How does anyone cut the leads on crystal oscillators? I used some wire
cutters and after a week the crystal failed.... I've heard before that you
need to do it a different way to avoid the shockwaves going up the lead and
cracking the crystal but how does anyone do it?? Is it better to do it
before soldering it in?? Alternatively if crystals are "weak" would a
resonator be a better option in an automotive setting? What does anyone
think??
TIA,
Ben
ps I've programmed some 16F877s using the modified NOPPP software and they
work just fine (the above was on 16F84s!).
Source code: website.lineone.net/~btbromilow/noppp2.c
Zipped executable website.lineone.net/~btbromilow/noppp2.zip
At 04:52 PM 8/16/01 +0100, Benjamin Bromilow wrote:
>Hi all,
>
>How does anyone cut the leads on crystal oscillators? I used some wire
>cutters and after a week the crystal failed.... I've heard before that you
>need to do it a different way to avoid the shockwaves going up the lead and
>cracking the crystal but how does anyone do it?? Is it better to do it
>before soldering it in?? Alternatively if crystals are "weak" would a
>resonator be a better option in an automotive setting? What does anyone
>think??
You would probably be better off to trim the leads after the xtal is
soldered in place.
However, I don't think that the lead trim is what caused the failure.
Crystals have a maximum drive power, and if you exceed that, you can
shatter them.
You can work out the drive power in circuit, using the voltage across the
crystal (scope both sides, and pay attention to phase) and the impedance of
the crystal at resonance, either from the spec sheet, or direct measurement.
I would have a link to http://www.findu.com/cgi-bin/find.cgi?KC6ETE-9 here
in my signature line, but due to the inability of sysadmins at TELOCITY to
differentiate a signature line from the text of an email, I am forbidden to
have it.
I have heard that it is better to shear crystal leads than to nip them.
Use cutters whose blade pass by like scissors, rather than blades that
meet head on. I have never noticed a difference, but a always trim the
leads AFTER they are soldered. I would expect any failure to occur
within a few seconds at most.
Absolutely, I thought the crystal would fail immediately. However, I'm now
in the situation that the oscillator only oscillates if I push it slightly
sideways. It then clicks and starts oscillating. I've checked and resoldered
it and the contact is good. I can only conclude that there is an intermitent
contact internally. When I push it sideways I even use insulated probes so
capacitance/resistance shouldn't be the answer. The clicking noise makes me
think more mechanical......
> I have heard that it is better to shear crystal leads than to nip them.
> Use cutters whose blade pass by like scissors, rather than blades that
> meet head on. I have never noticed a difference, but a always trim the
> leads AFTER they are soldered. I would expect any failure to occur
> within a few seconds at most.
>
> Sherpa Doug
>
> > {Original Message removed}
>How does anyone cut the leads on crystal oscillators? I used some wire
>cutters and after a week the crystal failed.... I've heard before that you
>need to do it a different way to avoid the shockwaves going up the lead and
>cracking the crystal but how does anyone do it?? Is it better to do it
>before soldering it in??
>Ben
Solder them in first, and the board will absorb some of the shock. The
way I cut the leads is to nick them with the cutters (squeeze lightly
and then go around the wire) and then bend them back and forth until
they break off. Still, as Dave pointed out, I'm not convinced this
is a big problem with crystals. I do it because it's easy to do.
A snip that makes a cut-off fly across the room will cause
the equivalent force to go into the part that didn't. Maybe
you weakened the crystal and made it susceptible to
over-driving failure. As noted in threads about using crystals
in rocketry and projectile applications, crystals will tolerate
mechanical stresses in some directions but not others
At 12:31 AM 8/17/01 +0100, you wrote:
>Absolutely, I thought the crystal would fail immediately. However, I'm now
>in the situation that the oscillator only oscillates if I push it slightly
>sideways. It then clicks and starts oscillating. I've checked and resoldered
>it and the contact is good. I can only conclude that there is an intermitent
>contact internally. When I push it sideways I even use insulated probes so
>capacitance/resistance shouldn't be the answer. The clicking noise makes me
>think more mechanical......
