Post by thread:[PIC] PIC18F452 ICSP Vdd Connection

'[PIC] PIC18F452 ICSP Vdd Connection'

Hi,

 I was reading the Microchip article on ICSP and implementing it in a circuit I'm working on, but I'm having doubts about the Vdd connection. Microchip says to put a resistor between +5V & the µC's Vdd pins, but I'm thinking a diode (preferable Schotky type) would be a lot better. The diode would make it so that the programmer doesn't see the 100 or so µF of smoothing caps, but that would mean that the PIC is fed by let's say 4.6V and all the other chips on the board by 5V. Would that be any problem?

 Regards, Anthony

Website: http://members.lycos.nl/anthonyvh

---------------------------------

Do you Yahoo!?

Shop for Back-to-School deals on Yahoo! Shopping.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Anthony Van Herrewege wrote:

> that would mean that the PIC is fed by let's say 4.6V and all the

> other chips on the board by 5V. Would that be any problem?

Hi Anthony,

I was a bit confused by your reference to "5C" in your original post. What

part is that? Also, when you mentioned "5F" did you mean 5 farads of

capacitance? Just to make sure I know what you are reading, what specific

"Microchip article on ICSP" are you referring to?

Going back to your question:

----------------------------

Please observe the PIC's electrical specifications in the data sheet. You

should see something along the lines of:

Voltage on any pin with respect to VSS

(except VDD, MCLR, and RA4):

 -0.3V to (VDD + 0.3V)

If any of the logic outputs powered by the 5V supply is feeding any of your

PIC's I/O lines (inputs), and your PIC is supplied 4.6 V (with respect to

Vss) then you would be violating the above specification (but I think you

just gave hypothetical voltages correct?).

It sounds like you are thinking about a schottky diode correct?

International Rectifier makes some pretty low Vf (forward voltage) schottky

diodes. For ICSP isolation, I've used this SMT schottky diode:

International Rectifier part# 10BQ015

(Digi-Key part#: 10BQ015-ND) (25 cents in single quantities)

Forward drop is only 0.3 volts at 0.5 amps at 25 degrees C -- expect a lower

forward drop with lower forward currents. You can interpolate this

information from the data sheet graphs. Up to a point, you'll see a lower Vf

at higher temperatures as well. Your PIC's Vdd pin should not be drawing

more than around 0.2 to 0.3 amps *maximum* depending on your PIC and your

application circuit, so your schottky's Vf should be much lower than 0.3

volts. Using this diode should allay your concerns.

Best regards,

Ken Pergola

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Hi Ken,

I checked my message and I can't find any "5C" or "5F", must be some error with Yahoo or something. The article is "In-Circuit Serial Programming Guide" (DS30277D).

My values aren't hypothetical, only +5V will be available in the circuit, so this was my only concern. (Well, actually there's a +12V as well, but it doesn't get anywhere near the PIC, so that doesn't matter.)

Thanks for the part number! I was thinking of using OnSemi's MBR130, but it has a typical Vf of 0.47V (at 1A), so the IR one's better :).

I think my circuit's almost complete now :).

Thanks, Anthony

Ken Pergola <spam_OUTno_spamTakeThisOuTlocalnet.com> wrote:

Hi Anthony,

I was a bit confused by your reference to "5C" in your original post. What

part is that? Also, when you mentioned "5F" did you mean 5 farads of

capacitance? Just to make sure I know what you are reading, what specific

"Microchip article on ICSP" are you referring to?

...

If any of the logic outputs powered by the 5V supply is feeding any of your

PIC's I/O lines (inputs), and your PIC is supplied 4.6 V (with respect to

Vss) then you would be violating the above specification (but I think you

just gave hypothetical voltages correct?).

It sounds like you are thinking about a schottky diode correct?

International Rectifier makes some pretty low Vf (forward voltage) schottky

diodes. For ICSP isolation, I've used this SMT schottky diode:

International Rectifier part# 10BQ015

(Digi-Key part#: 10BQ015-ND) (25 cents in single quantities)

...

Best regards,

Ken Pergola

Website: http://members.lycos.nl/anthonyvh

---------------------------------

Do you Yahoo!?

Yahoo! Mail - 50x more storage than other providers!

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Anthony Van Herrewege wrote:

> ...I can't find any "5C" or "5F"...

Hi Anthony,

I'm sorry, I should have quoted more from your original message. I saw the

following:

You wrote:

"Microchip says to put a resistor between +5V & the 5C's Vdd pins, but I'm

thinking a diode (preferable Schottky type) would be a lot better. The diode

would make it so that the programmer doesn't see the 100 or so 5F of

smoothing caps, but that would mean that the PIC is fed by let's say 4.6V

and all the other chips on the board by 5V."

