idea, and just sell propane torches to create heat....
The whole point here was to try and experiment with A DIFFERENT TYPE OF
PHASE CONTROL!!!! Why? READ my original post! Sorry, I didn't realize
that everyone here was so set in their ways.
That SMPS I was talking about wasn't my design, it was a highly
qualified "Engineer's" design, who was just as sure as you are that
SMPS's can, and should be used to do everything. Gee, do you dim the
lights in your house with a SMPS too? While we're at it, lets go out
the the power line pole and swap out that old inefficient transformer
with a SMPS, that'll work.....
And PWM's AC Phase control has little to do with a SMPS, other than it
would use MOSFETS or IGBT's instead of a triac/scr's. No Large
eletrolytic Caps involved, No rectifying the AC into DC (hence, no large
caps), no bootstrapping, no hiccup mode, etc, etc.... The Topology is
even totally different.
Inductor Inductor
Hot---------(((((--<-Mosfet---Mosfet->--)))))---LOAD
|
60hz AC Line |
|
Neutral-------------------------------------------|
Keep in mind that this a crude block diagram, not a detailed scematic,
so please don't go off on a tangent telling me how wrong it is.
Olin Lathrop wrote:
{Quote hidden}> Bob S. wrote:
>
>>To put it "simply", here are the basic specs.
>>
>>1) Able to generate voltages from 0.3 to 2 volts RMS (AC or DC, I
>>don't Care) in at least 256 distinct steps in the output range.
>>
>>2) Able to generate over 25 amperes on the output at the top end of
>>the range.
>>
>>3) Line voltage compensation (output is same for a given setting, no
>>matter what the line voltage is reading (90 to 140vac)).
>>
>>4) Linear output range (hard to do with AC triac phase control).
>>
>>5) It CANNOT generate gobs of heat (again: plastic box, limited sized
>>heatsink available, no cooling fans).
>>
>>6) It CANNOT generate gobs of RF.
>>
>>7) Doesn't have prohibitively high price tag, parts wise.
>
>
> So why not just full wave rectify the AC line, then PWM that into the
> transformer primary? A feedback circuit measures the actual power output,
> and the PWM duty cycle at some fixed frequency of maybe 100Hz is adjusted
> accordingly. Since you are apparently just trying to heat the final load,
> you don't need a nice sine wave coming out of the secondary.
>
>
>>I would appreciate it if someone here would help me investigate PWM AC
>>Phase control (AKA Sinewave Phase control), and not try to steer me
>>like some idiot sheep towards building a SMPS (already tried it,
>>doesn't work, end of story!).
>
>
> If it didn't work for you, that only proves your implementation was bad, not
> the SMPS is inherently bad. Don't look now, but "PWM AC Phase control" is
> really a SMPS dressed up with a different name.
>
>
> *****************************************************************
> Embed Inc, embedded system specialists in Littleton Massachusetts
> (978) 742-9014,
http://www.embedinc.com
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.8.2 - Release Date: 1/28/2005
2005\01\31@051749
by
p.cousens
Probably the best plan
>Then again, I could just
> scrap the whole idea, and just sell propane torches to create
> heat....
So tell us what stage your at, what you have done/ tried
> The whole point here was to try and experiment with A
> DIFFERENT TYPE OF
> PHASE CONTROL!!!!
PC
2005\01\31@054114
by
Bob S.
|
|
part 1 5591 bytes content-type:text/plain; charset=us-ascii; format=flowed (decoded 7bit)
Actually, my original post is dated 1/6/2005, at least according to my
email logs. I've also been extremely busy this last month, and I've
been having some stupid problems with sending out emails to this
listserver...
I wasn't asking for anyone to out and out design it for me at that time,
so there was no need to give detailed specs. But of course, people on
this listserver seem to have an unholy obsession with SMPS's, so next
thing ya know, everyone's telling me to make a SMPS, and me saying
"tried it, don't work", back and forth, etc, etc....
As I've already said, more than once, this is a different type of AC
Phase control (NOT SMPS). Was asking how to set up the mosfets to do
this (do I need free wheeling diodes, etc?), maybe get an idea what
values I would need for inductors, figuring out how to run the PWM
frequency high enough but still use a pic to control the darned thing.
Yea, I'm aware that I loose upwards of half the power in the cord
alone. Not much can be done about that. Can't make it shorter, Silver
is too expensive, and I can't use car battery cables... Did I mention
that it's a high quality, high stranding, oxygen free copper cord? ;-)
Don't worry about it, I just want to try a different type of gas in my
"engine", not totally re-invent the wheel, the engine, or the whole car
for that matter.
Well, that was the first basic question I asked. Here, I'll quote from
my original email:
Only big problem I've seen for doing this circuit so far, is being able
to generate a 48khz+ PWM using a PIC with fine enough resolution (I'd
prefer 10 bit resolution, especially if you want to do the universal
input). Any simple/easy/cheap way to do that trick with a PIC? Is
there a SMPS control chip circuit that could be modified or tricked to
do this (and could be controlled via a PIC)?
QUESTION ASKED! NEVER ANSWERED! We all went off on the SMPS tangent
instead.
Could you use some sort of frequency doubler or quadrupler. Can a
standard PIC's hardware PWM be sync'd to a different frequency than what
the rest of the uP is using? Or do I need to just go out and get some
sort of stand alone hardware PWM that is controlled with DC input? Any
suggestions? Is this spelled out enough? What part of these questions
am I not making clear here?
Just like when people here say RTFM, you first have to RTOFP 8^]
That last paragraph you quoted is not a question at all (I don't see
even one ? in it). It's more of a plea for people to get off of the
SMPS kick, and get back to what I was originally asking about.
I've been very persistent in asking for help, but now I'm starting to
see why I see a lot of one time posts from people who are never heard
from again on this listserver. So far it's been kind of like asking how
to get a sharper knife edge, and being told that I should go out and get
a chain saw instead (by the listserver's ample supply of chain saw
enthusiasts and dealers)....
Dave Tweed wrote:
{Quote hidden}>"Bob S." <
RemoveMEop1cwkTakeThisOuTpunkass.com> wrote:
>
>
>>Thanks Russell. I didn't understand myself why Olin was even talking
>>about that. Boy, people just love to go off on tangents that have
>>little or nothing to do with what the original post was about
>>(explaination below).
>>
>>
>
>Indeed. Such are the consequences of asking a bad question. It's only
>taken us 31 days (your original post was on Dec. 30) to get the *real*
>requirements of your application, and a clue about the things you've
>already tried. If you had presented this information up front, it would
>have saved everyone a lot of time and anguish.
>
>
>
>>Normally the power has to travel through a 4 foot 16 Gauge cord
>>
>>
>
>That's about 32 milliohms right there, almost half of your 80-milliohm
>working load. And that's not even considering the connectors in the
>path. But I guess you've worked all of that out and are satisfied with
>the results.
>
>Anyway, we're back where we started, which was you asking:
>
>
>
>>Only big problem I've seen for doing this circuit so far, is being
>>able to generate a 48khz+ PWM using a PIC with fine enough resolution
>>(I'd prefer 10 bit resolution, especially if you want to do the
>>universal input). Any simple/easy/cheap way to do that trick with a
>>PIC? Is there a SMPS control chip circuit that could be modified or
>>tricked to do this (and could be controlled via a PIC)?
>>
>>
>
>And now you're asking:
>
>
>
>>I would appreciate it if someone here would help me investigate
>>PWM AC Phase control (AKA Sinewave Phase control), and not try to
>>steer me like some idiot sheep towards building a SMPS (already
>>tried it, doesn't work, end of story!). SMPS's are not the "end
>>all - be all" design for every application! This is a new kind
>>of AC phase control method (not forward or reverse PC), so at the
>>very least it would be a learning experience for anyone looking
>>into it.
>>
>>
>
>Spell it out for us: Have you tried generating the 48 kHz 10-bit PWM
>by any means at all (ignoring a cheap PIC-based imnplementation for
>the moment) and were you satisfied with the results? It now appears
>that you're asking someone else to do that investigation for you.
>
>-- Dave Tweed
>
>
part 2 185 bytes content-type:text/plain; x-avg=cert; charset=us-ascii
(decoded quoted-printable)
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.8.2 - Release Date: 1/28/2005
part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)
2005\01\31@072444
by
p.cousens
What is a SMPS?
If you can answer this perhaps you would understand why
"We all went off on the SMPS tangent"
PC
{Quote hidden}> Only big problem I've seen for doing this circuit so far, is
> being able
> to generate a 48khz+ PWM using a PIC with fine enough resolution (I'd
> prefer 10 bit resolution, especially if you want to do the universal
> input). Any simple/easy/cheap way to do that trick with a PIC? Is
> there a SMPS control chip circuit that could be modified or tricked to
> do this (and could be controlled via a PIC)?
>
> QUESTION ASKED! NEVER ANSWERED! We all went off on the SMPS tangent
> instead.
>
2005\01\31@074252
by
Russell McMahon
|
|
OK - I'll take a little time here. Listen carefully. I'll try and be
useful technically and otherwise.
Good (hopefully) technical comment below, but 1st a very little
general advice.
People here are generally helpful, but not always. They don't always
get the point and some are ruder than others. You happen to have been
exchanging words with one of the ruder ones - but he is also extremely
competent technically and well experienced in power supplies. You are
far better off having him as a friend than an enemy. You have reacted
to his rudeness with rudeness and in the process of accusing him of
not listening and of offering stupid advice you have failed to listen
fully and have said technically stupid things about his (quite
possibly useful) advice because you didn't read what he said
carefully. Exchanging sarcasm with Olin is seldom profitable for any
concerned. The best you get is nothing and more often than not you get
two or more disgruntled people.
There are indeed a few one time posters who never come back but most
newcomers do in fact stay. Some people here have trouble reacting well
to newcomers but we try to stop them abusing them while they get used
to the friendly family atmosphere :-). In fact, a premium is placed on
politeness and treating each other well here. Levels of abuse which
are an accepted part of many lists are frowned on here and active
steps are taken to "discourage" such. (Try not to find this out
actively ;-) ).
I suggest we all start again and try and be nice to each other.
Your original proposal had merit. It is not especially new and has its
good and bad points, but it had the makings of a good idea.
I suggest that anyone who has been criticising what they THINK Bob has
been suggesting have a look at his original proposal. It was
competently described and commented on and it seems technically sound.
He didn't describe the end application but he didn't need to. While
it's always useful to know what people are trying to achieve end to
end, in this case he said he was keen in implementing a particular
system. It's entirely valid to suggest alternatives BUT trashing the
original concept without being aware of what it was is not an
especially good idea.
[[SUMMARY: PWMing mains AC with an AC switch in series with a
transformer primary presents variable voltage low frequency AC input
to the transformer. Advantages are no rectification loss, prospect of
OK overall efficiencies, excellent power factor, reasonable prospect
of open loop control, retrofitting to existing mains transformers,
.... . Still needs low frequency mains transformer. ]]
If I was trying to meet you spec from scratch I would try either
something like what you have suggested OR a transformer and rectifier
(or a SMPS) providing about 2.5 volts DC followed by a buck regulator.