Sounds like you fractured ("busted" to be technical about it) the glass-metal
seal on a lead. Sure you (or someone else) didn't crank the body of the
crystal
around, stressing the leads (after soldering)? Maybe in a misplaced attempt
to straighten it so it looked good? Just a though..
Best regards,
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..." "The Journey is the reward" speffKILLspaminterlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro,
It does sound like a "cold" solder/weld joint on the lead.....however,
I'd like to pass on a bit of "crystal" experiment of my own. Years ago, I
bought a VCR that had a clock that ran about an hour fast every day. My
previous wife said "your an electronics technician, fix it". Thinking about
it, I remembered an old electronics project book that had a clock with a
CD4049 divide by "N" chip and a crystal. I looked inside and saw two TV
colorburst crystals. Sure enough, there was a CD4049 next to one of them. I
took a screwdriver tip and knocked very hard on the crystal case and put the
VCR back together. Sure enough, the clock then worked accurately. The
crystal was operating in a second mode of resonance. I went to work and
told some other people this story and got alot of skeptical comments from
one guy. About 3 months later, same person humbly told me he had worked on
a similiar VCR with the same problem and that had been the fix.
Regards,
Mike
{Original Message removed}
At 10:57 PM 8/16/01 -0000, you wrote:
>Spehro,
> It does sound like a "cold" solder/weld joint on the lead.....however,
>I'd like to pass on a bit of "crystal" experiment of my own. Years ago, I
>bought a VCR that had a clock that ran about an hour fast every day. My
>previous wife said "your an electronics technician, fix it". Thinking about
>it, I remembered an old electronics project book that had a clock with a
>CD4049 divide by "N" chip and a crystal. I looked inside and saw two TV
>colorburst crystals. Sure enough, there was a CD4049 next to one of them. I
>took a screwdriver tip and knocked very hard on the crystal case and put the
>VCR back together. Sure enough, the clock then worked accurately. The
>crystal was operating in a second mode of resonance. I went to work and
>told some other people this story and got alot of skeptical comments from
>one guy. About 3 months later, same person humbly told me he had worked on
>a similiar VCR with the same problem and that had been the fix.
>Regards,
>Mike
Interesting. I've seen "dead" crystals of poor quality that were actually
shorted to the case. If the case had not been grounded, the crystal might
not have appeared to fail. The manufacturer, I believe, never tested that
aspect of the crystals.
Note here, I'm suggesting that the glass seal is broken, so that when the
crystal case is moved it's actually yanking the crystal (something like
a disk usually) back and forth inside the metal case. If you cut open an
old crystal, you'll see what I mean.
Best regards,
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam.....interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>How does anyone cut the leads on crystal oscillators? I used some wire
>cutters and after a week the crystal failed.... I've heard before that you
>need to do it a different way to avoid the shockwaves going up the lead and
>cracking the crystal but how does anyone do it?? Is it better to do it
>before soldering it in??
You can get shockless side cutters. These have blades that have a flat
surface on the side that you put towards the component. They will produce a
flat (or very nearly so) surface on the end of the component lead instead of
the peaked end that you will see on the other part of the lead. Most cutters
will produce peaked ends on both parts of the lead.
We have to use these shockless cutters in assembling PCB's for space flight
hardware to minimise the shock on component leads.
Having said that I doubt that this is the source of your problem unless
something has happened to damage the glass-metal seal the lead comes
through. If that has cracked then the wire going to the crystal will move
around relative to the case and may give problems.
I suspect that your real problem is the crystal has been dropped at some
stage (or suffered some other form of sudden shock during shipping maybe)and
the crystal itself is chipped or the support wires have bent so it touches
the case. Either way your only way to get reliable starting is to replace
it.