I was just confused by the "5C's Vdd pins" and the "5F of smoothing caps" in

the above.

Just be aware that the 10BQ015 reverse leaking current is much higher than

say a 1N5818 schottky, but the 10BQ015's Vf is much lower. Here's some

*anecdotal benchtop* results at room temperature, sample size = 1. Take

these as a guideline only.

10BQ015 Vf at 46 mA: 0.194 volts (with respect to ground)

1N5818 Vf at 46 mA: 0.289 volts (with respect to ground)

10BQ015 reverse leakage current at Vr = 5V: 3 uA

1N5818 reverse leakage current at Vr = 5V: 65 uA

Don't read too much into this except at the data points I measured. your

mileage may vary.

Best regards,

Ken Pergola

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

At 05:03 PM 9/11/2004, you wrote:

>Hi Ken,

>

>I checked my message and I can't find any "5C" or "5F", must be some error 

>with Yahoo or something.

Anthony managed to send his email with a "mu" character - prefix for micro. 

It showed up that way on my system but apparently on Kens system, it came 

through as a "5".  So uC became 5C and uF became 5F. Notice that I just 

used lower case "u" instead of "mu".

Andy Warren mentioned this just the other day.  Andy's advice was to not 

use special symbols like mu and ohmega, etc. even though they are 

correct.  The reason being that they *often* get misinterpreted and your 

message comes through as something you did not intend.

Of course, as Andy also says, your mileage may vary.

MD

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Marcel Duchamp wrote:

> Anthony managed to send his email with

> a "mu" character - prefix for micro. 

Hi Marcel,

Thanks for the clarification. No wonder I was confused. :)

Best regards,

Ken Pergola

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Hi,

I think I'm going to use OnSemi's MRB120VLSF (https://www.onsemi.com/pub/Collateral/MBR120VLSFT1-D.PDF), Vf Max is 0.34V, but probably a lot lower in my circuit. It's just that I can order a sample of those, to get the International Rectifier one, I'd need to order from RS or something and pay shipping costs that are like 10x higher than the price of the diode...

I won't use the mu symbol anymore ;).

Thanks, Anthony

Website: http://members.lycos.nl/anthonyvh

---------------------------------

Do you Yahoo!?

Yahoo! Mail - You care about security. So do we.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Anthony Van Herrewege wrote:

> I was reading the Microchip article on ICSP and implementing it in a

> circuit I'm working on, but I'm having doubts about the Vdd

> connection. Microchip says to put a resistor between +5V & the µC's

> Vdd pins, but I'm thinking a diode (preferable Schotky type) would be

> a lot better. The diode would make it so that the programmer doesn't

> see the 100 or so µF of smoothing caps, but that would mean that the

> PIC is fed by let's say 4.6V and all the other chips on the board by

> 5V. Would that be any problem?

Most PICs run at full speed down to at least 4.5V.  The extra 400mV

difference between the PIC power and that of the rest of the circuit is

probably not important, but it's your circuit and only you can answer that.

It's likely a bad idea if you are using the PIC A/D with Vdd as reference.

A better answer, at least in a production environment, is to use a

programmer specifically intended to tolerate reasonable capacitance on Vdd.

Is this a commercial product?  If so, you might want to seriously consider

my ProProg programmer (http://www.embedinc.com/proprog).  It can deliver up

to 500mA Vdd, and clamps Vdd to ground thru a 10ohm resistor when it wants

it to be low.  It also has analog readback of Vdd, so it can wait until the

level rises of falls sufficiently before proceeding.  It should be able to

tolerate 100s of uF on Vdd, but I haven't figured out the exact specs yet.

Part of it is how long you are willing to wait.  The 100uF spec on the web

page is just a placeholder for now.

For hobbyists, this can be solved by the EasyProg programmer

(http://www.embedinc.com/easyprog).  It has active high drive and passive

low drive on the target chip Vdd.  Since its open loop, the limiting factor

is how long the wait is after every high and low going Vdd change.  The

standard firmware waits 25mS, but there is also a version that waits 50mS on

Vdd rise and 300mS on Vdd fall.  I've got a customer using that on a 16F630

target circuit with 220uF on its Vdd line.

*****************************************************************

Embed Inc, embedded system specialists in Littleton Massachusetts

(978) 742-9014, http://www.embedinc.com

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Olin, have you thought about a gang programmer at all? My guess is

your response will be something along the lines of your primary

interest being design and consulting, not production, which is fair

enough. Right now we program using a single programmer, and it eats up

time which drives me nuts. We found a low cost 4 gang ISP for our PSoC

products, and it is amazing. It runs stand alone (once you load the

hex file into it), and you just connect everything up, press the

button, and 8 LEDs give you a status update. I wish we could find

something like that for the PIC.

