Searching \ for 'help me please!! ....' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=help+please
Search entire site for: 'help me please!! ....'.

Truncated match.
PICList Thread
'help me please!! ....'
2000\02\22@110630 by me

flavicon
face
Hello all,
I got some sleep!
Ok.
So I am now gonna put a 1000uF cap accross the power supply for the pic.

Now for the MCLR what should the exact condition be if I want to be able to
put a switch there too...?

Right now it is going STRAIGHT to high so this may indeed be the problem as
some have noted.

Thanks. Let's hope I get somewhere today!!


+--------------------------+
|  Sep Seyedi              |
|  -> spam_OUTmeTakeThisOuTspamsep.2y.net        |
|  -> http://sep.2y.net    |
+--------------------------+

2000\02\22@112451 by Chris Eddy

flavicon
face
Sep;

I gots some advice to toss on your pile.

Your Hbridge has a power and return wire coming from the power supply (2V and
ground).  The logic portions also have a 2V and ground.  Bring these two pairs
back to your source of power seperately.  Place the filter cap at the source of
power, where the wires come together.  Add a smaller filter, say 10uF and 0.1uF
at the pic itself.  Don't forget the 0.1, as the electrolytic is nt nearly as
useful for high frequencies.  Pull the MCLR up with  a 5 to 10K resistor.

When you lay out a board, try to implement this in the final PCB pattern as
well.  A nice solid ground trace is nice, but make sure that motor current does
NOT pass through that ground trace past the PIC itself.  They should arrive
seperately at the origin.

Also, try hard not to implement loops in the PCB pattern, such as a ground
trace all the way around the board.  This makes the ground trace susceptable to
magnetic fields in the near field (motor) and to a lesser degree,
electromagnetic (magnetic protion) in the far field (RF and florescent lights).
(A pure electric field or pure magnetic field transforms into an
electromagnetic field at some defined distance from the source of the field, in
a ratio that is related to the impedance of a free wave in space).

Chris Eddy
Pioneer Microsystems, Inc.

sep wrote:

{Quote hidden}

2000\02\22@113115 by James Paul

flavicon
face
Regarding MCLR...If it were me, I'd put a pullup resistor of
say 4.7K ohms from Vdd to MCLR, and a .1 uF cap from MCLR to
Ground.  For a manual RESET switch, put a N.O. PB across the
.1 uF cap.  So, when you first turn it on, the cap is
discharged and represents a short which pulls reset low, and
holds it there until the cap starts charging.  By the time the
cap has charged, the part has stabilized and is out of reset.
Likewise, when it is powered up and running, and you decide a
reset is in order, you press the pushbutton which discharges
the cap, and from here on out, it works the same as described
above.  Note that you can change the values of the cap and
resistor around to get the response you want.  The values
I gave have worked for me in the past, but may or may not be
optimum for your application.   Hope this helps out.

                                      Regards,

                                       Jim




On Tue, 22 February 2000, sep wrote:

{Quote hidden}

.....jimKILLspamspam.....jpes.com

2000\02\22@120631 by Dan Michaels

flavicon
face
At 08:30 AM 2/22/00 -0800, you wrote:
> Regarding MCLR...If it were me, I'd put a pullup resistor of
> say 4.7K ohms from Vdd to MCLR, and a .1 uF cap from MCLR to
> Ground.  For a manual RESET switch, put a N.O. PB across the
> .1 uF cap.  So, when you first turn it on, the cap is
> discharged and represents a short which pulls reset low, and
> holds it there until the cap starts charging.
.......
>                                       Regards,
>                                        Jim

Jim,

I'd add a mod here. First, (.1uf)*(4700) only gives a .00047 sec
RC time constant, which may not hold the MCLR line low long enough
for reset. You need a much bigger cap, say 22 uF, which gives .1 sec
time constant. Don't increase R by too much, as leakage can be a
factor.

Secondly, I've found, much to my chagrin (meaning $$$ wasted),
that you don't want to tie a capacitor straight onto MCLR on a PIC,
as this can damage the chip and lead to very large leakage
currents into the MCLR line (order of 100s of uA). Use at least
100 ohms between the cap and MCLR.

More info about this particular problem is in the Piclinks
page *PIC ERRATA* section on my website -

http://www.sni.net/~oricom/piclinks.htm

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom

2000\02\22@122052 by paulb

flavicon
face
Dan Michaels wrote:

> Secondly, I've found, much to my chagrin (meaning $$$ wasted),
> that you don't want to tie a capacitor straight onto MCLR on a PIC,
> as this can damage the chip and lead to very large leakage
> currents into the MCLR line (order of 100s of uA).

 What I think you mean is that a large cap on MCLR pulls it well above
Vcc when Vcc is shut off and that may cause strange behaviour.
--
 Cheers,
       Paul B.

2000\02\22@131106 by James Paul

flavicon
face
Dan,

Thanks for the tips.   I didn't think of the leakage current
before, but now that you bring it up, I can see that that could
cause some problems.  Also, regarding the cap/resistor
combination that I chose..It may not have actually been doing
much of anything as I have been using the Powerup Timer when
possible, so those MCLR components might have been just wasted
money.   Anyway, thanks again.

                                           Regards,

                                             Jim




On Tue, 22 February 2000, Dan Michaels wrote:

{Quote hidden}

EraseMEjimspam_OUTspamTakeThisOuTjpes.com

2000\02\22@144719 by Dan Michaels

flavicon
face
At 10:08 AM 2/22/00 -0800, you wrote:
> Dan,
>
> Thanks for the tips.   I didn't think of the leakage current
> before, but now that you bring it up, I can see that that could
> cause some problems.  Also, regarding the cap/resistor
> combination that I chose..It may not have actually been doing
> much of anything as I have been using the Powerup Timer when
> possible, so those MCLR components might have been just wasted
> money.   Anyway, thanks again.
>
>                                            Regards,
>                                              Jim

Yep, use of external RC is probably like a vestigal organ on
PICs. I no longer use RC, rather 2K pullup with 100 ohms to
gnd thru reset switch. No caps. They always start (with PWRup
Tmr enabled).

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom

2000\02\22@144727 by Dan Michaels

flavicon
face
At 04:19 AM 2/23/00 +1100, you wrote:
>Dan Michaels wrote:
>
>> Secondly, I've found, much to my chagrin (meaning $$$ wasted),
>> that you don't want to tie a capacitor straight onto MCLR on a PIC,
>> as this can damage the chip and lead to very large leakage
>> currents into the MCLR line (order of 100s of uA).
>
>  What I think you mean is that a large cap on MCLR pulls it well above
>Vcc when Vcc is shut off and that may cause strange behaviour.
>--
>  Cheers,
>        Paul B.
>

Paul,

Right. My take on it is, when you shutdown the system, Vcc goes
down fast and the cap discharges into MCLR, damaging internal
circuitry, probably the programming (rather than reset) stuff.

The chips are clearly damaged, as evidenced by large *quiescent*
leakage currents into MCLR, subsequently (cap or no cap). How/why
this causes flaky/hit-or-miss program execution, I have no idea.

Because of this susceptibility, I now use lower pullup R's (1-5K,
formerly used 20-50K), and NEVER tie any caps on MCLR.

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
==================

2000\02\22@152356 by Quitt, Walter

flavicon
face
Yep that 70ms typical (well usually like 90ms)
start up seems to more than enough for most of us.
Sometimes external circuitry doesn't like some
pins in certain states at start up, but that
usually can be designed around.  Just gotta
watch what kinda I/O you're connected to.  Those
few pesky open drain ones come to mind. he he he...

-Walt

{Original Message removed}

2000\02\22@175302 by Phillip Vogel

flavicon
face
> On Tue, 22 February 2000, Dan Michaels wrote:
>

> Secondly, I've found, much to my chagrin (meaning $$$ wasted),
> that you don't want to tie a capacitor straight onto MCLR on a PIC,
> as this can damage the chip and lead to very large leakage
> currents into the MCLR line (order of 100s of uA). Use at least
> 100 ohms between the cap and MCLR.

Actually, what you want to do here is put a diode (1N4148, 1N914, etc.) across
the pullup resistor (cathode to Vdd). When Vdd falls below Vmclr, the cap will
discharge through the diode, instead of into the mclr pin. An added advantage
of this configuration is that the reset circuit will have better transient
response.

Phillip
--
Phillip M. Vogel, President   | "It's not what you've been taught,
Bartal Design Group, Inc.     |  it's what you've learned." (me)
318 Marlboro Road             | +1-201-567-1343 FAX:+1-201-568-2891
Englewood, NJ 07631  USA      | phillipspamspam_OUTbartal.com

2000\02\22@194455 by Dan Michaels

flavicon
face
At 05:52 PM 2/22/00 -0500, you wrote:
>> On Tue, 22 February 2000, Dan Michaels wrote:
>>
>
>> Secondly, I've found, much to my chagrin (meaning $$$ wasted),
>> that you don't want to tie a capacitor straight onto MCLR on a PIC,
>> as this can damage the chip and lead to very large leakage
>> currents into the MCLR line (order of 100s of uA). Use at least
>> 100 ohms between the cap and MCLR.
>
>Actually, what you want to do here is put a diode (1N4148, 1N914, etc.) across
>the pullup resistor (cathode to Vdd). When Vdd falls below Vmclr, the cap will
>discharge through the diode, instead of into the mclr pin. An added advantage
>of this configuration is that the reset circuit will have better transient
>response.
>
>Phillip
>

Phil,

As noted in another response, the Microchip databook also shows an
R1 between the cap and MCLR. It's there for a reason, I think.

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
==================

2000\02\22@194506 by Dan Michaels

flavicon
face
At 12:20 PM 2/22/00 -0800, you wrote:
>Yep that 70ms typical (well usually like 90ms)
>start up seems to more than enough for most of us.
>Sometimes external circuitry doesn't like some
>pins in certain states at start up, but that
>usually can be designed around.  Just gotta
>watch what kinda I/O you're connected to.  Those
>few pesky open drain ones come to mind. he he he...
>
>-Walt
>

Yeah, and how many times have those durn external signals
kept the chip turned on (after Vcc is down) via leakage
thru the clamping diodes on the I/O pins? 1, 2, 3, ....,
(10)powerN.

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
==================

2000\02\23@020758 by Dr. Imre Bartfai

flavicon
face
Hi,
for being correct, put a Schmitt-trigger or at least a CMOS buffer between
MCLR and that RC chain. PIC hates slowly increasing level on the MCLR pin!

Regards,
Imre


On Tue, 22 Feb 2000, Dan Michaels wrote:

{Quote hidden}

2000\02\23@111908 by Dan Michaels

flavicon
face
At 08:07 AM 2/23/00 +0100, you wrote:
>Hi,
>for being correct, put a Schmitt-trigger or at least a CMOS buffer between
>MCLR and that RC chain. PIC hates slowly increasing level on the MCLR pin!
>
>Regards,
>Imre
>

Actually, the Mchp dataheet mentions using the R-C startup ckt
"... (for slow Vdd powerup)". Apparently, the internal powerup timer
might not fire in this case.

Has anyone actually *ever* seen a startup problem when either:
(a) tying MCLR straight to Vcc (Vdd), or
(b) using a pullup resistor only (no caps, diodes, etc)?

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom

2000\02\23@150745 by Dwayne Reid

flavicon
face
<x-flowed>>Has anyone actually *ever* seen a startup problem when either:
>(a) tying MCLR straight to Vcc (Vdd), or
>(b) using a pullup resistor only (no caps, diodes, etc)?
>
>- Dan Michaels

Absolutely!  It was a major problem for me with 16c71 PICs.  I came up with
ways to ensure that I met mchip's  specs for all subsequent designs and
have never been bothered since.  That means that EVERY design that I have
done since then has taken VDD rise time into account and ensured that
either MCLR is held LO while VDD is below the operating voltage (external
reset circuit) *or* that VDD rises very quickly (battery powered systems)
and that VDD *never* hovers around 0.6 - 0.7 Vdc (small filter caps with
discharge resistors).

I really don't know what the current crop of PICs is like without proper
reset control and I can't afford to take a chance that a design will fail
in the field because of it.  All I can say is that I got bit when I didn't
deal with it properly.

The reset circuit I use in my oven controllers uses a 10K pullup on MCLR
with a small (1n0) bypass cap right at the pin.  MCLR ties directly to the
reset controller and also feeds my interlock timer via a 1n4148 diode.  The
timer is reset anytime MCLR goes LO and disables dangerous outputs.  The
bypass capacitor is needed as per mchip's current errata sheets.

I posted a longer response with more detail to the same question on 14 Feb
2000 - I can repost if it got missed.

I agree with one of Dan's comments - a large cap tied directly to MCLR
might cause damage if VCC is ever lowered suddenly (shorted to gnd).  This
is not a normal situation and I am not aware of any failures in my products
that can be attributed to this.

dwayne




Dwayne Reid   <@spam@dwaynerKILLspamspamplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax

Celebrating 16 years of Engineering Innovation (1984 - 2000)

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
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.

</x-flowed>

2000\02\23@173012 by Dan Michaels

flavicon
face
Dwayne Reid wrote:

>>Has anyone actually *ever* seen a startup problem when either:
>>(a) tying MCLR straight to Vcc (Vdd), or
>>(b) using a pullup resistor only (no caps, diodes, etc)?
>>
>>- Dan Michaels
>
>Absolutely!  It was a major problem for me with 16c71 PICs.  I came up with
>ways to ensure that I met mchip's  specs for all subsequent designs and
>have never been bothered since.  That means that EVERY design that I have
>done since then has taken VDD rise time into account and ensured that
>either MCLR is held LO while VDD is below the operating voltage (external
>reset circuit) *or* that VDD rises very quickly (battery powered systems)
>and that VDD *never* hovers around 0.6 - 0.7 Vdc (small filter caps with
>discharge resistors).
>

Dwayne,

1. Any idea what the "critical" Vdd rise-time is - to preclude startup
problems?

2. Also, I wonder if your problems weren't largely a factor of how big
the initial load on Vdd was, with larger loads slowing risetime down,
or possibly with using large value caps (2000 uF) on the power supply?

[Maybe I've been lucky - only ever do low-power design, 7805 with 10-22
uF filter cap].

3. Also, I wonder if the newer 'A' series of chips aren't better in
this area.
=============

>
>I posted a longer response with more detail to the same question on 14 Feb
>2000 - I can repost if it got missed.
>

Looked at it, and you also talk about Vdd dipping into the
twilight zone causing hangups. On newer chips, Mchp now includes the
brownout reset ckt. Which leads to the next question,

4. Anybody have problems with that?
==============

>I agree with one of Dan's comments - a large cap tied directly to MCLR
>might cause damage if VCC is ever lowered suddenly (shorted to gnd).  This
>is not a normal situation and I am not aware of any failures in my products
>that can be attributed to this.
>
>dwayne
>

Personally, I don't think you need the "sudden" qualifier, but it's
a potential problem every time you power down the system.

You know, this whole thing is a problem because we are all used to
using other controllers that worked just fine with simple R-C reset
ckts, but now Mchp comes along and uses MCLR for both reset and
programming, removes the upper clamp diode, and some us of just
tie the ole cap to the ole pin, and never give it a second thought
- until one day ....
----------

After mulling thru comments from about 100 people, I wonder if the
best approach isn't simply to (unswaveringly) use the ckt Mchp shows
in their datasheet [son of a gun]. Use the reset cap (to hold MCLR
low while Vdd rises), but just make sure you've got R1 in there too
(to prevent the cap from zapping the chip on powerdown) - and
probably include the pullup diode too (to shorten retriggering time
and help pull the cap down).

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
too.

2000\02\24@022021 by Dwayne Reid

flavicon
face
<x-flowed>At 03:28 PM 2/23/00 -0700, Dan Michaels wrote:

>Dwayne,
>
>1. Any idea what the "critical" Vdd rise-time is - to preclude startup
>problems?

mchip data sheets specified that the total time from 0Vdc thru 3V had to be
less than 50 mSec - I was considerably faster than that at about 10 mSec
worst case, <1 mSec best case.  Still had lockup problems if VDD did not
dip below about 0.5 - 0.6 Vdc.

>2. Also, I wonder if your problems weren't largely a factor of how big
>the initial load on Vdd was, with larger loads slowing risetime down,
>or possibly with using large value caps (2000 uF) on the power supply?

That was one of the early things I looked at - I don't recall it being an
issue.


>[Maybe I've been lucky - only ever do low-power design, 7805 with 10-22
>uF filter cap].

That may have something to do with it - not the rise time, but whether VDD
hovers at that magical voltage of 0.7 - 1 Vdc.  In my case, when I lost AC
mains, VDD collapsed quickly to less than a couple of volts, then decayed
slowly down to about 0.5 Vdc over a period of hours or days.  Try to power
the pic back up during that time - I would get lockups where MCLR would NOT
revive the pic.  Discharge the filter cap so that VDD was below 0.3 Vdc or
so (I never did characterize the exact voltage) - power up reliably every time.