>> >shockwaves going up the lead
>
>A snip that makes a cut-off fly across the room will cause
>the equivalent force to go into the part that didn't. Maybe
>you weakened the crystal and made it susceptible to
>over-driving failure. As noted in threads about using crystals
>in rocketry and projectile applications, crystals will tolerate
>mechanical stresses in some directions but not others
Not when you use shockless cutters (see my previous post on this thread) The
cutters produce a flat cut on one side and that means there is no wedge
formed to push the lead back in that direction. All the energy goes into
pushing the portion of the lead with the wedge on it. I will try some ASCII
art.
| /
--------------| |/ /--------
component | / excess
lead | \ lead
--------------| |\ \--------
| \
<- no force cutter force applied ->
blades
> Use cutters whose blade pass by like scissors, rather than blades that
> meet head on. I have never noticed a difference.....
The best cutters are completely flat one one side, and tapered on the other.
The flat side causes no shock at all. The tapered side will shoot the
off-cut across the room.
Try trimming your toe-nails with either side. You will DEFINITELY notice the
difference !!!
> Try trimming your toe-nails with either side. You will
> DEFINITELY notice the difference !!!
Remind me never to ask to borrow your cutters Zim. Not as
bad as shaking hands with an old friend you meet leaving
a sperm bank but.......
I've a very nice pair of ultra-flush cutters that are so sharp
and well-aligned they just breeze through leads. I don't
think I could make anything less than spring steel fly. They
cost, but it was worth it to be able to make a nice flat cut
as close to the PCB as you want to get. What's the old
saying - "a bad workman blames his tools" ?
Barry Gershenfeld wrote:
>
> >How does anyone cut the leads on crystal oscillators? I used some wire
> >cutters and after a week the crystal failed.... I've heard before that you
> >need to do it a different way to avoid the shockwaves going up the lead and
> >cracking the crystal but how does anyone do it?? Is it better to do it
> >before soldering it in??
> >Ben
>
> Solder them in first, and the board will absorb some of the shock. The
> way I cut the leads is to nick them with the cutters (squeeze lightly
> and then go around the wire) and then bend them back and forth until
> they break off. Still, as Dave pointed out, I'm not convinced this
> is a big problem with crystals. I do it because it's easy to do.
I always use hot melt glue to hold the can down,
THEN solder it, THEN cut the leads with nice sharp
cutters. Never had a failed crystal. Recently i've
switched to Murata ceramic resonators for anything
that doesn't need a perfect clock speed, these
resonators are very robust.
-Roman
Benjamin Bromilow wrote:
>
> Absolutely, I thought the crystal would fail immediately. However, I'm now
> in the situation that the oscillator only oscillates if I push it slightly
> sideways. It then clicks and starts oscillating. I've checked and resoldered
> it and the contact is good. I can only conclude that there is an intermitent
> contact internally. When I push it sideways I even use insulated probes so
> capacitance/resistance shouldn't be the answer. The clicking noise makes me
> think more mechanical......
In the reapir industry any crystal behaviour
like this is instantly declared a "bad" crystal
and goes straight in the bin. :o)
-Roman
Mike Kendall wrote:
>
> Spehro,
> It does sound like a "cold" solder/weld joint on the lead.....however,
> I'd like to pass on a bit of "crystal" experiment of my own. Years ago, I
> bought a VCR that had a clock that ran about an hour fast every day. My
> previous wife said "your an electronics technician, fix it". Thinking about
> it, I remembered an old electronics project book that had a clock with a
> CD4049 divide by "N" chip and a crystal. I looked inside and saw two TV
> colorburst crystals. Sure enough, there was a CD4049 next to one of them. I
> took a screwdriver tip and knocked very hard on the crystal case and put the
> VCR back together. Sure enough, the clock then worked accurately. The
> crystal was operating in a second mode of resonance. I went to work and
> told some other people this story and got alot of skeptical comments from
> one guy. About 3 months later, same person humbly told me he had worked on
> a similiar VCR with the same problem and that had been the fix.