Josh

-- 

A common mistake that people make when trying to design something

completely foolproof is to underestimate the ingenuity of complete

fools.

        -Douglas Adams

On Sun, 12 Sep 2004 10:26:31 -0400, Olin Lathrop

<.....olin_piclistKILLspam@spam@embedinc.com> wrote:

> A better answer, at least in a production environment, is to use a

> programmer specifically intended to tolerate reasonable capacitance on Vdd.

> Is this a commercial product?  If so, you might want to seriously consider

> my ProProg programmer (http://www.embedinc.com/proprog).  It can deliver up

> to 500mA Vdd, and clamps Vdd to ground thru a 10ohm resistor when it wants

> it to be low.  It also has analog readback of Vdd, so it can wait until the

> level rises of falls sufficiently before proceeding.  It should be able to

> tolerate 100s of uF on Vdd, but I haven't figured out the exact specs yet.

> Part of it is how long you are willing to wait.  The 100uF spec on the web

> page is just a placeholder for now.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Josh Koffman wrote :

> Right now we program using a single programmer, and it eats up

> time which drives me nuts. We found a low cost 4 gang ISP for our PSoC

> products, and it is amazing. It runs stand alone (once you load the

> hex file into it), and you just connect everything up, press the

> button, and 8 LEDs give you a status update. I wish we could find

> something like that for the PIC.

Some time ago (a couple of months I think) I seems to remember that

Wouter made a comment about one of his Wisp628 customers that was

using a bunch of Wisp628's for programming. I don't remember the

exact amount, could have been anything between 5 and 20.

I'm sure Wouter will "fill in the blanks" here :-) :-)

Regards,

Jan-Erik.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Josh Koffman wrote:

> Olin, have you thought about a gang programmer at all? My guess is

> your response will be something along the lines of your primary

> interest being design and consulting, not production, which is fair

> enough. Right now we program using a single programmer, and it eats up

> time which drives me nuts. We found a low cost 4 gang ISP for our PSoC

> products, and it is amazing. It runs stand alone (once you load the

> hex file into it), and you just connect everything up, press the

> button, and 8 LEDs give you a status update. I wish we could find

> something like that for the PIC.

My answer to a gang programmer is multiple individual programmers.  I put an

uncommitted user button and LED on the ProProg so that it can appear to run

unattended although the top level logic is being performed on a host

computer.  Once programming is complete, you swap in the next device or

target chip then hit the button.  The host software would then automatically

repeat the programming procedure.  You would need a separate serial port

connection bewteen the host and each programmer.  That's not too hard to do

nowadays via USB serial port adapters.  The ProProg and EasyProg only use

RX, TX, and ground, so even the dumb ones should work although I haven't

tried it.

*****************************************************************

Embed Inc, embedded system specialists in Littleton Massachusetts

(978) 742-9014, http://www.embedinc.com

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Hi Olin,

> The extra 400mV difference between the PIC power and

> that of the rest of the circuit is probably not 

> important, but it's your circuit and only you can 

> answer that. It's likely a bad idea if you are using

> the PIC A/D with Vdd as reference.

I plan on using 1 pin as an A/D and I have no external

reference. However, the absolute value is not really

important. I'm using it to connect +- 8 pushbuttons to

1 pin, so I guess it should be OK.

> Is this a commercial product?

No, it's a controller for a 7.1 channel preamp I'm

trying to make. You can see some details about it here

on my site:

http://members.lycos.nl/anthonyvh/index.php?page=preamp-main

> For hobbyists, this can be solved by the EasyProg

programmer.

I'm using the PPWin programmer

(http://www.semis.demon.co.uk/Pics/PICmain.htm), but

this is the first project that I'm using a PIC for

(amongst heaps of other things I use for the first

time, I'm crossing my finger that it will work), so I

don't know if it could handle the capacitance on the

board.

Anthony

* I got one very stupid question for which I certainly

don't want to send a separate mail to the list. It'd

be cool if someone helped me out.

Ok, here goes (attention, stupid question ahead): to

protect the controller board from damage, I'd like to

include a zener in the circuit (which probably won't

be necessary at all, but I guess it's better to be

safe than sorry). Can I just connect it between the

+5V & gnd or should it be like: +5V -> low value

resistor -> circuit & zener? I don't quite know how

much current the circuit will need, although it

shouldn't be more than 1A (probably even less than

0.5A, but the 1A figure is just to be safe).

=====

Website: http://members.lycos.nl/anthonyvh

__________________________________

Do you Yahoo!?