The output could be very precisely controlled by feedback.
Let's have a quick look at what Olin suggested. He proposed a full
wave rectified unsmoothed DC supply followed by PWM into a
transformer. The act of PWMing will provide the AC component that the
transformer seeks. The transformer will be able to be a very small one
compared to one used on 50 Hz due to the high frequency of the PWM.
When turned off the transformer primary will "flyback" and will need a
recirculating diode across it. Lo and behold - you now have the SMPS
you were trying to avoid :-). Efficiency should be able to be
reasonably high (= lowish heat) as rectification and switching is done
at high voltage and low current. What he is suggesting is similar in
end result to yours but has the very significant advantage of a small
high frequency transformer and it also allows a simpler switch.
Let's try and address you original circuit.
You can chop the AC as proposed with 2 FETS *BUT* they cannot be in
parallel antiphase due to the body diodes - they must be in series -
gate to gate, source to source, input and outputs from the 2 drains.
Switching is achieved by PWMing the gate to source as required. The
whole switch floats across the whole AC range so the gate drive needs
to be floating. There are good ways of doing this BUT for
experimenting it's better to use an "EASY" way first and come back and
improve it technically once the rest works OK. An easy way is to use a
9v volt battery plus an opto to drive the FET gates. The battery will
last for weeks of use and a floating isolated supply (or other
solution) can be implemented to replace it in due course.
If you have a conventional mains transformer AND you PWM the mains
supply AND you filter the output to remove the PWM component BUT not
the 50 Hz AC component THEN you will get a variable amplitude AC
waveform which the transformer will happily accept. This could be a
very nice solution but still has a standard mains transformer.
Switching efficiency is goodish, and there are no rectifiers to
dissipate power. The PWM filter need not be excessively large. The
transformer itself will form a reasonably good filter in its own right
due to its significant inductance.
Let's jump to the PIC question and then look at other things. If you
use 10 bit control then you have 2^10= 1024 steps. If you want 48 kHz
then you get about 50 million possible steps/second which is VERY fast
PWM. It's also humungously fast for FET or other switches. I would be
surprised if you need either 48 KHz frame rate or 10 bit resolution
BUT in the absence of knowledge of the final application I may, of
course, be wrong. I have a design where inaudibility is important and
I am PWMing hundreds of watts in an electromechanical device that
attempts to be a speaker. I find 20 to 25 kHz is high enough for most
people. If you are working with something that heats as its main task
and it is in air and if it is working into a heat sink (person, wood,
liquid, other) and if it is moving and if it is fed via leads that
constitute a significant percentage of the load THEN variations in
convection/conduction/radiation, air currents, thermal paths, working
medium and the rest all work to ensure that far far less than 10 bits
accuracy in the outcomes is possible. Real world engineering often
doesn't need such. As we don't know the application .... .
I am not familiar with all of the latest PIC range but some modern
processors (eg AVR ATtiny26 have a phase lock loop controlled PWM
where the PWM frequency is locked at many times the processor clock
frequency. Without looking at the data sheet AFAIR the PWM can be at
up to about 50 MHz data rate ! Switching a power FET at such rates is
another matter. I would not recommend much above 1 Mhz fastest
switching time unless you are desperately unable to do it any other
way.
Feedback control will often make raw control precision less important
as you can alter the drive to suit what the output is doing.
If you want to try the basic idea you can use far far far lower than
10 bits accuracy to trial the basic concept. An analog 555 oscillator
would do just fine to drive the series connected back to back FETs
into the transformer primary. If I was doing this I'd start with low
voltage :-)
If you are able/willing to tell us the actual application is it would
help. Sounds something like either a poker-work type heater or a
person heater/therapy unit.
Russell McMahon
2005\01\31@085635
by
Dave Tweed
|
|
"Bob S." <spamBeGoneop1cwkspamBeGonepunkass.com> wrote:
> I've also been extremely busy this last month, ...
And you assume that the rest of us have gobs of free time to figure out
what you're really asking...
> ... and I've been having some stupid problems with sending out emails
> to this listserver...
Yeah, well, up until now, you've been sending to both server addresses,
so we've been getting everything that gets through twice.
> I wasn't asking for anyone to out and out design it for me at that
> time, so there was no need to give detailed specs.
but you are now?
> But of course, people on this listserver seem to have an unholy
> obsession with SMPS's, so next thing ya know, everyone's telling me
> to make a SMPS, and me saying "tried it, don't work", back and forth,
> etc, etc....
Look carefully at your own quote of your own post. Who mentioned SMPS
first?
> What part of these questions am I not making clear here?
It doesn't matter any more. It's clear that you cannot be worked with
in a civil manner, even though you have a moderately interesting
technical problem. Now I know where that domain name came from.
>>> PLONK <<<
-- Dave Tweed
2005\01\31@085928
by
Martin K
Bob S. wrote:
> QUESTION ASKED! NEVER ANSWERED! We all went off on the SMPS tangent
> instead.
I believe if someone doesn't know how to answer your question the way
you want it answered they have the right to ask for more information.
You have the right to ignore people that "go off on the SMPS tangent".
Nobody gets paid either way here, it's quite free.
That being said, what sort of heating element are you using and why
can't it be made higher resistance so you don't need mega amps at low volts?
--
Martin K
http://wwia.org/sgroup/biofuel/
2005\01\31@092118
by
olin_piclist
Bob S. wrote:
> Hmm, unless I'm totally missing something here, or the basics of AC
> power transformers has suddenly changed through some quirk of quantum
> physics... Last time I checked, 60 hz power transformers kinda liked
> AC, not DC, or pulsed DC.
Who said anything about DC? I was suggesting to rectify the AC line, then
PWM that into the transformer primary with feedback adjusting the duty cycle
to get the desired power transfer. The rectified AC line is DC, but that
doesn't necessarily translate to DC into the transformer primary.
There are several ways to deal with the issue of DC into the transformer.
First, you can ignore the DC component and the transformer takes it out.
This appears to be the only method you were thinking of. This is usually
not the preferred method, especially at high power levels, since the DC
current transfers no power to the output, reduces overhead to the magnetic
saturation limit of the core, and represents wasted power. Another way is
to put a capacitor in series, although this is again unwieldy at high power
levels. If you can't control the transformer, then an H bridge input drive
would work. If you have a center tapped primary, then you can use a
push-pull configuration with only two switching elements.
> I think it would probably have the same
> effects as asymmetrically gating a triac (core saturation, followed by
> the smell of burnt wire varnish). Hmm, 100hz PWM'd 60hz wave form, that
> would sound really interesting coming from a transformer (AC or DC).
Again, only in the limited interpretation of "PWM" to mean single ended
pulses from ground to the supply voltage. Clearly that is not a suitable
approach in this case.
> Sigh.... No, I do not necessarily need a nice sine wave to the load. I
> could just use a triac hooked up to a PIC. Then again, I could just
> scrap the whole idea, and just sell propane torches to create heat....
> The whole point here was to try and experiment with A DIFFERENT TYPE OF
> PHASE CONTROL!!!! Why? READ my original post! Sorry, I didn't realize
> that everyone here was so set in their ways.
Well excuuuuuse me! You came here asking for ideas. If you've already made
up your mind what you want, don't waste everyone's time asking. I threw out
one idea. I don't know if it's optimum for your design. It's your design.
However there is no point ridiculing it, especially since you didn't bother
to understand it first. This is hardly a way to entice people to give you
free advice.
*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
2005\01\31@094143
by
Mike Hawkshaw
|
|
----- Original Message -----
From: "Russell McMahon"
To: "Microcontroller discussion list - Public." <TakeThisOuTpiclistEraseMEspam_OUTmit.edu>
Sent: Monday, January 31, 2005 12:42 PM
Subject: Re: [EE] Single Phase AC control via PWM (maybe a challenge?)
> [[SUMMARY: PWMing mains AC with an AC switch in series with a
> transformer primary presents variable voltage low frequency AC input
> to the transformer.
We have some bench power supplies made by a company called Roband. Having
repaired one of these in my dim past, If I remember correctly, the basic
topology was mains transformer followed by a half controlled recifier, then
the smoothing and linear regulator. The idea was that the 1/2 controlled
bridge would keep the input to the regulator high enough for good
regulation, but low enough that there wasn't a massive amount of wasted
power.
The 1/2 controlled bridge used thyristers, I think, and had a fairly crude
control circuit. Very neet idea in it's day, but now surpassed by - sorry -
SMPSs.
I don't know if this would help but it would keep the electronics isolated
from the live mains.
By the way, the transformer used to buzz violently as the load changed!
Mike.
--
Outgoing mail certified virus free.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.8.2 - Release Date: 28/01/2005
2005\01\31@095043
by
olin_piclist
Bob S. wrote:
> But of course, people on
> this listserver seem to have an unholy obsession with SMPS's, so next
> thing ya know, everyone's telling me to make a SMPS, and me saying
> "tried it, don't work", back and forth, etc, etc....
But you didn't give any reason for why it didn't work, which leaves the
probability that your implementation was flawed. After all, you came here
asking for help, and therefore your statement of "SMPS didn't work" is
suspect and not terribly relevant without more specifics.
> As I've already said, more than once, this is a different type of AC
> Phase control (NOT SMPS).
I thought you wanted to control the temperature of a heating element. Why
does this have to be done with "AC phase control", whatever it is you might
think that term means? Also SMPS is not rigidly defined. Anything that
switches (as apposed to constant "linear" operation) and produces or handles
power can be considered a SMPS. You seem to have some allergic reaction to
that term, which makes your other statements less credible and therefore
mostly ignored.
> Was asking how to set up the mosfets to do
> this (do I need free wheeling diodes, etc?), maybe get an idea what
> values I would need for inductors, figuring out how to run the PWM
> frequency high enough but still use a pic to control the darned thing.
Perhaps if you showed a schematic of this approach these questions could be
answered. It's still not clear to me what your proposed topology looks
like.
> Well, that was the first basic question I asked. Here, I'll quote from
> my original email:
>
> Only big problem I've seen for doing this circuit so far, is being able
> to generate a 48khz+ PWM using a PIC with fine enough resolution (I'd
> prefer 10 bit resolution, especially if you want to do the universal
> input). Any simple/easy/cheap way to do that trick with a PIC? Is
> there a SMPS control chip circuit that could be modified or tricked to
> do this (and could be controlled via a PIC)?
Frankly I didn't really understand the topology you were referring to, so I
just ignore this. Besides, I thought the problem was controlling power to a
heating element, so it makes sense to look at the bigger picture than think
inside your poorly defined box.
> QUESTION ASKED! NEVER ANSWERED! We all went off on the SMPS tangent
> instead.
If you feel you didn't get good value from the free advice offered on this
list, I'm sure everyone would be happy to give you your money back.
> Could you use some sort of frequency doubler or quadrupler.
Huh!!? Now I'm really lost. I can't understand what this has to do with
anything. Again, show a schematic or block diagram. Otherwise "frequency"
doubler makes no sense, so it will likely just be ignored again.
> Just like when people here say RTFM, you first have to RTOFP 8^]
No, I don't have to do anything for you when you're asking me for free
advice.
> So far it's been kind of like asking how
> to get a sharper knife edge, and being told that I should go out and get
> a chain saw instead
Very often the right solution to a question is to ask a different question
in the first place. It's still not clear that this doesn't apply in your
case as well.
All in all you need to remember you are asking 2000 people to do you a
favor. You came accross like an arrogant ingrate in this post, which isn't
likely to get you what you want.
*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
2005\01\31@121738
by
Dwayne Reid
|
At 02:09 AM 1/31/2005, Bob S. wrote:
>And PWM's AC Phase control has little to do with a SMPS, other than it
>would use MOSFETS or IGBT's instead of a triac/scr's. No Large
>eletrolytic Caps involved, No rectifying the AC into DC (hence, no large
>caps), no bootstrapping, no hiccup mode, etc, etc.... The Topology is
>even totally different.
>
>
>
> Inductor Inductor
>Hot---------(((((--<-Mosfet---Mosfet->--)))))---LOAD
> |
>60hz AC Line |
> |
>Neutral-------------------------------------------|
>
>Keep in mind that this a crude block diagram, not a detailed scematic, so
>please don't go off on a tangent telling me how wrong it is.
I've been toying with that concept off and on for quite a while now - no
real work done on it, just back of napkin stuff and such. I really do have
to get off my duff and get serious about it, though.
Just out of curiosity, what was your original application? It almost
sounds like one of those wood-burning pens you see in craft shops or at Lee
Valley Tools. I've built a couple of those over the years - the pen and
nichrome elements came from Lee Valley (Razortip comes to mind) and it was
child's play to build my own phase-controlled transformer supplies. I keep
one handy near my shop bench for cutting and shaping plastic.
dwayne
--
Dwayne Reid <RemoveMEdwaynerTakeThisOuTplanet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
Celebrating 20 years of Engineering Innovation (1984 - 2004)
.-. .-. .-. .-. .-. .-. .-. .-. .-. .-
`-' `-' `-' `-' `-' `-' `-' `-' `-'
Do NOT send unsolicited commercial email to this email address.
This message neither grants consent to receive unsolicited
commercial email nor is intended to solicit commercial email.
2005\01\31@122220
by
Dwayne Reid
|
|
At 03:41 AM 1/31/2005, Bob S. wrote:
>Only big problem I've seen for doing this circuit so far, is being able
>to generate a 48khz+ PWM using a PIC with fine enough resolution (I'd
>prefer 10 bit resolution, especially if you want to do the universal
>input). Any simple/easy/cheap way to do that trick with a PIC? Is
>there a SMPS control chip circuit that could be modified or tricked to
>do this (and could be controlled via a PIC)?
I'm pursuing an analog solution for my version - I've already got a dirt
cheap technique for high frequency ramp and comparitor that I use in a
couple of my other SMPS designs. You could, of course, use generate the
control signal from a PIC by using low frequency PWM and filtering that to
near DC.
dwayne
--
Dwayne Reid <dwaynerEraseME.....planet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
Celebrating 20 years of Engineering Innovation (1984 - 2004)
.-. .-. .-. .-. .-. .-. .-. .-. .-. .-
`-' `-' `-' `-' `-' `-' `-' `-' `-'
Do NOT send unsolicited commercial email to this email address.
This message neither grants consent to receive unsolicited
commercial email nor is intended to solicit commercial email.
2005\01\31@172403
by
Russell McMahon
> I'm pursuing an analog solution for my version - I've already got a
> dirt cheap technique for high frequency ramp and comparitor that I
> use in a couple of my other SMPS designs. You could, of course, use
> generate the control signal from a PIC by using low frequency PWM
> and filtering that to near DC.
If you do the switching at the load end then you can post filter to
DC. If you do the switching at the input prior to the transformer (as
he was proposing) you need to filter the PWM but leave the resultant
low frequency AC for the transformer to handle.
RM.
2005\01\31@172409
by
Russell McMahon
|
|
{Quote hidden}>> [[SUMMARY: PWMing mains AC with an AC switch in series with a
>> transformer primary presents variable voltage low frequency AC
>> input
>> to the transformer.
> We have some bench power supplies made by a company called Roband.
> Having
> repaired one of these in my dim past, If I remember correctly, the
> basic
> topology was mains transformer followed by a half controlled
> recifier, then
> the smoothing and linear regulator. The idea was that the 1/2
> controlled
> bridge would keep the input to the regulator high enough for good
> regulation, but low enough that there wasn't a massive amount of
> wasted
> power.
>
> The 1/2 controlled bridge used thyristers, I think, and had a fairly
> crude
> control circuit. Very neet idea in it's day, but now surpassed by -
> sorry -
> SMPSs.
The key difference is that no attempt is made to preserve the waveform
in the above method. When thyristors are used, with switch off being
caused by cessation of current flow at the end of a mains half cycle,
the waveform gets progressively distorted and pulse shaped as the
output is reduced and phase angle approaches 180 degrees.
With the method that Bob is proposing the AC is chopped finely across
the full half cycle and the filtered output is a reasonably accurate
facsimile of the original but of reduced amplitude. This allows input
to a following 50 or 60 HZ transformer in the usual way.
> I don't know if this would help but it would keep the electronics
> isolated
> from the live mains.
His method is mains-live at the switching point but then typically
would use a normal mains transformer for isolation.
> By the way, the transformer used to buzz violently as the load
> changed!
That's due to pulsed load. Bob's method provides a proportionately
reduced amplitude sinusoidal load across trhe whole cycle and would
probably be extremely benign.
I have used FETs in series opposed mode for on off switching of high
power (hundreds of watts) high frequency AC in an inductive power
transfer application. It worked well.
RM
{Quote hidden}>
> Mike.
>
>
>
> --
> Outgoing mail certified virus free.
> Checked by AVG Anti-Virus.
> Version: 7.0.300 / Virus Database: 265.8.2 - Release Date:
> 28/01/2005
>
> --
2005\01\31@175423
by
Russell McMahon
|
|
Here we have several strong willed people (N against 1 - not usually
very fair) with good ideas talking past each other, failing to
understand what the other person has said, missing the good points the
other has offered, being gratuitously rude to each other, incorrectly
criticising technical input from the other person and generally having
a good time.
Why not start again and try a little harder. All can learn from this
technically ( if not socially :-) )
Bob may yet learn that SMPS is not a total swear word and that what he
is proposing has SOME similarities.
And 'we all' may realise that Bob is proposing a not new but less
familiar technique which may be very useful on occasion and which is
being used extremely successfully in some large real world
applications.
A confusion is that bob has used the term "phase control" consistently
when it's not phase control as our mothers understood the term. The
method simply consists of PWMing an input (usually mains) 50/60 Hz AC
waveform so that its amplitude is effectively reduced. The reduction
is NOT achieved by converting to DC and then using high frequency Ac
to reconstitute the waveform. There is no intentional flyback or
energy storage at high frequency in the following transformer. The
chopped AC is either filtered before it gets to the mains transformer
OR the chopping rate is such that the transformer does not "see" the
high frequency as the transformer acts as part of an low pass filter.
There will be some issues with transformer response to high frequency
components and back of envelope thinking suggests that there MAY be a
need for some active flyback switching but the idea has merit and does
overcome a number of issues not well addressed by other means. There
is NO switching at the output side of the mains transformer and the
output can see smooth sinusoidal 50/60 Hz AC if desired. Power factor
is excellent.
The alternative to everyone starting again is that sooner or later
(probably sooner) James gets mad and bangs some heads together. There
have been "faults" all round. Why not try again with the extra
information and understanding niow available?
Russell McMahon
2005\01\31@180937
by
Dwayne Reid
|
|
At 03:05 PM 1/31/2005, Russell McMahon wrote:
>>I'm pursuing an analog solution for my version - I've already got a dirt
>>cheap technique for high frequency ramp and comparitor that I use in a
>>couple of my other SMPS designs. You could, of course, use generate the
>>control signal from a PIC by using low frequency PWM and filtering that
>>to near DC.
>
>If you do the switching at the load end then you can post filter to DC. If
>you do the switching at the input prior to the transformer (as he was
>proposing) you need to filter the PWM but leave the resultant low
>frequency AC for the transformer to handle.
Sorry, what I meant (but probably wasn't clear enough) is that I have a
very inexpensive high-frequency PWM analog circuit block I was planning to
use with this project. The DC input that controls the PWM level in that
circuit could come from a filtered (to near DC) low frequency PWM output
from the PIC.
dwayne
--
Dwayne Reid <EraseMEdwaynerplanet.eon.net>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
Celebrating 20 years of Engineering Innovation (1984 - 2004)
.-. .-. .-. .-. .-. .-. .-. .-. .-. .-
`-' `-' `-' `-' `-' `-' `-' `-' `-'
Do NOT send unsolicited commercial email to this email address.
This message neither grants consent to receive unsolicited
commercial email nor is intended to solicit commercial email.
2005\01\31@193531
by
p.cousens
It's still a SMPS as the controlling element is operating in switch mode
not linear, certainly not phase control
I did propose a phase controlled very simple low cost solution which
seemed to fit most of his requirements
He didn't even bother to comment
Seems more intent on complaining how misunderstood he is.
I wonder if there is a hidden agenda
PC
> {Original Message removed}
2005\01\31@194554
by
p.cousens
I did propose a phase controlled very simple low cost solution
Should have read
I did propose a phase control of a SMPS module
> {Original Message removed}
2005\01\31@203736
by
Russell McMahon
|
|
> It's still a SMPS as the controlling element is operating in switch
> mode
> not linear, certainly not phase control
It's a SMPS by basic definition ie SWITCHED mode power supply.
BUT it differs from what people almost always MEAN when they say SMPS.
SMPS means to most a device that uses reactive storage as a means of
transferring energy at one impedance level to another impedance level.
eg High voltage low current to lower voltage higher current.
His is EFFECTIVELY a resistive switch. ie it does not transform the
energy to a different impedance level by storing it in a reactive
element and then using the element differently. Instead it JUST chops
the input voltage to decrease it. Iin to the switch block = Iout.
By that measure it is not a SMPS as we usually understand the term.
It suffers the large disadvantage of needing a conventional main
transformer as well as a switch. This disadvantage is accepted by some
in exchange for the advantages.
> I did propose a phase controlled very simple low cost solution which
> seemed to fit most of his requirements
He isn't really after a phase controlled solution - that is an
unfortunate terminology which he probably picked up from others using
the technique. The waveform is chopped multiple times per phase, not
once only per half cycle as in traditional techniques. The method has
several advantages over conventional phase control.
> He didn't even bother to comment
> Seems more intent on complaining how misunderstood he is.