>3. Also, I wonder if the newer 'A' series of chips aren't better in
>this area.

Indeed - they may be better.  But I found low cost ways to deal with the
original problem at my end - I simply have included them in my designs.

>you also talk about Vdd dipping into the
>twilight zone causing hangups. On newer chips, Mchp now includes the
>brownout reset ckt. Which leads to the next question,
>
>4. Anybody have problems with that?

Nope - mchip engineers assure me that the brownout circuit will completely
eliminate the lockups that I experienced.  Couple of caveats - 12 bit core
pics don't have that brownout circuit so I still need to do that on AC
powered PIC designs.  Higher end controllers (14 bit core) want the extra
security of an external watchdog anyways (some of my ovens are 30 million
BTUs) and the PSU supervisor / reset control is part of that chip.


>I agree with one of Dan's comments - a large cap tied directly to MCLR
> >might cause damage if VCC is ever lowered suddenly (shorted to gnd).  This
> >is not a normal situation and I am not aware of any failures in my products
> >that can be attributed to this.
>
>Personally, I don't think you need the "sudden" qualifier, but it's
>a potential problem every time you power down the system.

I don't see how!  If VDD collapses slowly (ie - you remove AC), the cap
will begin to discharge either thru the substrate or the pullup resistor or
both.  If the power supply can supply all the current the circuit needs,
the cap on MCLR will supply only enough current to discharge - usually a
very low current.

dwayne



Dwayne Reid   <KILLspamdwaynerKILLspamspamplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax

Celebrating 16 years of Engineering Innovation (1984 - 2000)

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
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.

</x-flowed>

2000\02\24@042630 by Russell McMahon

picon face
-----Original Message-----
From: Dwayne Reid <RemoveMEdwaynerTakeThisOuTspamPLANET.EON.NET>
>That may have something to do with it - not the rise time, but whether VDD
>hovers at that magical voltage of 0.7 - 1 Vdc.  In my case, when I lost AC
>mains, VDD collapsed quickly to less than a couple of volts, then decayed
>slowly down to about 0.5 Vdc over a period of hours or days.  Try to power
>the pic back up during that time - I would get lockups where MCLR would NOT
>revive the pic.  Discharge the filter cap so that VDD was below 0.3 Vdc or
>so (I never did characterize the exact voltage) - power up reliably every
time.


re Vcc dropping to 0.5v ish and staying there (as on IC diodes stop
conducting) causing problems.for some ICs.



I had a battery powered product which wanted lowest possible power but an IC
on board sometimes locked up if repowered seconds to minutes after powering
off. This was a bug in the IC which the manufacturers knew about

It needed Vcc to drop to very near true gnd (say < 0.1v?) to reset properly.

A resistor across Vcc would have done but would add to current drain while
active so I incorporated a "Vcc discharger":

- An NPN  transistor from Vcc to gnd
- A 2nd npn transistor clamping base of transistor 1 to ground.
(C1 to Vcc, C2 to B1, E1 & E2 to gnd)

- Pullup resistor from base of Q1 to a capacitor (100uF) powered by diode
from Vcc.

- Pullup resistor R1 from base of 2nd resistor to Vcc plus a pulldown
resistor R2 from base of 2nd transistor to gnd.

As long as Vcc is high enough that R2/R1  turn on Q2 then nothing much
happens.
R1/R2 can be large.
When Vcc drops to eg 1 volt (or whatever you like) Q2 turns off and Q1 turns
on from 100 uF capacitor supply.
This crowbars Vcc to a true level of << 0.1 volt.
Resistors can be designed for very low quiescent current.
Q1 should be high beta.

Worked well.


     Russell McMahon
_____________________________

>From other worlds - http://www.easttimor.com
                               http://www.sudan.com

What can one man* do?
Help the hungry at no cost to yourself!
at  http://www.thehungersite.com/

(* - or woman, child or internet enabled intelligent entity :-))

2000\02\24@130819 by Dan Michaels

flavicon
face
I love those crowbar circuits, and think they are grossly
under utilized. Much more reliable than using WDT only,
especially in environments with a lot of external noise (such
as from switching inductive loads), which can cause the cpu to
latch up. We added one - to great advantage - to an industrial
controller that was suffering lots of latchups.