Yahoo! Mail Address AutoComplete - You start. We finish.

http://promotions.yahoo.com/new_mail 

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Josh Koffman writes:

> Olin, have you thought about a gang programmer at all? My guess is

> your response will be something along the lines of your primary

> interest being design and consulting, not production, which is fair

> enough. Right now we program using a single programmer, and it eats up

> time which drives me nuts. We found a low cost 4 gang ISP for our PSoC

> products, and it is amazing. It runs stand alone (once you load the

> hex file into it), and you just connect everything up, press the

> button, and 8 LEDs give you a status update. I wish we could find

> something like that for the PIC.

I'm just a newbie, but if there is one thing I know about right now it

is PIC programmers!  It seems like it would be nearly trivial to wire

up two Wisp628s (or something similar) and just run two instances of

the programming software on separate com ports.  The advantage of this

is that you can be programming one while removing, filing, and

inserting the other.  True you wouldn't get the multi-voltage

verification, but after looking at other circuits the multi-voltage

stuff doesn't really seem all that complicated.  The real issue here

would be at the PC end, as you'd need to modify the programming code

to support multi-voltage control, and the source may not be available.

I guess it really depends on what your time is worth and whether it's

just cheaper to go out any buy something ready-made than develop

yourself...

-p.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

> We found a low cost 4 gang ISP for our PSoC

> products, and it is amazing. It runs stand alone (once you load the

> hex file into it), and you just connect everything up, press the

> button, and 8 LEDs give you a status update. I wish we could find

> something like that for the PIC.

Can't you use multiple programmers with one PC? I recall a client of me

used (IIRC) 36 Wisp628's this way.

Wouter van Ooijen

-- -------------------------------------------

Van Ooijen Technische Informatica: http://www.voti.nl

consultancy, development, PICmicro products

docent Hogeschool van Utrecht: http://www.voti.nl/hvu

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Peter Johansson wrote :

> I'm just a newbie, but if there is one thing I know about right now it

> is PIC programmers!  It seems like it would be nearly trivial to wire

> up two Wisp628s (or something similar) and just run two instances of

> the programming software on separate com ports.

Hi.

Wouter once wrote (http://www.piclist.com doesn't answer right know so I can't

check the archives) about a setup with 10 or 20 (or, if it was a binary

numer, maybe 8 or 16 :-) ) Wisp628's running at a customer of his.

As I remember it, they was running from a single PC...

Regards,

Jan-Erik.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Anthony Van Herrewege wrote :

> Ok, here goes (attention, stupid question ahead): to

> protect the controller board from damage,...

It might help knowing what kind of damage and from what

source it should be protected. What is the potential problem ?

Overvoltage ?

5v and gnd connected the wrong way ?

Jan-Erik.

> I'd like to

> include a zener in the circuit (which probably won't

> be necessary at all, but I guess it's better to be

> safe than sorry). Can I just connect it between the

> +5V & gnd or should it be like: +5V -> low value

> resistor -> circuit & zener? I don't quite know how

> much current the circuit will need, although it

> shouldn't be more than 1A (probably even less than

> 0.5A, but the 1A figure is just to be safe).

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Hi Jan-Erik, 

> Jan-Erik Soderholm wrote:

>

> It might help knowing what kind of damage and from

> what source it should be protected. What is the 

> potential problem ?

> 

> Overvoltage ?

> 5v and gnd connected the wrong way ?

> 

> Jan-Erik.

The potential problem is overvoltage, connecting the

wrong way around is not really possible, because the

connectors are polarized, so I can't connect the power

connector wrong. So, do I need a resistor between the

+5V & the circuit or not?

Anthony

=====

Website: http://members.lycos.nl/anthonyvh

_______________________________

Do you Yahoo!?

Shop for Back-to-School deals on Yahoo! Shopping.

http://shopping.yahoo.com/backtoschool

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Peter Johansson wrote:

> I'm just a newbie, but if there is one thing I know about right now it

> is PIC programmers!  It seems like it would be nearly trivial to wire

> up two Wisp628s (or something similar) and just run two instances of

> the programming software on separate com ports.  The advantage of this

> is that you can be programming one while removing, filing, and

> inserting the other.  True you wouldn't get the multi-voltage

> verification,

If by "something similar" to the Wisp628 you include the EasyProg, then you

do get multi-voltage verification.

> but after looking at other circuits the multi-voltage

> stuff doesn't really seem all that complicated.  The real issue here

> would be at the PC end, as you'd need to modify the programming code

> to support multi-voltage control, and the source may not be available.

The PC source for the EasyProg is available, but it already handles the

multi-voltage stuff anyway.

> I guess it really depends on what your time is worth and whether it's

> just cheaper to go out any buy something ready-made than develop

> yourself...

I don't recommend the EasyProg in a time=money environment since you have to

build it up yourself.

*****************************************************************

Embed Inc, embedded system specialists in Littleton Massachusetts

(978) 742-9014, http://www.embedinc.com

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist