Idea for removing potting compound

[Whorse]

This is a really small thing, but I'm working on one right now. I don't have a toothbrush to sharpen, so had to comprimise.

Used a small exacto blade and started slicing 'layers' off the top. This was a reman board but the chip I want hasn't been cut out, so doing that (practice board mostly). When I got to the bottom nearly to the traces, I needed something that wouldn't damage the traces.

I grabbed a medium sized tie wrap/zap strap/cable tie/whatever you call it, and cut it so the weak end was gone and had a nice flush sharp edge on it. Seems to be working well for removing this stuff. Just a thought. I'm going to see what chemicals we have at work for removing this crap.

Is this stuff even epoxy? Seems silicone based.

[looneytuner]

Try soaking in acetone.

[Whorse]

:p gotta take it to work to get some

[87glhs232]

I don't have a toothbrush to sharpen, so had to comprimise.
.
You don't have a toothbrush? :o

I'm can't remember the chemical make up of that compound, but I do remember it's NASTY stuff. I wear gloves when handling them. If memory serves it had cyanide in it, or a compound that became cyanide when you melted it (ie. with your soldering iron). Be carefull!

[GLHNSLHT2]

a nylon/rubber brush on the dremel works great after cutting most of it off with my Swiss Army knife :)

[Whorse]

I'm not melting it right now. I did however find that isopropal with a toothbrush (that is rubber and can't be sharpened) and my zap strap is doing an amazing job. It's been less than an hour and I've almost entirely done both sides .I wouldn't hesitate to solder this, but I have time to clean it up nicer.

Update: Added a few photos. Before, During, and Now.

[cordes]

Nice work.

[Whorse]

Thanks! I've been reading your bit on the SBEC calibration. I think your writeups are excellent and a big help to people learning.

I am going to take this to work and hopefully remove the chip as is and keep it just in case. I'd like to see if we have any equivalent chips at work that might fit this, and I want to see what I might be able to order in for sockets.

I will see if we have anything that cleans this compound up any more.

[butchsuppe]

It,s almost impossible ( or maybe impossible ) to remove the chip in one peice since the potting glues it to the board. I,ve always used a dramel with a cutout whl. to cut the pins off the chip as close to the to top as possible. I don,t think you should use any chemical as it may damage the board, just my 2 cts. 1st board I did took 4 hrs, got it down to 2 hrs after doing 10. Buy a toothbrush to clean the board after the chip is removed. Good luck + enjoy the fruits of your labor.

[Tony Hanna]

Not to hijack your thread Jason, but does anybody make a remote socket/ribbon cable setup that a person could use to to mount the chip externally? I'm cleaning the potting on one myself (if I can find the box it's packed away in) and figured it'd be neat to do an external socket mounted inside a project case for easy access when swapping chips.

[sdac guy]

I use a small wooden stick that was made for electronic probing where non conductive nature is important. Then I use the business end of an old toothbrush to clear most of the crumbs, then I hit it with compressed air from a can (holding the board over a wastebasket to catch the crumbs).

Do not ever use any chemical to disolve the compound. The fumes from the compound in liquid form are extremely toxic.

Una has a very good process posted on his site (I don't recall the link) for doing the SMEC/SBEC's including many pics of progress along the way.

He mentions using side cutters to cut the pins on the chip and that is what I do. There is no metal dust doing it that way as there is when using a dremel to cut them. Probably takes a little longer than the dremel though.

Barry

[ShelGame]

Not to hijack your thread Jason, but does anybody make a remote socket/ribbon cable setup that a person could use to to mount the chip externally? I'm cleaning the potting on one now myself and figured it'd be neat to do an external socket mounted inside a project case for easy access when swapping chips.

www.moates.net

He has a lot of really cool stuff that we could use...

[87glhs232]

It's almost impossible ( or maybe impossible ) to remove the chip in one peice

Butch- I have to disagree with you there. True the compound is under the chip as well, but I have NEVER had to cut the pins. Last one I did was for mech1nxh, he didn't want a custom cal, but rather verify what the cal on it was (thought it was a mexican cal). I pulled the chip, read it, socketed the board and popped the chip in the socket then water proofed it. But to each his own. I'm sure it's a lot easier to cut the chip out and deal with the remains.

[sdac guy]

Butch- I have to disagree with you there. True the compound is under the chip as well, but I have NEVER had to cut the pins. Last one I did was for mech1nxh, he didn't want a custom cal, but rather verify what the cal on it was (thought it was a mexican cal). I pulled the chip, read it, socketed the board and popped the chip in the socket then water proofed it. But to each his own. I'm sure it's a lot easier to cut the chip out and deal with the remains.I agree that I've never had a broken chip as long as all the pins have had 90% or more of the solder removed.

I cut the pins so there is much less chance of ruining the hole or solder trace near it by overheating it trying to remove the solder.