> Regards,
> Mike
Hi Mike, not doubting your story, but many of the VCRs
have crystals held down by glue, and the glue goes
conductive and causes osc speed changes and stopped
oscillation. I have seen similar things which were
fixed when the crystal was "cracked" away from the glue
or the glue removed. :o)
-Roman
> Not when you use shockless cutters (see my previous post on this thread)
The
> cutters produce a flat cut on one side and that means there is no wedge
> formed to push the lead back in that direction. All the energy goes into
> pushing the portion of the lead with the wedge on it. I will try some
ASCII {Quote hidden}
> art.
>
>
> | /
> --------------| |/ /--------
> component | / excess
> lead | \ lead
> --------------| |\ \--------
> | \
> <- no force cutter force applied ->
> blades
Your physics doesn't add up. If "all the energy" goes into the excess lead,
it will presumably be converted to kinetic energy of that lead causing it to
fly off. Since the excess lead goes from still to a moving state, its
momentum was increased, which means some impulse (force over a time period)
was applied to it. According to Newton and conservation of momentum,
EXACTLY THE SAME impulse is applied in the other direction to whatever
pushes the excess lead away. In other words, whenever a lead flys off, a
"shock" has been delivered to the board, regardless of the cutting
mechanism.
The shockless cutters I've seen are shockless because they cut the leads in
such a way that they don't fly off. They work more like scissors than wire
cutters. The lead is sheared sideways instead of pinched.
This all being said, I agree that it is unlikely that trimming the crystal
lead on the other side of the board after soldering caused the problem.
How about using a jewelers file. I've done this for cutting steel cable
going through housings. The cutters would fray the wire no-matter how it
was cut. A piece of heat shrink, then filing through the heat shrink and
cable (in this case a lead) will really work wonders as far as shock
elimination and fraying.
Regards,
Mike
{Original Message removed}
Funny tread this have become.
I might eat what I'm about to say but I really think we are filtering
mosquitoes and let trough elephants.
I'm sure that brute force can damage crystals but generally I believe that
problem with bad crystal is just a bad crystal. I also believe that you need
to take a look at the oscillator design to find out if there is to high
amplification that might damage the crystal. There is a bunch of bad
crystals in the market and you need to find a crystal that is good enough
for your design and test it extensively before you decide to use it. When
you have found a crystal that do not brake for as little as cutting the wire
you are on the right track. Do some electrical testing and then use that
crystal type from that specific manufacturer in the future. This is my
suggestion to the problem. Some people might argue about that to high
amplification can damage a crystal but I'm just superstitious and have that
belief.
Niklas Wennerstrand
How about using a jewelers file. I've done this for cutting steel cable
going through housings. The cutters would fray the wire no-matter how it
was cut. A piece of heat shrink, then filing through the heat shrink and
cable (in this case a lead) will really work wonders as far as shock
elimination and fraying.
Regards,
Mike
{Original Message removed}
Having spent far too many years repairing rf equipment and replacing many crystals over the years, I can say without hesitation that if you solder the xtal in place before cutting the leads, you are not going to harm the crystal by cutting the leads. If the xtal does not work, just remove it, install a new one, solder, cut leads with "SHARP" cutters, preferably then ones mentioned below but if they are not available, use flush cutting diagonals when possible and cut, not snip the leads. (There IS a difference!) The crystal unit often breaks internally where the crystal wire attaches to the lead that exits the case. Occasionally, it is just a poor (solder??)connection at that point. It can sometimes be temporarily FIXED by soldering (it reheats the junction of the lead to the small wire that attaches to the crystal itself) but is best to replace it cause you can expect it to come back on you. {Quote hidden}
>
> From: "Alan B. Pearce" <KILLspamA.B.PearceKILLspamRL.AC.UK>
> Date: 2001/08/17 Fri AM 09:02:05 EDT
> To: RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU
> Subject: Re: [PIC]: Oscillator leads
>
> >> >shockwaves going up the lead
> >
> >A snip that makes a cut-off fly across the room will cause
> >the equivalent force to go into the part that didn't. Maybe
> >you weakened the crystal and made it susceptible to
> >over-driving failure. As noted in threads about using crystals
> >in rocketry and projectile applications, crystals will tolerate
> >mechanical stresses in some directions but not others
>
> Not when you use shockless cutters (see my previous post on this thread) The
> cutters produce a flat cut on one side and that means there is no wedge
> formed to push the lead back in that direction. All the energy goes into
> pushing the portion of the lead with the wedge on it. I will try some ASCII
> art.
>
>
> | /
> --------------| |/ /--------
> component | / excess
> lead | \ lead
> --------------| |\ \--------
> | \
> <- no force cutter force applied ->
> blades
>
> --
> http://www.piclist.com hint: The PICList is archived three different
> ways. See http://www.piclist.com/#archives for details.
>
>
>
Of forgot to mention, too much heat when soldering can damage the crystal as well as moving around , as mentioned elsewhere, to make it LOOK PRETTY. Unless you are having a good percentage of boards with crystal failure, just fix it and move on.
> You can get shockless side cutters. These have blades that have a flat
> surface on the side that you put towards the component. They will produce a
> flat (or very nearly so) surface on the end of the component lead instead of
> the peaked end that you will see on the other part of the lead. Most cutters
> will produce peaked ends on both parts of the lead.
You can get scissor cutters and flush cutters.
I prefer the flush cutters, we have some small
and VERY sharp ones. The component legs fall
off, so little energy is imparted. It's a bit of
a pain sometimes as the legs often get stuck
in the cutters with a bit of flux etc, they
really don't shoot off like blunt old side
cutters. :o)
-Roman
Roman,
Which Murata resonators do you use? How much do they cost?
TIA
Looking forward:
Alan Shinn
Experience the
beginnings of microscopy.
Make your own replica
of one of Antony van Leeuwenhoek's microscopes.
visit http://www.mindspring.com/~alshinn/
I always use hot melt glue to hold the can down,
THEN solder it, THEN cut the leads with nice sharp
cutters. Never had a failed crystal. Recently i've
switched to Murata ceramic resonators for anything
that doesn't need a perfect clock speed, these
resonators are very robust.
>-Roman
--
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
Alan Shinn wrote:
>
> Roman,
> Which Murata resonators do you use? How much do they cost?
> TIA
> Looking forward:
> Alan Shinn
We have used a lot of the Murata CSA parts,
2-leg, in 10MHz and 16MHz. Never had one
fail, and they cop a hiding in automotive
devices. Cost is relative, Farnell well
under $1 US for smaller quantities, but you
can order larger quantities direct from
Asia once you need 500+.
-Roman
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
Absolutely, that would surely be a very shock free way of doing things. But
if the oscillators are THAT weak, wouldn't they be unusable in most
situations. I'm going to try a new crystal and hope that it was a one-off
and that the module won't fail in it's eventual environment....
Sorted it out! After replacing the oscillator, the problem remained, so I
started looking elsewhere. In the end I found it. It retrospect the mistake
was obvious. The design prototype was built on stripboard but lack of space
up top required a wire link, which for the sake of simplicity I had created
with some under board wiring. The only problem was one of these wires had
pocket through the PCB and touched the wire connecting the PIC to the
crystal. The heat of the soldering had melted the insulation, shorting the
wire (which was +ve) to the oscillator. If I pushed of pulled the crystal
this seperated the two wires by just enough to break the short and allow the
crystal to oscillate.
So infact, this had nothing to do with oscillators and more to do with
avoiding under-board wiring.
One of those mistakes where looking back it seems so obvious (like when I
spent two hours debugging code only to find I'd put movlw 12 instead of
d'12')....
Ben
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
>Sorted it out! After replacing the oscillator, the problem remained, so I
>started looking elsewhere. In the end I found it.
...
>So infact, this had nothing to do with oscillators and more to do with
>avoiding under-board wiring.
Yes, but look at the wonderful thread about 'how to cut wires'
this adventure spawned :-)
IMETRIX.COM>
