Searching \ for 'Weldbond - Weird Problem with the 16C64A' 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/microchip/devices.htm?key=16C
Search entire site for: 'Weldbond - Weird Problem with the 16C64A'.

Truncated match.
PICList Thread
'Weldbond - Weird Problem with the 16C64A'
1998\08\30@220636 by myke predko

flavicon
face
Hi Folks,

I just found a weird one that you should be aware of:

I am building a circuit with a 16C64A and I am putting a 40 Pin DIP onto a
Wire Wrap Board.  I couldn't find a 40 pin W/W socket, so I made one by
cutting up two 16 pin and one 8 pin W/W DIP Sockets.

The board itself is just a phenolic board with no copper traces on either
side, so I couldn't solder in the sockets.  I did what I normally do in this
situation and that was to glue them down with "Weldbond".  The sockets were
glued in last Wednesday and the glue seemed to have cured (ie it's clear -
from a white liquid originally).

Putting in the 16C64A, I found it wouldn't run.  I spent most of this
weekend trying to get it to work until I finally put a scope on it.  I found
that the pins adjacent to Vdd (Pin 10, RE2 and Pin 33 RB0) were at about 3
Volts.  This wasn't high enough, or have enough current to set off my logic
probe or be detected by my DMM, but it seemed to be holding the chip reset.

I built a test circuit on a proto-board and found these pins were normally
at zero volts.

Of course, for the six spots where I put on the glue, two were straddling
these pins and Vdd.  I scraped off the glue and lo and behold, it started to
work (and the voltages around the pins was back to around zero volts).

When I checked the impedance between the pins before removing the glue, I
found that for Pin 33 to Pin 32 it was above 2 M (ie as high as my DMM goes)
and for Pin 10 and Pin 11, it was around 250K.

This seems to be a case of a high-impedance short holding the PICMicro
Reset.  Neither or these pins are related to programming (Reset, Pin 1 & RB7
and RB6, Pins 39 and 40, respectively, are).

With regard to "Weldbond", I've been using it for years for this type of
application.  When it's wet, it has a resistance of about 50K, which goes up
into the 100s of Kilo-ohms after a couple of days (just as I saw here).

In case you're wondering why I know about the electrical properties of
Weldbond, I was using it to glue down wires on a Polaroid sonar module
earlier this year and it stopped working for a few days (the module produces
400 volts DC during the ultrasonic "chirp") until the glue cured and the
resistance became high.

So, I guess I should be looking for some other glue for this type
application.  Does anybody have any suggestions?  I have tried hot-melt, but
that can get inside connectors and gum them up or the long strings cause
opens in the connectors that are just about impossible to find.

Any ideas why this happened with the PICMicro?  I would have thought that
the marginal voltages wouldn't have affected the start up - What if these
lines where driven high when the application started up, I wouldn't think
that would cause a problem.

Curious in Toronto,

myke

Hunter S. Thompson's quest for the American Dream, this week in the Book Room.

http://www.myke.com/Book_Room/book1a.htm

1998\08\30@222932 by Jason Tuendemann

flavicon
face
Many Glue Problems have been found in the broadcast industry. Sony used to
glue wireing mods down on there boards, 3-5 years later the glue broke down
under operation temperature and there were hundreds of unexplainable
problems. Remove the glue and wash boards and the problems dissappeared. The
other day my fathers Microwave went bizerk and I mean completely out of
control. Pulled out the controller board, removed the glue around the
resonator and the clock ran again and all was happy.
I have used hot glue most of the time with no trouble or a type of epoxy.

Jason.

1998\08\31@021342 by Mark Willis

flavicon
face
Hmm.  The only types of glue I've used lots are hot glue, and
superglue (to hold down wiring mods on re-worked boards.  Usually dot
the board with catalyst, dab a little Gel-type Superglue onto the wire,
& toothpick the wire into place, and stand back as it heats up & sets
quickly <G>)  But I haven't measured impedance of superglue (I'd suggest
the gel type and keep it away from the electrical connectors & use the
catalyst, it sets ON CONTACT, that way.  Sorta fizzes from heat as it
sets instantly <G>)  Hope that helps!

 Mark, spam_OUTmwillisTakeThisOuTspamnwlink.com

myke predko wrote:
{Quote hidden}

1998\08\31@033915 by Lee Jones

flavicon
face
> I just found a weird one that you should be aware of:
>
> I am building a circuit with a 16C64A and I am putting a 40 Pin DIP onto a
> Wire Wrap Board.  I couldn't find a 40 pin W/W socket, so I made one by
> cutting up two 16 pin and one 8 pin W/W DIP Sockets.

> glue [sockets] down with "Weldbond".

> Putting in the 16C64A, I found it wouldn't run.

> glue, two were straddling these pins and Vdd.  I scraped off
> the glue and lo and behold, it started to work

Very interesting tidbit that I'll have to remember.

On the subject of resistances of liquids... We were trying to
figure out how some soft drink dispensers stop filling the cups
when they reach the top.  Our guess was dispenser measured the
resistance from liquid stream to the metal (always) push bar.
So I measured flowing carbonated soft drinks.  Got 10K ohms for
diet and 40-50K for sugar ones.


> So, I guess I should be looking for some other glue for this type
> application.  Does anybody have any suggestions?

I've tried hot-melt too.  I switched to Devcon 5 minute epoxy.
Been a while since I've done any wire-wrap though I still use
the wire to jumper pins on prototype boards.  A couple wraps
holds it nicely while you're soldering.

                                               Lee Jones

-------------------------------------------------------------------
Jones Computer Communications             .....leeKILLspamspam@spam@frumble.claremont.edu
509 Black Hills Dr, Claremont, CA 91711         voice: 909-621-9008
-------------------------------------------------------------------

1998\08\31@044249 by Russell McMahon

picon face
Hot melt is a bad choice for most things, ultimately. Its quick and
easy to use and sets quite quickly but it doesn't last as a glue.
Many people use it in electronic assemblies and my universal
experience is that it deteriorates in a year or two. If the glue is
load bearing the glued object drops off and does whatever the glue
was initially meant to stop it doing. (Its excellent for use in
products with a one year warranty :-)).

A much nicer adhesive which lasts longer than I have been using it
(the manufacturer of some versions says it will last 20 years out in
the weather) is "Silicon Rubber" (RTV etc). This sets in hours rather
than minutes but doesn't let go. Messier, gap filling and slowish
setting. You MUST use "neutral cure" versions - MOST versions use
acetic acid cure which is potentially corrosive, bad on eyes etc and
unnecessary. I use a neutral cure "marine" version which is marvelous
for secondary anchoring of components which will be subject to
mechanical shock. The glue sets to a "rubber" with excellent adhesion
and a some flexibility which stops shock fracturing.

Very useful is "household epoxy" (as its called here in NZ) which is
epoxy resin with about  a 1 minute pot life. Sets to good tack
strength in 2 minutes (firm and unworkable - would easily hold an IC
socket) and is very very well set in 10 or 15 minutes. Its very
probably not as strong as regular epoxies but works superbly for
anything that needs strength and speed. Not quite as fast as
"superglue", gap fills (so no good on crockery) and is 2 part but
still useful. The resistance between probe tips placed as close as I
can get them without touching (0.5mm / 0.020"???) is over 30 megohms
(standard el cheapo electronic meter, low voltage) whether fully
cured (scrap from bin) or just mixed. I've never tried it in a power
on situation but it sounds like the glue you want.



{Original Message removed}

1998\08\31@063826 by Chip Weller

flavicon
face
myke predko wrote:


>Hi Folks,
>
>I just found a weird one that you should be aware of:
>
>I am building a circuit with a 16C64A and I am putting a 40 Pin DIP onto a
>Wire Wrap Board.  I couldn't find a 40 pin W/W socket, so I made one by
>cutting up two 16 pin and one 8 pin W/W DIP Sockets.
>
>The board itself is just a phenolic board with no copper traces on either
>side, so I couldn't solder in the sockets.  I did what I normally do in
this
>situation and that was to glue them down with "Weldbond".  The sockets were
>glued in last Wednesday and the glue seemed to have cured (ie it's clear -
>from a white liquid originally).
>
>Putting in the 16C64A, I found it wouldn't run.  I spent most of this
>weekend trying to get it to work until I finally put a scope on it.  I
found
>that the pins adjacent to Vdd (Pin 10, RE2 and Pin 33 RB0) were at about 3
>Volts.  This wasn't high enough, or have enough current to set off my logic
>probe or be detected by my DMM, but it seemed to be holding the chip reset.

> <snipped>

>When I checked the impedance between the pins before removing the glue, I
>found that for Pin 33 to Pin 32 it was above 2 M (ie as high as my DMM
goes)
>and for Pin 10 and Pin 11, it was around 250K.

> <snipped>

>Any ideas why this happened with the PICMicro?  I would have thought that
>the marginal voltages wouldn't have affected the start up - What if these
>lines where driven high when the application started up, I wouldn't think
>that would cause a problem.


This is an odd problem. Was the oscillator running? If so my first guess
would be you had enabled the INT pin (RB0) which ended up continously
interrupting. If you are curious and care to probe your working system using
high value resistors I would be interested to learn which pin was the cause
of the problem.