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
==========================
==========================

At 09:07 PM 02/24/2000 +1300, you wrote:
>-----Original Message-----
[snip]

>This crowbars Vcc to a true level of << 0.1 volt.
>Resistors can be designed for very low quiescent current.
>Q1 should be high beta.
>
>Worked well.
>      Russell McMahon

2000\02\24@130822 by Dan Michaels

flavicon
face
Dwayne,

Good info.

Also, you certainly have an interesting case here. When I used
the word "sudden", I didn't realize that in your particular systems
it may take *days* for Vdd decay. Oy vay.

Again, I was thinking about lower power systems (probably much
more common), where there isn't all that much capacitance on the
Vdd bus, and MCLR stays up enough longer that the cap discharges
into MLCR.

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
===========================
===========================

At 11:53 PM 02/23/2000 -0700, you wrote:
>At 03:28 PM 2/23/00 -0700, Dan Michaels wrote:
>
>>Dwayne,

[snip]
{Quote hidden}

2000\02\24@133351 by Erik Reikes

flavicon
face
Now, I'm no electrical engineer, but it seems to me that you have two paths
for the discharge of this capacitor, one is through the pullup to VCC, and
the other is through a protection diode to VCC (MCLR line).  As VCC is
dropping won't the resistor start discharging first because of the step
function nature of the forward bias voltage on the diode (i.e. until there
is at least a .6V drop on the resistor ...60uA current... the diode does
nothing)?

So, in other words, as long as VCC is decaying at a rate that is equal to
or slower than the time constant of your MCLR RC combination you should be
ok...  Even a little faster might be ok because then you have two parallel
paths (as well as the .6v 'hysteresis' forward voltage of the MCLR diode).
For me this time constant is .1uF * 10k Ohm = 1ms.  I've got decouplers on
my VCC line on the order of 22uF +.  If for example, my micro typically
burns 1mA or less that gives an equivalent load at 3.6V of 3.6KOhm giving a
time constant for VCC decay of : 3.6k * 22uF = 79mS.  Seems like it should
never be a problem unless some yahoo figures out how to yank out a battery
and short VCC to GND in less than a couple of mS.  I'm not saying its
impossible, I'd just like to see it!

Well, its all academic as next spin of board gets a schottky diode to VCC
anyhow.

Let me know if there is something wrong with my logic...  In practice, my
EE skills stop around the V=IR stage.

-E

>At 11:53 PM 02/23/2000 -0700, you wrote:
>>At 03:28 PM 2/23/00 -0700, Dan Michaels wrote:
>>
>>>Dwayne,
>
>[snip]
>>
>> >I agree with one of Dan's comments - a large cap tied directly to MCLR
>>> >might cause damage if VCC is ever lowered suddenly (shorted to gnd).  This
>>> >is not a normal situation and I am not aware of any failures in my
products
{Quote hidden}

Erik Reikes
Software Engineer
Xsilogy, Inc.

spamBeGoneereikesspamBeGonespamxsilogy.com
ph : (858) 535-5113
fax : (858) 535-5163
cell : (858) 663-1206

2000\02\24@141335 by Dan Michaels

flavicon
face
Erik,

First, there is probably no (pullup) protection diode inside
the MCLR pin, this line has to be pulled up to Vpp (13.5v or so)
during programming.

And I think your logic is correct. But it's all *relative*, Vdd
rise/falltimes vs MCLR rise/falltimes.

If you want the RC to reliably reset the cpu, then its risetime
has to be significantly longer than that at Vdd, so there's a
good chance its falltime will also be longer - especially
since Vdd is driving all the loads on the bus.

[Note some systems *will* have slow Vdd decay, due to using
large filter caps, etc. Every case will be a little different,
and these differences will change, due to changing loads, etc.]

Now, in those cases where Vdd *DOES* decay significantly faster
than MCLR, *AND* you don't use an external pullup diode, then the
cap will discharge into the MCLR pin (possibly damaging the part).
[this is the case where the reset circuitry is a simple RC only
- just like we've been using for years on other processors].

Adding the pullup diode will help pull MCLR down faster, but
the way to *GUARANTEE* limited current into MCLR is to use
R1 between the cap and the pin - just as Mchp recommends on
their datasheets. [everything else is variable, but this is
an absolute].

- Dan Michaels
Oricom Technologies
http://www.sni.net/~oricom
==========================
==========================

At 10:37 AM 02/24/2000 -0800, you wrote:
{Quote hidden}

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...