I think that if you put yourself in his position and read the full
thread that you would feel a bit got at as well. If you have grown a
thick Olin proof skin, or have an Olin proof filter or are one of his
mates who he never attacks then its easy to not see how someone new
feels about it. He WAS misunderstood, and improperly. He was also rude
and sarcastic in exchange (and technuically wrong in ctriticisng Olins
advice). He initially gave a clear technical description of what he
wanted to achieve and most did not bother to try to understand the
technical description. (I have the advantage of having used this basic
method previously). Olin admits that he didn't understand it so he
skipped the description. If he had also skipped comment and gone on to
be incredibly helpful to someone else, as he often is, then all may be
well. Better to ignore than to wrongly criticise technical input. It
makes you look silly and doesn't help anyone.
> I wonder if there is a hidden agenda
Doubt it. It think he's just a bit like me and Olin :-). A bit
pig-headed and not willing to be pushed around without reacting in
some way or other. We all react differently - Olin gets rude and stops
listening to good input, I spout long diatribes, others .... .
If we keep this up much longer you can cut another notch on Olin's gun
barrel - maybe not a full kill but certainly a powerful assist.
Russell McMahon
2005\01\31@205018
by
David P Harris
|
Hi-
Well, if we can do this then can we also discard the transformer, pass
the reduced voltage AC through a full-wave rectifier and get controlled
DC? I guess the downside is that there is no isolation. Is adding
isolation with a 1:1 transformers, and are they cheap?
David
Russell McMahon wrote:
{Quote hidden}> Here we have several strong willed people (N against 1 - not usually
> very fair) with good ideas talking past each other, failing to
> understand what the other person has said, missing the good points the
> other has offered, being gratuitously rude to each other, incorrectly
> criticising technical input from the other person and generally having
> a good time.
>
> Why not start again and try a little harder. All can learn from this
> technically ( if not socially :-) )
>
> Bob may yet learn that SMPS is not a total swear word and that what he
> is proposing has SOME similarities.
> And 'we all' may realise that Bob is proposing a not new but less
> familiar technique which may be very useful on occasion and which is
> being used extremely successfully in some large real world applications.
>
> A confusion is that bob has used the term "phase control" consistently
> when it's not phase control as our mothers understood the term. The
> method simply consists of PWMing an input (usually mains) 50/60 Hz AC
> waveform so that its amplitude is effectively reduced. The reduction
> is NOT achieved by converting to DC and then using high frequency Ac
> to reconstitute the waveform. There is no intentional flyback or
> energy storage at high frequency in the following transformer. The
> chopped AC is either filtered before it gets to the mains transformer
> OR the chopping rate is such that the transformer does not "see" the
> high frequency as the transformer acts as part of an low pass filter.
> There will be some issues with transformer response to high frequency
> components and back of envelope thinking suggests that there MAY be a
> need for some active flyback switching but the idea has merit and does
> overcome a number of issues not well addressed by other means. There
> is NO switching at the output side of the mains transformer and the
> output can see smooth sinusoidal 50/60 Hz AC if desired. Power factor
> is excellent.
>
> The alternative to everyone starting again is that sooner or later
> (probably sooner) James gets mad and bangs some heads together. There
> have been "faults" all round. Why not try again with the extra
> information and understanding niow available?
>
>
>
> Russell McMahon
2005\01\31@224119
by
Bob S.
|
|
Thank you Russell,
It looks like you RTOFP (Read The Original ******* Post ;-)). Again, I
thank you!
I'll comment below, between your paragraphs, to make it easier for
others to follow.
Russell McMahon wrote:
{Quote hidden}> OK - I'll take a little time here. Listen carefully. I'll try and be
> useful technically and otherwise.
> Good (hopefully) technical comment below, but 1st a very little
> general advice.
>
> People here are generally helpful, but not always. They don't always
> get the point and some are ruder than others. You happen to have been
> exchanging words with one of the ruder ones - but he is also extremely
> competent technically and well experienced in power supplies. You are
> far better off having him as a friend than an enemy. You have reacted
> to his rudeness with rudeness and in the process of accusing him of
> not listening and of offering stupid advice you have failed to listen
> fully and have said technically stupid things about his (quite
> possibly useful) advice because you didn't read what he said
> carefully. Exchanging sarcasm with Olin is seldom profitable for any
> concerned. The best you get is nothing and more often than not you get
> two or more disgruntled people.
Yes, I'm accutely aware of how cetain people on this list tend to
interact with others. I try not to speak to people rudely, unless
spoken to rudely first. But I, like most people, do tend to treat
people the way they have treated me. As you know, I did NOT start the
sacrasm. And your right, I probably should have just ignored it
outright. But nobody likes to be spoken to as if they were an idiot
(especially people with high IQ's). Most people forget, that unless
they upload a schematic or block diagram, you pretty much have to paint
the "picture" of what you are describing with your words, inside the
other person's head. If you do a poor job of that, and the other person
doesn't understand it (because you forgot one or more key elements, or
he/she mis-read your post), the other person may have a totally
different picture painted on their head that what you had originally
intended (See example of what I was thinking Olin was trying to describe
below, as a good example of that).
I have belonged to several various listservers (and some that no longer
exist), and the one common thing that will kill any listserver, is
having even just one or two habitually rude/sarcastic people swat down
any "new comer" questions as if they had no right to ask such a stupid
question in the first place. Most people won't say anything, and duck
and shut their mouth. Some, will take one side or the other, only
pointing out why they think other person's brain was put in backwards
(the ORIGINAL TOPIC is immediately forgotten). Pretty soon (if it's
allowed to continue all the time), nobody is posting anymore, and the
membership just withers away. This listserver's admin is pretty good
about canning people who do nothing but bicker, no matter how smart they
are (or think they are). But there are always a few that run just
inside the ragged edges of what you can get away with, without getting
kicked off. God knows I wouldn't want to have to deside what's over the
line on a daily, or even weekly, basis (thanks for your time and effort
James ;-)
>
> There are indeed a few one time posters who never come back but most
> newcomers do in fact stay. Some people here have trouble reacting well
> to newcomers but we try to stop them abusing them while they get used
> to the friendly family atmosphere :-). In fact, a premium is placed on
> politeness and treating each other well here. Levels of abuse which
> are an accepted part of many lists are frowned on here and active
> steps are taken to "discourage" such. (Try not to find this out
> actively ;-) ).
>
I've actually been on this list since 1999 or 2000, and have seen a few
of the rude'r ones get canned. Like momma always said, " if you don't
have something nice (read constructive) to say, then it's best not to
say anything at all". I've also seen many say "scr*w this, I'm outta
here" after asking about something for a month or two, with either no
response, or rude / non constructive replys.
> I suggest we all start again and try and be nice to each other.
>
Yes, let's do that.
{Quote hidden}> Your original proposal had merit. It is not especially new and has its
> good and bad points, but it had the makings of a good idea.
> I suggest that anyone who has been criticising what they THINK Bob has
> been suggesting have a look at his original proposal. It was
> competently described and commented on and it seems technically sound.
> He didn't describe the end application but he didn't need to. While
> it's always useful to know what people are trying to achieve end to
> end, in this case he said he was keen in implementing a particular
> system. It's entirely valid to suggest alternatives BUT trashing the
> original concept without being aware of what it was is not an
> especially good idea.
>
Praise be Jebus (Pun), he's seen the lighting fixture ;-). Thank you
for acknowledging that.
> [[SUMMARY: PWMing mains AC with an AC switch in series with a
> transformer primary presents variable voltage low frequency AC input
> to the transformer. Advantages are no rectification loss, prospect of
> OK overall efficiencies, excellent power factor, reasonable prospect
> of open loop control, retrofitting to existing mains transformers,
> .... . Still needs low frequency mains transformer. ]]
>
Now if I could have only come up with a description that succinct... ;-).
> If I was trying to meet you spec from scratch I would try either
> something like what you have suggested OR a transformer and rectifier
> (or a SMPS) providing about 2.5 volts DC followed by a buck regulator.
> The output could be very precisely controlled by feedback.
>
I don't know if I mentioned it before, but I did look into doing
something like that. It was more on the order of using a cheap standard
off the shelf 12VAC 4 Amp rated transformer hooked right to the AC line,
and then putting some sort of DC/DC regulator after it. A few web sites
like National.com have online applications (like webench) for doing such
a thing, but the programs I tried didn't like going down to the
extremely low voltages that I needed, or the high amperages either. And
none of them were variable output. I was looking for an "off the shelf"
solution, as I didn't feel comfortable making any type of SMPS from
scratch (I understand the theory, I just have no concept of the math
involved).
{Quote hidden}> Let's have a quick look at what Olin suggested. He proposed a full
> wave rectified unsmoothed DC supply followed by PWM into a
> transformer. The act of PWMing will provide the AC component that the
> transformer seeks. The transformer will be able to be a very small one
> compared to one used on 50 Hz due to the high frequency of the PWM.
> When turned off the transformer primary will "flyback" and will need a
> recirculating diode across it. Lo and behold - you now have the SMPS
> you were trying to avoid :-). Efficiency should be able to be
> reasonably high (= lowish heat) as rectification and switching is done
> at high voltage and low current. What he is suggesting is similar in
> end result to yours but has the very significant advantage of a small
> high frequency transformer and it also allows a simpler switch.
>
Okay, here's what was confusing me about Olin's suggestion. AC line,
rectify it to DC (UNSMOOTHED, NO CAP), and then PWM that at 100hertz
(NOT 100 Kilohertz) into a regular 50/60 hz transformer....
At a 52.9% duty cycle, would it not look something like this on a 60hz
rectified unsmoothed (AKA DC) waveform???
+120VDC
/|
/ / |
/ /| / |
/| |\ / | / |
/ | | \ / | |\ / |
0v / |_________| \/ |_________| \/ |_________|\
ZC ZC ZC ZC
Okay, maybe he meant 100khz (kilohertz) instead I thought. Well, it's
still DC (Direct Current), even if parts of the VOLTAGE waveform have
been chopped up to "look like" high frequency AC, it's still chopped up
DC as far as a "60hz power transformer" is concerned (feed unfiltered
60hz DC into into a tranformer, see what happens...;-). If you read his
and my original posts carefully, you might see where I came to this
conclusion (also look at the basic circuit topology I sent on my last
post). Your paragraph above just happened to kill that confusion for
me, thanks.
Now, if your talking about a "fully blown SMPS topology" (transformer
and all), that setup probably works just fine (and Olin's offering makes
a bit more sense then ;-). I only realized it, when YOU mentioned (your
paragraph above) using a much smaller transformer than the one I was
using. I was still thinking about my original "Phase control" circuit
topology (and transformer), Olin was still pushing a SMPS topology (and
SMPS transformer). This is why Olin's idea didn't work well in my head,
as he forgot to say he was going with a fully blown SMPS at that point,
and not using my existing transformer. If you then think about it, my
comments about his solution are valid, if your under the same assumption
that I was, in that we were all still talking about using a 60hz
transformer.
I know I've been stuborn about sticking to my guns, but keep in mind
that over two or three years ago I had asked how to do this, was steered
(like the proverbial sheep) directly toward SMPS's (went willingly),
spent over a year talking to an engineer, and ended up with a prototype
power supply that was too big, too hot, too many parts, too expensive,
etc, ect.... Basically, not my cup of tea. You've got to understand, at
a certain point I felt like John Kleese (in Monty Python's "Spam" skit),
when being asked by the incipid waiter if he'd like a bit of Spam (for
the umpteenth time) with his tea. Just replace "Spam" with "SMPS", and
it starts to look like the same skit.
Now a 100khz SMPS that is controlled from the primary side only (high
freq. AC output to the load) probably would work great. BUT, wouldn't
that radiate a lot of unwanted RF, especially on that 4 foot cord going
to the actual load??? I don't want to worry about killing people with
pace makers, or have the FCC try to shoot legal bullets at me ;-). In
fact, I believe this SMPS topology was originally looked at during the
design phase for the SMPS prototype that I had built for me, and it was
dropped for that reason. If that's not a problem, or there are simple
ways to get around it (other than shielding everything going to the
load), then PLEASE let me know about it (at which point I'd have to
apologize to Olin for my ignorance of his "better tasting" brand of spam
;-).
{Quote hidden}> Let's try and address you original circuit.
> You can chop the AC as proposed with 2 FETS *BUT* they cannot be in
> parallel antiphase due to the body diodes - they must be in series -
> gate to gate, source to source, input and outputs from the 2 drains.
> Switching is achieved by PWMing the gate to source as required. The
> whole switch floats across the whole AC range so the gate drive needs
> to be floating. There are good ways of doing this BUT for
> experimenting it's better to use an "EASY" way first and come back and
> improve it technically once the rest works OK. An easy way is to use a
> 9v volt battery plus an opto to drive the FET gates. The battery will
> last for weeks of use and a floating isolated supply (or other
> solution) can be implemented to replace it in due course.
>
Yes, two fets in series and at least one inductor to kill the high freq.
PWM from going back into the line. Yep, planned on using a floating
source anyways to power the PIC (VB409 from ST).
> If you have a conventional mains transformer AND you PWM the mains
> supply AND you filter the output to remove the PWM component BUT not
> the 50 Hz AC component THEN you will get a variable amplitude AC
> waveform which the transformer will happily accept. This could be a
> very nice solution but still has a standard mains transformer.
> Switching efficiency is goodish, and there are no rectifiers to
> dissipate power. The PWM filter need not be excessively large. The
> transformer itself will form a reasonably good filter in its own right
> due to its significant inductance.
>
Very close to the topology I was looking to do. On the one's like this
I've read about, they actually remove the PWM component on the high
voltage side with an smallish inductor, before it even gets to the
load. Keep in mind, that it was originally developed for stage lighting
and sound studios (in the k-watt range), so that you didn't have the
bulb element "singing" from using a conventional triac circuit (you'd
absolutely need to filter it before the load then). I just seen other
aspects of this topology that lent themselves well to my application.
This is why I mentioned having small high freq. inductors on either side
of those fet switches.
{Quote hidden}> Let's jump to the PIC question and then look at other things. If you
> use 10 bit control then you have 2^10= 1024 steps. If you want 48 kHz
> then you get about 50 million possible steps/second which is VERY fast
> PWM. It's also humungously fast for FET or other switches. I would be
> surprised if you need either 48 KHz frame rate or 10 bit resolution
> BUT in the absence of knowledge of the final application I may, of
> course, be wrong. I have a design where inaudibility is important and
> I am PWMing hundreds of watts in an electromechanical device that
> attempts to be a speaker. I find 20 to 25 kHz is high enough for most
> people. If you are working with something that heats as its main task
> and it is in air and if it is working into a heat sink (person, wood,
> liquid, other) and if it is moving and if it is fed via leads that
> constitute a significant percentage of the load THEN variations in
> convection/conduction/radiation, air currents, thermal paths, working
> medium and the rest all work to ensure that far far less than 10 bits
> accuracy in the outcomes is possible. Real world engineering often
> doesn't need such. As we don't know the application .... .
>
Yea, 10 bit control is a bit much, at the time I hadn't sat down to do
those basic calculations (was a long post to begin with). I could
probably do it with just 8 bit control. Okay, you've confused me again
;-) 50 million possible steps per second? I thought we were talking
about PWM'ing at around 48khz (per second)? At 10 bits (1024 possible
duty cycles) at 48khz, yea, then your talking about having an extremely
fast 50mhz PIC, but the PWM is still only 48khz.
I think I calculated that a PIC running at 20mhz / 4 = 5mhz actual speed
would only give me something like a 4882.8125 hertz PWM (5mhz / 1024
steps). Way too slow to make this work (I'm assuming the PIC hardware
PWM is sync'd to the uP's code speed, and 1 step = 1 code operation).
At 8 bits (256 steps), I'd get something on the order of only 19607.84
hertz PWM (you could probably still hear a high pitch transformer hum,
but it would probably work). Keep in mind, that the faster the PWM, the
smaller the smoothing inductors can be.... At 48khz PWM at 1024 steps,
your uP (and it's hardware PWM, per step) would have to operate at about
50 million operations per second (49,152,000 hertz), something beyond
most PIC's.
I think I'm doing my math right here? Please correct me if it's wrong.
{Quote hidden}> I am not familiar with all of the latest PIC range but some modern
> processors (eg AVR ATtiny26 have a phase lock loop controlled PWM
> where the PWM frequency is locked at many times the processor clock
> frequency. Without looking at the data sheet AFAIR the PWM can be at
> up to about 50 MHz data rate ! Switching a power FET at such rates is
> another matter. I would not recommend much above 1 Mhz fastest
> switching time unless you are desperately unable to do it any other way.
>
> Feedback control will often make raw control precision less important
> as you can alter the drive to suit what the output is doing.
>
> If you want to try the basic idea you can use far far far lower than
> 10 bits accuracy to trial the basic concept. An analog 555 oscillator
> would do just fine to drive the series connected back to back FETs
> into the transformer primary. If I was doing this I'd start with low
> voltage :-)
Heh, definitely. That's what that varibable transformer thingy sitting
on my bench is fer ;-).
>
> If you are able/willing to tell us the actual application is it would
> help. Sounds something like either a poker-work type heater or a
> person heater/therapy unit.
>
>
Well, lets just say it's closer the the first one you mentioned there.
If this does eventually lead to a new type of power supply, I don't want
any of my competitors to be able to easily just look it up on google,
and then copy it..... I have enough problems with them copying my
design ideas already ;-).
>
>
> Russell McMahon
Again, thanks for reading the original post, and thanks for taking the
time to go over this in a very civil manner. Sorry about my long posts,
but I do try to be as descriptive as possible, so that there is less
confusion about what I am trying to convey.
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.8.2 - Release Date: 1/28/2005
2005\01\31@224826
by
Russell McMahon
|
|
>> ... The method simply consists of PWMing an input (usually mains)
>> 50/60 Hz AC waveform so that its amplitude is effectively reduced.
>> The reduction is NOT achieved by converting to DC and then using
>> high frequency Ac to reconstitute the waveform. There is no
>> intentional flyback or energy storage at high frequency in the
>> following transformer.
> Well, if we can do this then can we also discard the transformer,
> pass the reduced voltage AC through a full-wave rectifier and get
> controlled DC? I guess the downside is that there is no isolation.
> Is adding isolation with a 1:1 transformers, and are they cheap?
Yes, you could definitely (try to) do this transformerless, but there
are large advantages in doing it with a transformer when peak output
voltage is much lower than mains voltage. If you want a 25 amp output
at 2 volts then if you switch direct off mains you are trying to draw
25 amp peaks for short periods off the mains. This is effectively
supplied by the filter, but it's not nice. If you supply the 2.5V
demand from say a 3 or 4v supply then the current drawn at peak load
matches more what you'd see when the switch was run at 100% duty
cycle. High power short term current peaks are much reduced.
RM
2005\01\31@234013
by
William Chops Westfield
On Jan 31, 2005, at 2:35 PM, Russell McMahon wrote:
> The method simply consists of PWMing an input (usually mains) 50/60 Hz
> AC
> waveform so that its amplitude is effectively reduced.
Like a lamp dimmer, right?
I've often wondered whether something like that would be useful for
those
low-current transformer-less supplies we occasionally talk about.
In this case, where we want (IIRC), a low-voltage, high current supply,
I'm not sure I see how it's possible to get there. If you're simply
going to take the ~3V P-P part of the ~300V P-P waveform, you're
discarding
(or ignoring, actually) an awful lot of the power, and the instantaneous
current that you draw during that part of the phase would have to be
very high indeed to average out to the desired current for the rest of
the phase. You can improve on that by doing some additional filtering
and post-regulation, but that would get you closer to an SMPS...
BillW
'[EE] Single Phase AC control via PWM (maybe a chal'
2005\02\01@014748
by
Bob S.
|
|
part 1 7564 bytes content-type:text/plain; charset=us-ascii; format=flowed (decoded 7bit)
They say a picture is worth a 1000 words. Here's a bit over 4000, but
easier to read/understand.
Forward phase control (FPC) using a triac or two SCR's
120+ |\
| \
| \
| \
0 ___________| \___________ ZC
ZC | /
| /
| /
| /
120- |/
"Reverse" phase control (RPC) using a fet or igbt. Don't think ANYONE
considers this a SMPS, just because it uses a Fet or IGBT?
120+ /|
/ |
/ |
/ |
0 / |__________ZC ___________ZC
\ |
\ |
\ |
\ |
120- \|
PWM phase control before any inductors, again, using a fet or igbt (not
totally accurate, can't show high freq. with ansi graphics). Looks like
a smaller amplitude, but smooth, AC sinewave after going through filter
inductor. To me, it just looks like the next logical step from RPC.
120+ /|
/| ||
|| || |\
/| || || || ZC ZC
0 ||_||_||_||_|\_ _ _ _ _ _
|| || || || |/
\| || || ||
|| || |/
\| ||
120- |/
Note that "The reduction is NOT achieved by converting to DC and then
using high frequency Ac to reconstitute the wave form.". Also "There is
no intentional flyback or energy storage at high frequency in the
following transformer." (thanks Russell for those very concise
descriptions/definitions). Put small inductors at both ends of the
Fets, and you can run it right into a light bulb (w/o singing of the
bulb element), or a 60hz transformer. It should also be inherently more
load independent than FPC, RPC, or SMPS, in that it don't care what the
load is (resistive, capacitive, inductive, no load, whatever). The only
two things that I can see in it that are similar to SMPS, is that it is
using a Fet or IGBT (just like Reverse Phase control does), and turning
it on & off at a relatively fast pace. You seem to be equating
relatively high switching frequency with SMPS. If that's the case... At
what magic frequency does the lightswitch on my wall turn into a SMPS?
Assuming I can drink enough coffee to make me jittery enough to attain
that frequency (LOL ;-) . A relatively high switching frequency is
integral to a SMPS, but I don't think I'd call it "exclusive". Maybe
the moniker "Phase control" isn't the proper way to describe it, as
your not turning on/off only once each half cycle delaying the "phase
angle" (hence the name) of where you would turn on a triac. Maybe a
better moniker would be "AC Amplitude Control", or AC/AC converter, as
that better describes it.
What does every SMPS have in common? They first rectify the AC line
voltage into DC, and then remodulate it into high frequency AC. Or in
the case of DC/DC converters, start out with DC to begin with... This
topology is not doing that (no DC involved, no rectifiers used). It
runs in series between the load and the AC line (just like a triac, or
RPC circuit would). It doesn't look like any SMPS topology I've ever seen.
Inductor Inductor
Hot---------(((((--<-Mosfet---Mosfet->--)))))---LOAD
|
60hz AC Line |
|
Neutral-------------------------------------------|
$3000.00 is only "low cost" if your Billy Gates ;-). Just out of
curiosity, what requirements did your design not meet with in my
detailed posted specs?
Sorry, I haven't had time to reply to each and every post, and I
apologize if anyone feels slighted for me not responding to each and
every post on this subject.
I'll just ignore your last two belligerent comments.
p.cousens wrote:
{Quote hidden}>It's still a SMPS as the controlling element is operating in switch mode
>not linear, certainly not phase control
>
>I did propose a phase controlled very simple low cost solution which
>seemed to fit most of his requirements
>
>He didn't even bother to comment
>
>Seems more intent on complaining how misunderstood he is.
>
>I wonder if there is a hidden agenda
>
> PC
>
>
>
>
>
>
>>{Original Message removed}
2005\02\01@032931
by
Russell McMahon
|
|
Nice pictures - they show what you mean well.
> 120+ /|
> /| ||
> || || |\
> /| || || || ZC ZC
> 0 ||_||_||_||_|\_ _ _ _ _ _
> || || || || |/
> \| || || ||
> || || |/
> \| ||
> 120- |/
>
> Note that "The reduction is NOT achieved by converting to DC and
> then
> using high frequency Ac to reconstitute the wave form.". Also
> "There is
> no intentional flyback or energy storage at high frequency in the
> following transformer."
Quick comment here, more anon probably.
Note that the load is subject to a series of peaks with nothing
between. The "reduced" voltage waveform occurs if you average the
power delivered. if you put it on a scope the voltage will be a sparse
sinusoid at FULL voltage.
If the load is pure resistive this will contribute to heating which
will deliver average power in a sinusoidal manner. You show series
inductors to filter this waveform to produce an average voltage to the
load. The inductors will be subject to pulsating voltage which will be
effectively pulses of DC in the short term. Some flyback control may
be required depending on what form the filter takes. As the voltage
reverses polarity each mains half cycle you may also have to gate any
flyback "diodes" accordingly. It may be possible to use a filter which
does not use flyback diodes - effectively it's a low pass filter
subjected to AC at the PWM frequency. I should sit don and draw a
circuit taking account of this. Later.
RM
2005\02\01@032931
by
Russell McMahon
|
|
>> The method simply consists of PWMing an input (usually mains) 50/60
>> Hz AC
>> waveform so that its amplitude is effectively reduced.
>
> Like a lamp dimmer, right?
No - not in the usual case. Lamp dimmers tend to switch once per half
cycle. This switches many times per cycle. I do have a speed
controller which does do just this though. Electronics Australia
circuit - I'll dig up the reference.
> In this case, where we want (IIRC), a low-voltage, high current
> supply,
> I'm not sure I see how it's possible to get there. If you're simply
> going to take the ~3V P-P part of the ~300V P-P waveform,
It's not, fortunately. Power is taken across the whole aveform ina
series of "slices".
> ... and the instantaneous
> current that you draw during that part of the phase would have to be
> very high
As it's taken across the whole cycle current draw peak is load
current. As I notes previously, using a transformer to bring peak
voltage down to slightly above max load voltage helps reduce current.
RM
2005\02\01@041301
by
p.cousens
OK let me start from the beginning
Buy a dimmable electronic l2v halogen lighting transformer
www.skot.co.uk/lighting/transfor/transfor.htm
For around $12 (this is the SMPS part).
Rewind the 12 turns or so on the output transformer
Power with mains via a dimmer
(phase control)
Give me $3000 and I will supply you 10 modified units guaranteed and
ready to go
(did you see the smiley ?)
PC
2005\02\01@070024
by
Russell McMahon
|
|
> Buy a dimmable electronic l2v halogen lighting transformer
> www.skot.co.uk/lighting/transfor/transfor.htm
> For around $12 (this is the SMPS part).
>
> Rewind the 12 turns or so on the output transformer
>
> Power with mains via a dimmer
> (phase control)
Most of the SMPS experienced people on the list would agree that the
task can be satisfied with a suitable SMPS. But Bob, quite apart from
his bad experiences therewith, has stated that he would specifically
like to try the chopped low frequency sine wave approach. It's fine to
suggest alternatives (and I agree that your one has the potential to
work) but it is also useful to discuss his proposed method.
He wants lowest possible heat dissipation in the "plastic box" he is
using, no metal etc. At 50 watts load a 90% efficient will dissipate 5
watts and at 80%, 10 watts. Realistically a SMPS will probably lie in
that range. 10 watts can get annoying. I have a small plastic box with
1 metal face dissipating about 4 watts with no air holes etc (as he
specified AFAIR) and it gets warmer than I'm happy with. Whether his
method is going to produce superior efficiency is uncertain, but
efficiency should be able to be "highish".
RM
2005\02\01@075242
by
olin_piclist
David P Harris wrote:
> Well, if we can do this then can we also discard the transformer, pass
> the reduced voltage AC through a full-wave rectifier and get controlled
> DC?
No. The "reduced voltage AC" is only reduced in its average, not its peak.
A rectifier will ride the peaks. The chopping method was partly relying on
time constants in the transformer to average out the input, and partly on
the final load reacting slowly to RMS voltage anyway (it's a heating
element).
*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
2005\02\01@132513
by
olin_piclist
Bob S. wrote:
> Okay, here's what was confusing me about Olin's suggestion. AC line,
> rectify it to DC (UNSMOOTHED, NO CAP),
No, you would certainly want a cap there. I never said to leave it out.
> and then PWM that at 100hertz
> (NOT 100 Kilohertz) into a regular 50/60 hz transformer....
Yes, this was a compromise just to float the initial idea. If I were
designing this from scratch, I would do what Russell talked about, which is
chopping this rectified AC into a transformer at high frequency (probably
10-100KHz) to allow for a small transformer. However, I was somehow under
the impression that you were trying to make a small modification to an
existing design that ran directly off 60Hz power now, just wanted to add a
regulation front end.
A great deal of the confusion resulted from your not being clear about what
the final goal was, and mixing the problem statement with suggested
solutions. Maybe the details were in there somewhere, but as a general rule
I don't spend my time digging thru a post to understand the problem when I'm
doing it for free. You're going to get a quick read thru, and if I have a
misconception because more work is required to truly understand the
question, my feeling is that's your fault, not mine. Like I said before, if
you don't like my free advice, you can have a full refund.
Anyway, the thought was that this transformer was designed for a 120V 60Hz
sine wave input. We've now got 170V DC available to drive it. There is a
lot more power in a +-170V square wave than in a 120V RMS sine wave. This
means that less than 50% duty cycle would be needed at full power output.
However, the transformer would probably saturate and generally react badly
to a 60Hz pulse train, even if the tips were shorter than 1/2 cycle and
there was dead time in between. The solution is to increase the frequency a
bit, making use of the natural time constants in the transformer. I was
actually thinking a few 100Hz at the time. I started typing that and
realized that I didn't want to get into the frequency issue, since that
would require additional explanation I didn't have time for, and I wasn't
sure this whole solution was even in the right direction in the first place.
So "100Hz" ended up in the message (300Hz would have sounded too specific),
and I figured this would be elaborated on in the ensuing discussion if this
solution was even roughly in the right direction.
> At a 52.9% duty cycle, would it not look something like this on a 60hz
> rectified unsmoothed (AKA DC) waveform???
You are missing two key points, one of which was pointed out in a previous
post.
First, the duty cycle would never be as high as 50%. Even if the
transformer could tolerate a +-170V square wave input, that would be about
twice the power of a 120V RMS sine wave. The maximum duty cycle therefore
would be about 25% for each of the tips.
Second, as was pointed out previously, you wouldn't want to put DC into the
transformer primary. I was envisioning an H bridge drive, although I simply
said "PWM" just to get the basic point accross and not get into the details
at that time. You somehow equated PWM with single ended drive, which is
incorrect.
> Well, it's
> still DC (Direct Current), even if parts of the VOLTAGE waveform have
> been chopped up to "look like" high frequency AC, it's still chopped up
> DC as far as a "60hz power transformer" is concerned
No, this need not be the case at all. Again, I was envisioning an H bridge
drive. I also pointed out that you could use two switching elements instead
of 4 if the primary was center tapped. However, if you get to pick a new
transformer, I'd rather chop at higher frequencies and make it much smaller.
> as he forgot to say he was going with a fully blown SMPS at that point,
I did say "PWM" which pretty much implies SMPS, but I was leaving the
details to subsequent discussion. I figured the details were pointless
until it was determined the overall concept fit the problem.
> I know I've been stuborn about sticking to my guns,
Yes, which can be approriate in some circumstances, but usually not when you
are asking others for a favor. Humility is important when you've got your
hand out. How would you react to someone on the street saying "Yo, ---hole,
gimme a dollar!".
*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
2005\02\01@142340
by
Bob S.
|
|
Hmmm, I forgot to mention one thing in the specs... The user may (at
any time) unplug the end load, usually without turning off the power
supply. I see that the smallest one there has a minium load of 20VA.
Don't know if that would hurt it, or it would just shut down with no
load (and if user would have to turn the PS off and on again).
Also, Because that dimmable halogen light transformer obviously doesn't
compensate for line voltage fluctuations (because it is dimmable), the
Dimmer would have to do that. At which point, why would I even want
this thing? Am I missing the point here? If so, what is the point of
using a device like this?
p.cousens wrote:
{Quote hidden}>OK let me start from the beginning
>
>Buy a dimmable electronic l2v halogen lighting transformer
>www.skot.co.uk/lighting/transfor/transfor.htm
>For around $12 (this is the SMPS part).
>
>Rewind the 12 turns or so on the output transformer
>
>Power with mains via a dimmer
>(phase control)
>
>Give me $3000 and I will supply you 10 modified units guaranteed and
>ready to go
>(did you see the smiley ?)
>
> PC
>
>
>
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.8.2 - Release Date: 1/28/2005
2005\02\01@163643
by
Bob S.
|
|
Comments between your paragraphs below dude.
Olin Lathrop wrote:
{Quote hidden}> Bob S. wrote:
>
>> Hmm, unless I'm totally missing something here, or the basics of AC
>> power transformers has suddenly changed through some quirk of quantum
>> physics... Last time I checked, 60 hz power transformers kinda liked
>> AC, not DC, or pulsed DC.
>
>
> Who said anything about DC? I was suggesting to rectify the AC line,
> then
> PWM that into the transformer primary with feedback adjusting the duty
> cycle
> to get the desired power transfer. The rectified AC line is DC, but that
> doesn't necessarily translate to DC into the transformer primary.
Um, you did? ;-) Maybe my terminology is totally mis-matched here, but
I thought if you "rectify" AC (i.e., run the AC power through a bridge
rectifier), you typically get DC out??? If you then PWM it with one
Fet, you get chopped up DC.
>
> There are several ways to deal with the issue of DC into the transformer.
> First, you can ignore the DC component and the transformer takes it out.
> This appears to be the only method you were thinking of.
Yep, pretty much.
> This is usually
> not the preferred method, especially at high power levels, since the DC
> current transfers no power to the output, reduces overhead to the
> magnetic
> saturation limit of the core, and represents wasted power.
Hence my "smell of burnt wire varnish" comment. 8^]
> Another way is
> to put a capacitor in series, although this is again unwieldy at high
> power
> levels. If you can't control the transformer, then an H bridge input
> drive
> would work. If you have a center tapped primary, then you can use a
> push-pull configuration with only two switching elements.
>
Ah, you didn't mention any H-Bridge. Sorry, ya gotta spell these things
out for me. Don't have a center tapped primary.
>
> Again, only in the limited interpretation of "PWM" to mean single ended
> pulses from ground to the supply voltage. Clearly that is not a suitable
> approach in this case.
>
And that's what I was thinking, running a switch(s) in between the AC
line and the load, just like a Reverse Phase Control (RPC) circuit
would. I thought I had described that on my 3rd post.
{Quote hidden}>> Sigh.... No, I do not necessarily need a nice sine wave to the load. I
>> could just use a triac hooked up to a PIC. Then again, I could just
>> scrap the whole idea, and just sell propane torches to create heat....
>> The whole point here was to try and experiment with A DIFFERENT TYPE OF
>> PHASE CONTROL!!!! Why? READ my original post! Sorry, I didn't realize
>> that everyone here was so set in their ways.
>
>
> Well excuuuuuse me! You came here asking for ideas. If you've
> already made
> up your mind what you want, don't waste everyone's time asking. I
> threw out
> one idea. I don't know if it's optimum for your design. It's your
> design.
> However there is no point ridiculing it, especially since you didn't
> bother
> to understand it first. This is hardly a way to entice people to give
> you
> free advice.
>
Sorry, but if you read my last big reply to Russell, I was feeling a bit
like John Kleese in the Monty Python "Spam skit" at that point (you
being the waiter, and me being John Kleese ;-). Yes, I had a "not so
good experience" with SMPS's in the past, it left a somewhat bad taste
in my mouth, and I didn't particularly want to have it shoved down my
throat again. You weren't the only one expousing the virtures of SMPS's
to me at the time, so I now realize that I did overreact a bit. Now,
If your wanting to talk SMPS's, I'm ready to listen. Please keep in
mind, that I am nowhere near as familiar with them as you are, so try to
put it into laymen's terms if possible (draw pictures if need be, use of
crayon is optional ;-)
Russell mentioned in one of his emails that he thought your idea should
work good for what I wanted (Control the output power from the primary
side only, AC output, small, not hot, etc...). If you read my reply to
that paragraph, you'll see I was worried that the 4 foot cord going to
the load would radiate too much RF, and that I had looked at that option
in the afore mentioned SMPS prototype (we didn't use it for that
reason). Would RF on that output cord be a problem at 2VAC RMS @ 25
amps? Or is that something you wouldn't know unless you actually tried
it? Keep in mind that I'm unable to shield that cord, as I already have
10,000 feet of it sitting on spools already. Also, although you said
100hz, I thought that you maybe meant 100khz.
Also, I forgot to mention one specification, important, but overlooked
at the time. The user, at any given time, may unplug the load
momentarily. Most, if not all, of the SMPS topologies I've look at,
require a minimum load of some sorts. Is that a problem, or is it
something easy to get around?
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.8.2 - Release Date: 1/28/2005
2005\02\01@171359
by
olin_piclist
Bob S. wrote:
> I was worried that the 4 foot
> cord going to
> the load would radiate too much RF, and that I had looked at that
> option in the afore mentioned SMPS prototype (we didn't use it for
> that
> reason). Would RF on that output cord be a problem at 2VAC RMS @ 25
> amps?
I don't know. The switching is happening at the other side of the
transformer, which will filter it a bit. You can certainly put capacitors
accross the transformer output and from each side to ground. Keeping the
sides symmetric will help because the differential mode signals will largely
cancell, and the common mode signal comes only from capacitive coupling thru
the transformer and therefore is much easier to squash. Series inductors
would be nice, but at 25A those will probably be too big and expensive for
this application.
> Also, I forgot to mention one specification, important, but overlooked
> at the time. The user, at any given time, may unplug the load
> momentarily. Most, if not all, of the SMPS topologies I've look at,
> require a minimum load of some sorts. Is that a problem, or is it
> something easy to get around?
It needs to be considered up front, but it should be possible to deal with
it. A lot has to do with how the feedback works. Ideally the PWM gets
cranked back to nearly nothing if the load is removed.
*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
2005\02\01@193259
by
Russell McMahon
> Ah, you didn't mention any H-Bridge. Sorry, ya gotta spell these
> things out for me. Don't have a center tapped primary.
Clarification (possibly unneeded) - an H bridge will drive a single
winding primary. Each end has a switching device going to high and low
side of device. Pairs of diagonally opposite sides are switched on
alternately. Looks like an H.
RM
2005\02\01@200740
by
p.cousens
Open circuit would not be a problem, do you have to change the unit
every time a lamp blows?
No just change the bulb and the bulb lights, they don't shutdown for
open circuit.
They do shutdown/foldback for shorts
They also softstart but only at max load
Why use one? Because there cheep, efficient, light, small, plastic cased
and almost a drop in replacement for your transformer and I thought you
wanted to avoid metal
> {Original Message removed}
2005\02\01@215538
by
Chen Xiao Fan
|
It is not really like lamp dimmer (often use SCR or similar with phase
control). Though it is correct that this can be used in the
low-current transformer-less supplies.
Supertex SR036/SR037 (Inductorless, Dual Output Off-Line Regulators) are
doing exactly this. They are used after the rectify and chop out the
voltage above the threshold voltage. It can be used for drive
relay or LEDs directly from the mains. They are very interesting ICs.
Of course we can also doing it with discretes.
Xiaofan
{Quote hidden}>Date: Mon, 31 Jan 2005 20:40:16 -0800
>From: William "Chops" Westfield <
RemoveMEwestfwEraseMEEraseMEmac.com>
>On Jan 31, 2005, at 2:35 PM, Russell McMahon wrote:
>> The method simply consists of PWMing an input (usually mains) 50/60 Hz
>> AC waveform so that its amplitude is effectively reduced.
>
>Like a lamp dimmer, right?
>I've often wondered whether something like that would be useful for
>those low-current transformer-less supplies we occasionally talk about.
>
>In this case, where we want (IIRC), a low-voltage, high current supply,
>I'm not sure I see how it's possible to get there. ...
><snip>
>
>BillW
>
2005\02\01@223401
by
Chen Xiao Fan
|
|
I have not follow this thread closely so hopefully I am not talking
nonsense here.
However looking from the specifications it is definitely a challenge!
The output voltage is so low and the output current is so high.
Therefore it is very much like VRM for CPUs like Pentium 4.
I will say (1) and (2) are already very hard with constraints like
(5) (6) and (7). If you can get all these fulfilled, Intel and AMD
will be very interested.
In the VRM (voltage regulator module) application, one of the approach
is AC VRM and once Intel was promoting it but finally dropped the idea
since the technology is not mature and the DC VRM still works. For
AC VRM, input is rather fixed (high frequency 20V AC? or 48V AC bus)
and the output voltage is in the range of (0.5 to 1.5V?) with
32/64 steps (??). I do not have the paper with me so I am not
100% sure about the values. The output current may be higher than
your requirement but if 25A can be reached, 100A can be achieved
through multi-phase as well.
Anyway it is very interesting things to do though I think it is not
that simple. There are ideas around for AC VRM which may be useful.
There are already some applications in the server/workstation arena.
Of course, in the AC VRM application, the AC bus is high frequency
and the output regulation is very important. This is different
from your application.
Xiaofan
Date: Sun, 30 Jan 2005 15:38:02 -0600
From: "Bob S." <RemoveMEop1cwkspam_OUTKILLspamPunkAss.com>
<snip>
To put it "simply", here are the basic specs.
1) Able to generate voltages from 0.3 to 2 volts RMS (AC or DC, I don't
Care) in at least 256 distinct steps in the output range.
2) Able to generate over 25 amperes on the output at the top end of the
range.
3) Line voltage compensation (output is same for a given setting, no
matter what the line voltage is reading (90 to 140vac)).
4) Linear output range (hard to do with AC triac phase control).
5) It CANNOT generate gobs of heat (again: plastic box, limited sized
heatsink available, no cooling fans).
6) It CANNOT generate gobs of RF.
7) Doesn't have prohibitively high price tag, parts wise.
Some "features" that would be nice to have, but are not requirements
per-se....
A) Can detect an output fault condition (too many amps being drawn?
Shut down, maybe throw out an error code to an LED display).
B) Can automatically work with either 110 or 220 VAC input. (PWM'd AC
phase control could easily do this w/o an extra transformer winding).
C) Could have the output range cut in half (0.15 to 1 V RMS) via an
adjustment to PIC program (user could toggle range by pressing a button)
D) Minimal parts design! Less parts = less things that can go wrong,
easier to build, easier to fix. I.e., the "KISS" rule....
<snip>
2005\02\01@230651
by
Chen Xiao Fan
|
|
Good explanations, Russell. It is very difficult to directly
derive anything above 48V to something around 1V. The efficiency
suffered a lot here due to low duty ratio.
This is the same in the VRM world. 5V (1st generation Pentium I/II)
VRMs are easier, 12V VRMs are much more difficult (Pentium III/IV)
but can still be achieved without a transformer. In 48V VRMs (
server/workstation/dual core??), you need a transformer here if
not resorting back to (48V -> 12V --> 1V). Efficiency is the
key here.
Sorry about the VRM stuff if you think they are not relevant here
but I think they are really related.
Xiaofan
{Quote hidden}>Date: Tue, 01 Feb 2005 16:32:27 +1300
>From: Russell McMahon <
RemoveMEapptechTakeThisOuTspamparadise.net.nz>
>
>>> ... The method simply consists of PWMing an input (usually mains)
>>> 50/60 Hz AC waveform so that its amplitude is effectively reduced.
>>> The reduction is NOT achieved by converting to DC and then using
>>> high frequency Ac to reconstitute the waveform. There is no
>>> intentional flyback or energy storage at high frequency in the
>>> following transformer.
>
>> Well, if we can do this then can we also discard the transformer,
>> pass the reduced voltage AC through a full-wave rectifier and get
>> controlled DC? I guess the downside is that there is no isolation.
>> Is adding isolation with a 1:1 transformers, and are they cheap?
>
>Yes, you could definitely (try to) do this transformerless, but there
>are large advantages in doing it with a transformer when peak output
>voltage is much lower than mains voltage. If you want a 25 amp output
>at 2 volts then if you switch direct off mains you are trying to draw
>25 amp peaks for short periods off the mains. This is effectively
>supplied by the filter, but it's not nice. If you supply the 2.5V
>demand from say a 3 or 4v supply then the current drawn at peak load
>matches more what you'd see when the switch was run at 100% duty
>cycle. High power short term current peaks are much reduced.
2005\02\02@013809
by
Chen Xiao Fan
|
|
I think there problem is that SMPS is too widely (or too narrowly?)
defined. Actually it is better called Power Converters (to include
AC/AC, AC/DC, DC/AC and DC/DC, switching or phase control, etc). All
are studied under the subject of Power Electronics.
So your approach is as following.
60Hz AC Input -> Chopped 60Hz AC -> ||60Hz Tran}||
-> Low Voltage 60Hz AC output.
By controlling the chopping of two back to back MOSFET/IGBT
you will get the desired AC output voltage. Do you need a feedback?
Maybe not since you do not need a very accurate output voltage.
The other approach is as following.
60Hz AC Input -> rectifier -> DC -> PWM -> High f AC ->
||High f Tran ||->Low Voltage DC.
or
60Hz AC Input -> rectifier -> DC -> PWM -> High f AC ->
||High f Tran ||->Low Voltage high f AC.
I think both approach are possible and both are not so easy to
achieve using single stage considering the output voltage
is so low. Maybe a two stage is easier/better.
60Hz AC Input -> Chopped 60Hz AC -> ||60Hz Tran}||
-> intermediate low voltage (say around 12V) 60Hz AC
-> synchronous rectifier -> Low voltage DC (around 12V)
-> synchronous rectifier -> Low Voltage DC
(can use single phase VRM circuits in the last step)
or
60Hz AC Input -> PFC -> High V DC -> PWM -> ||High f Tran ||
-> synchronous rectifier -> intermediate low voltage (say around 12V) DC
-> synchronous rectifier -> Low Voltage DC
(can use single phase VRM circuits in the last step)
Notice I put synchronous rectifier for all the low voltage side
(high current side) to get higher efficiency. The first
synchronous rectifier may be left out for simplicity but
but the efficiency will be a bit lower.
Now the difference of the two approaches is the front end.
In the first approach, a relative big 60Hz transformer is
used but it seems the circuit will a bit easier. In the
second approach, the high frequency transformer is smaller
but the circuit is a bit more complicated. PFC may not
be necessary if we do not care about the harmonics since
the power level is not that high. The first approach may
be better in this aspect as well.
However, with both approaches I will doubt if we can achieve
very high efficiency with small budget. The second approach
looks like the PC silver box (power supply) + VRM and we sure
know that the efficiency is not that fantastic.
Xiaofan
{Quote hidden}> Date: Tue, 01 Feb 2005 00:47:29 -0600
> From: "Bob S." <
EraseMEop1cwkspamspamBeGonePunkAss.com>
>
> They say a picture is worth a 1000 words. Here's a bit over
> 4000, but
> easier to read/understand.
>
> <snip>
>
> PWM phase control before any inductors, again, using a fet or
> igbt (not
> totally accurate, can't show high freq. with ansi graphics).
> Looks like
> a smaller amplitude, but smooth, AC sinewave after going
> through filter
> inductor. To me, it just looks like the next logical step from RPC.
>
> 120+ /|
> /| ||
> || || |\
> /| || || || ZC ZC
> 0 ||_||_||_||_|\_ _ _ _ _ _
> || || || || |/
> \| || || ||
> || || |/
> \| ||
> 120- |/
> <snip>
> What does every SMPS have in common? They first rectify the AC line
> voltage into DC, and then remodulate it into high frequency
> AC. Or in
> the case of DC/DC converters, start out with DC to begin
> with... This
> topology is not doing that (no DC involved, no rectifiers used). It
> runs in series between the load and the AC line (just like a
> triac, or
> RPC circuit would). It doesn't look like any SMPS topology
> I've ever seen.
>
> Inductor Inductor
> Hot---------(((((--<-Mosfet---Mosfet->--)))))---LOAD
> |
> 60hz AC Line |
> |
> Neutral-------------------------------------------|
>
>
> $3000.00 is only "low cost" if your Billy Gates ;-). Just out of
> curiosity, what requirements did your design not meet with in my
> detailed posted specs?
>
> Sorry, I haven't had time to reply to each and every post, and I
> apologize if anyone feels slighted for me not responding to each and
> every post on this subject.
> <snip>
2005\02\02@114753
by
Martin K
|
|
Bob S. wrote:
> Comments between your paragraphs below dude.
Same here dude.
>> Bob S. wrote:
>
> Um, you did? ;-) Maybe my terminology is totally mis-matched here, but
> I thought if you "rectify" AC (i.e., run the AC power through a bridge
> rectifier), you typically get DC out??? If you then PWM it with one
> Fet, you get chopped up DC.
A transformer doesn't have to have AC strictly speaking. It can be a
flyback or forward converter (with "chopped up DC", of course). But this
is an SMPS, and the transformer is not your standard laminate core type.
{Quote hidden}> Sorry, but if you read my last big reply to Russell, I was feeling a bit
> like John Kleese in the Monty Python "Spam skit" at that point (you
> being the waiter, and me being John Kleese ;-). Yes, I had a "not so
> good experience" with SMPS's in the past, it left a somewhat bad taste
> in my mouth, and I didn't particularly want to have it shoved down my
> throat again. You weren't the only one expousing the virtures of SMPS's
> to me at the time, so I now realize that I did overreact a bit. Now,
> If your wanting to talk SMPS's, I'm ready to listen. Please keep in
> mind, that I am nowhere near as familiar with them as you are, so try to
> put it into laymen's terms if possible (draw pictures if need be, use of
> crayon is optional ;-)
<http://schmidt-walter.fbe.fh-darmstadt.de/smps_e/smps_e.html>
Switching Power Supply Design
by Abraham I. Pressman "There are about 14 basic topologies (basic block
diagrams) commonly used to implement a switching power supply..."
ISBN: 0070522367
Switchmode Power Supply Handbook
by Keith Billings "Although the various switchmode conversion techniques
are often very different in terms of circuit design, they have, over
many years, developed very similar basic functional characteristics
which have become generally accepted industry standards."
ISBN: 0070067198
Power Supply Cookbook
by Marty Brown
ISBN: 075067329
And if you have a specific question about SMPS you are welcome to ask
on-list.
{Quote hidden}>
> Russell mentioned in one of his emails that he thought your idea should
> work good for what I wanted (Control the output power from the primary
> side only, AC output, small, not hot, etc...). If you read my reply to
> that paragraph, you'll see I was worried that the 4 foot cord going to
> the load would radiate too much RF, and that I had looked at that option
> in the afore mentioned SMPS prototype (we didn't use it for that
> reason). Would RF on that output cord be a problem at 2VAC RMS @ 25
> amps? Or is that something you wouldn't know unless you actually tried
> it? Keep in mind that I'm unable to shield that cord, as I already have
> 10,000 feet of it sitting on spools already. Also, although you said
> 100hz, I thought that you maybe meant 100khz.
There should be a low pass filter on anything that could potentially
radiate a lot of noise (usually power supplies). If you're looking to
get approval by a regulatory agency (it sounds like you're mass
producing these) it would have to have such a filter.
> Also, I forgot to mention one specification, important, but overlooked
> at the time. The user, at any given time, may unplug the load
> momentarily. Most, if not all, of the SMPS topologies I've look at,
> require a minimum load of some sorts. Is that a problem, or is it
> something easy to get around?
An SMPS that requires a minimum load is a poorly designed SMPS. I'm not
pushing the SMPS topic again, you brought it up again.
--
Martin K
http://wwia.org/sgroup/biofuel/
2005\02\02@122152
by
Alan B. Pearce
> Sorry, but if you read my last big reply to Russell, I was feeling a bit
> like John Kleese in the Monty Python "Spam skit" at that point (you
> being the waiter, and me being John Kleese ;-). Yes, I had a "not so
> good experience" with SMPS's in the past, it left a somewhat bad taste
> in my mouth, and I didn't particularly want to have it shoved down my
> throat again. You weren't the only one expousing the virtures of SMPS's
> to me at the time, so I now realize that I did overreact a bit. Now,
> If your wanting to talk SMPS's, I'm ready to listen. Please keep in
> mind, that I am nowhere near as familiar with them as you are, so try to
> put it into laymen's terms if possible (draw pictures if need be, use of
> crayon is optional ;-)
The biggest problem with a SMPS is that when something goes wrong, things go
bang very fast, not just a gentle letting the smoke out of one component. If
it is an off-mains SMPS then if one component fails on the mains side, then
all semiconductors on the mains side will probably head south with it. It
requires a well versed knowledge of device operating characteristics to
ensure that a design is good and safe.
For your purposes I suspect that a switching regulator off the secondary of
a conventional transformer may be an easier way to go, if you are not well
versed in doing SMPS stuff direct off mains. I see no real reason why you
could not have a DC supply around 12V or so, and then do a switching
regulator to the very low voltage/high current level you seem to be seeking.
This would be much more easily achievable, and I doubt that the losses would
be that much different to having an off-line SMPS, as most of the losses
will be in the low voltage rectifier area anyway.
2005\02\02@140904
by
Peter Johansson
Alan B. Pearce writes:
> The biggest problem with a SMPS is that when something goes wrong, things go
> bang very fast, not just a gentle letting the smoke out of one
> component.
Anyone who has ever worked with enough PCs with cheap power supplies
can attest to how frequently this happens, often releasing the smoke
not only from the power supply but at least one chip on every
component in the PC.
-p.
2005\02\02@182908
by
Roland
Pulled off a newsletter, may be of interest. Have not looked at the article;
***Power Sources***
Dense amps
Artesyn Technologies offers its ultracompact POL converter modules
that deliver up to 30 amps at 0.8 to 3.63 volts for densely packed
ASICs, network processor and DSP applications. The company says
that's 75 percent more current than provided by comparably sized
products.
http://newsletter.eeproductcenter.com/cgi-bin4/DM/y/elwj0Fza5o0FrC0DBEg0Ao
Regards
Roland Jollivet
More... (looser matching)
- Last day of these posts
- In 2005
, 2006 only
- Today
- New search...