Chip Weller

1998\08\31@070114 by tjaart

flavicon
face
Russell McMahon wrote:

> Hot melt is a bad choice for most things, ultimately. Its quick and
> easy to use and sets quite quickly but it doesn't last as a glue.
> Many people use it in electronic assemblies and my universal
> experience is that it deteriorates in a year or two. If the glue is
> load bearing the glued object drops off and does whatever the glue
> was initially meant to stop it doing. (Its excellent for use in
> products with a one year warranty :-)).
>
> A much nicer adhesive which lasts longer than I have been using it
> (the manufacturer of some versions says it will last 20 years out in
> the weather) is "Silicon Rubber" (RTV etc). This sets in hours rather
> than minutes but doesn't let go. Messier, gap filling and slowish
> setting. You MUST use "neutral cure" versions - MOST versions use
> acetic acid cure which is potentially corrosive, bad on eyes etc and
> unnecessary. I use a neutral cure "marine" version which is marvelous
> for secondary anchoring of components which will be subject to
> mechanical shock. The glue sets to a "rubber" with excellent adhesion
> and a some flexibility which stops shock fracturing.

Silicon rubber attack any plastic close to it. It can also
cause headaches with 'nowash' solder fluxes after some
time. It's use is dissallowed in all automobile electronics
for this very reason. We even specify our injection
mould release agent to contain no silicon.

--
Friendly Regards

Tjaart van der Walt
tjaartspamKILLspamwasp.co.za

|--------------------------------------------------|
|                WASP International                |
|R&D Engineer : GSM peripheral services development|
|--------------------------------------------------|
|SMS .....tjaartKILLspamspam.....sms.wasp.co.za  (160 chars max)|
|     http://www.wasp.co.za/~tjaart/index.html     |
|Voice: +27-(0)11-622-8686  Fax: +27-(0)11-622-8973|
|          WGS-84 : 26¡10.52'S 28¡06.19'E          |
|--------------------------------------------------|

1998\08\31@110641 by Matt Bonner

flavicon
face
Tjaart van der Walt wrote:
>
> Silicon rubber attack any plastic close to it. It can also
> cause headaches with 'nowash' solder fluxes after some
> time. It's use is dissallowed in all automobile electronics
> for this very reason. We even specify our injection
> mould release agent to contain no silicon.
>
We use a "no headache inducing" silicon adhesive from Dow Corning (3145
RTV).  This expensive stuff is non-acidic and doesn't react with
anything.  Takes an hour or two to set up and is fully cured in 24
hours.  Great for WW sockets - glue them on then go for lunch.  When you
get back it has cured enough to start wire wrapping.  The real question,
though, is: who's got the time for a one hour lunch?

--Matt

1998\08\31@185740 by Dennis Plunkett

flavicon
face
>Very useful is "household epoxy" (as its called here in NZ) which is
>epoxy resin with about  a 1 minute pot life. Sets to good tack
>strength in 2 minutes (firm and unworkable - would easily hold an IC
>socket) and is very very well set in 10 or 15 minutes. Its very
>probably not as strong as regular epoxies but works superbly for
>anything that needs strength and speed. Not quite as fast as
>"superglue", gap fills (so no good on crockery) and is 2 part but
>still useful. The resistance between probe tips placed as close as I
>can get them without touching (0.5mm / 0.020"???) is over 30 megohms
>(standard el cheapo electronic meter, low voltage) whether fully
>cured (scrap from bin) or just mixed. I've never tried it in a power
>on situation but it sounds like the glue you want.


This is invariably a "Locktite" product, sometimes called Arldite. This IS
NOT a good choice, as most epoxy resins have the following properties:-

1/      Hygroscopic
2/      Melt (Yes that's correct) at over 70 degrees or loss of bond strength

There are other epoxies that are better, and like all things you pay the
price. 3M have a 2 part mix, name I can't remember but it is gray in colour,
that is perfect for electronics.


Dennis


'Weldbond - Weird Problem with the 16C64A'
1998\09\01@013921 by Peter L. Peres
picon face
> Weldbond, other glue

 The clear 'hot stuff' is very good electrically but not for RF (not for
Tx tank coils etc). The secret of thread-less gluing is, to cut the bar up
into tiny slices or squares or flakes and place the flakes on the place to
be glued, then apply heat to melt. The heat is best applied with a 250
deg. C hot air gun or with a large soldering iron (without touching). When
melted, press opposite part onto it and leave alone for 1 minute. That's
it. The bond can be re-opened by heating the whole board to ~100 deg. C
and pulling the part off.

 Most glues have notorious problems when combined with high impedance
circuits. This is one of the few areas where it does NOT pay to improvise.
Buy the best possible quality electronics grade glue, and you'll never
have a problem. Cyanacrylate and many epoxies sold in 'hobby shops' come
to my mind... I don't know why cyanacrylate, it should be a good insulator
when dried.

 I could tell you long stories about cheap latex solution that corrodes
chips to the point where they have no pins left and fall off (ca. 2
years...), cheap non-remove flux that has 100 kohm/cm after 1 year in
humid air, contact cement that becomes dust when dried properly, and flaky
peeling solder stop masks that can stop the most expensive servos at
improper times...

Peter

1998\09\01@053034 by - Underwater Acoustics Group

flavicon
face
At 22:04 30/08/98 -0400, you wrote:
>that the pins adjacent to Vdd (Pin 10, RE2 and Pin 33 RB0) were at about 3
>Volts.  This wasn't high enough, or have enough current to set off my logic
>probe or be detected by my DMM, but it seemed to be holding the chip reset.

Had you set these pins as outputs?  Added pullup or pulldown resistors?
That might solve your problem if it is possible to do so.  If an input pin
is left floating, and the impedance is high, it doesn't take much leakage
to give strange inputs (I learnt the hard way, spending a long time
debugging strange signals until I realised they correlated with a local MW
station broadcasting from about 1/2 mile away...)

Nigel
--
Nigel Orr                  Research Associate   O   ______
       Underwater Acoustics Group,              o / o    \_/(
Dept of Electrical and Electronic Engineering     (_   <   _ (
    University of Newcastle Upon Tyne             \______/ \(

1998\09\01@102600 by John Shreffler

flavicon
face
part 0 1540 bytes
John Shreffler

This is invariably a "Locktite" product, sometimes called Arldite. This IS
NOT a good choice, as most epoxy resins have the following properties:-

1/      Hygroscopic
2/      Melt (Yes that's correct) at over 70 degrees or loss of bond strength

There are other epoxies that are better, and like all things you pay the
price. 3M have a 2 part mix, name I can't remember but it is gray in colour,
that is perfect for electronics.


Dennis


{Original Message removed}

1998\09\01@124426 by myke predko

flavicon
face
Nigel Orr Asked:

>At 22:04 30/08/98 -0400, you wrote:
>>that the pins adjacent to Vdd (Pin 10, RE2 and Pin 33 RB0) were at about 3
>>Volts.  This wasn't high enough, or have enough current to set off my logic
>>probe or be detected by my DMM, but it seemed to be holding the chip reset.
>
>Had you set these pins as outputs?  Added pullup or pulldown resistors?
>That might solve your problem if it is possible to do so.  If an input pin
>is left floating, and the impedance is high, it doesn't take much leakage
>to give strange inputs (I learnt the hard way, spending a long time
>debugging strange signals until I realised they correlated with a local MW
>station broadcasting from about 1/2 mile away...)

RE2 is going to be an input in the application while RB0 is set to an output
by the application.  Now, the problem was, that the application never
started, so RB0 could never be set to an output.

I've done this before with Weldbond with other PICMicros (take a look at the
Project pictures in "Programming and Customizing the PIC Microcontroller");
this was the first time with a 16C64A.  Does anybody have any ideas about
why this one would be more sensitive to this?


I also wanted to thank everyone for their suggestions and ideas.  It was a
lot more than I was expecting!

myke

Check out the "Handbook of Microcontrollers" as a reference for embedded
microcontrollers including information on the Intel 8051, Motorola 68HC05,
Microchip PICMicro, Atmel AVR and Parallax Basic Stamp:

http://www.myke.com/#MCUHand


Hunter S. Thompson's quest for the American Dream, this week in the Book Room.

http://www.myke.com/Book_Room/book1a.htm

1998\09\01@182059 by Dennis Plunkett

flavicon
face
At 10:12 AM 1/09/98 -0400, you wrote:
>I have encased short term underseas electronics in epoxy, so
>although it does have a hygroscopic number not suitable for
>long term dunkings, the question was about a suitable
>glue for such tasks as assembling sockets to perf board.
>The clear 5 min curing stuff is perfect for most applications
>where it is OK for the unset epoxy to run.  For non-sag, I would
>recommend another widely available epoxy paste product
>called PC-7.  Epoxys, especially the clear  5 minute runny
>ones *do* melt when you touch them with a solder iron tip,
>so I suppose the 70 degrees Dennis cited may be 70 degrees C.
> But again, in non critical, room temperature applications, you
>will not see the product start melting.  Epoxies are an excellent
>choice for electronic work.  Many components are encased with
>it, in fact.  For potting, as strange as this sounds, I choose a
>pourable epoxy that is filled with metal dust, called Liquid
>Steel by Devcon.  For low voltage applications, including 120V
>it has never shown any tendancy to conduct, although I would
>suspect that would change as you got into HV applications.
>

Try, using glass beads to fill the epoxy instead, they are cheeper.

Dennis




>John Shreffler
>

1998\09\02@095244 by Andy Kunz

flavicon
face
>Try, using glass beads to fill the epoxy instead, they are cheeper.

Microballoons are cheaper still.

Andy


==================================================================
Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
==================================================================

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...