Cutting the pins, removing the chip, then removing the pins with a soldering iron, and then sucking the solder from the hole is far easier than trying to get all the solder out with the chip and pins in place. New chips are cheap enough.

Barry

[Tony Hanna]

www.moates.net

He has a lot of really cool stuff that we could use...

Nice site!
Looks like the EMUC 2818 cable combined with the HDR1 memcal header would work.
Thanks Rob!

[butchsuppe]

Nice site!
Looks like the EMUC 2818 cable combined with the HDR1 memcal header would work.
Thanks Rob!
If you build this let us know how it works, I think you got a good idea there.

[Tony Hanna]

If you build this let us know how it works, I think you got a good idea there.

If/when I build it, I'll start a thread with all the details and pics.:thumb:
It's only $20 for the parts to do it, so if it tanks, I won't be out much.
I think I'm getting ahead of myself though. I haven't even got the potting cleaned off a board yet. :o I started, but it got boxed up when I moved and now I can't find the box.:( This was just an idea that hit me while I was considering what a pain it's going to be to remove and disassemble the smec every time I want to change the cal. I figure I'll be doing that pretty often as I learn.

[Whorse]

Which parts are you adding exactly? I have a board here with the potting compound all removed that I could try on and see if it works?

On a different note, I'm looking at a 90s TBI SBEC and a 91 Turbo SBEC. TBI has one less heat sink on it. Anyone know the physical differences?

[Tony Hanna]

Which parts are you adding exactly? I have a board here with the potting compound all removed that I could try on and see if it works?


The first pic is the ribbon cable (EMUC2818) from the Moates link Rob posted.
The second pic is the memcal header (HDR1) also from Moates.

The idea is to plug the ribbon cable into the socket on the smec/sbec on one end and the connector on the header on the other end. Then plug your eprom into the socket on the header. If it would work, it would allow you to mount the eprom externally (I was thinking of a small project case velcroed to the side of the smec/sbec) for easy access.

If you decide to try this, keep in mind that it's definately not a sure thing and may turn out to be $20 + shipping down the drain. I don't know what (if any) effect the length of the ribbon cable would have, and I'm not entirely sure what kind of socket is on the other side of the header since that's the only picture. That could be fixed easy enough with a zif socket though.:)
I'm curious to see if it'll work, but it'll be awhile before I'm ready to try it.

[sdac guy]

... On a different note, I'm looking at a 90s TBI SBEC and a 91 Turbo SBEC. TBI has one less heat sink on it. Anyone know the physical differences?The TBI engine doesn't have as many injectors (only at the throttle body) as a turbo car (4-MPI) so the SBEC for the TBI will not work on a turbo car without heavy modification. That's why it has only one heat sink.

Barry

[Whorse]

The TBI cal is fried supposedly, so that's why I have it (practicing removing the compound).

Does anyone know the chemical makeup up this stuff?

[Tony Hanna]

The TBI cal is fried supposedly, so that's why I have it (practicing removing the compound).

Does anyone know the chemical makeup up this stuff?

Pretty sure I read somewhere that it's isocyanate (sp?).

[Whorse]

http://en.wikipedia.org/wiki/Isocyanate

Here's what Wiki says. Sounds similar to what we're working with. Any chemists here know what eats this stuff away and doesn't hurt electronic components :nod:

[ShelbyMotorsports]

The TBI cal is fried supposedly, so that's why I have it (practicing removing the compound).

Does anyone know the chemical makeup up this stuff?

The potting compound I use is epoxy based. 50/50 mix of part-a & part-b.

On a related note I just opened up a Mopar Reman SMEC and they do it the same way we do which is removing just enough of the potting compound to get at whatever compnent thats being replaced and filling the hole back up with new compoundwhen your done.

Steve

[cordes]

Geoff over on moparchem.com says that it would be a good idea to keep the ribbon cable pretty short, as the length can affect the operation of it. I also like the idea of having the case velcroed to the SBEC.

[Tony Hanna]

Geoff over on moparchem.com says that it would be a good idea to keep the ribbon cable pretty short, as the length can affect the operation of it. I also like the idea of having the case velcroed to the SBEC.

Did he give any guideline as to how short? The shortest one moates has listed is 18".
I figured the length of the cable might have an effect, but I wasn't sure how long it would have to be to start causing problems.

[cordes]

Did he give any guideline as to how short? The shortest one moates has listed is 18".
I figured the length of the cable might have an effect, but I wasn't sure how long it would have to be to start causing problems.

He was suggesting about 4". I am sure that mouser would have something like that.

[Tony Hanna]

He was suggesting about 4". I am sure that mouser would have something like that.
That's cool. I figured it was a specialty item with the one end being made to plug into a 28 pin socket.

[Whorse]

I tried looking up info based on the chemical compound Isocyanate, and mostly found only datasheets on why it's great for potting circuit boards :p.

[Tony Hanna]

I tried looking up info based on the chemical compound Isocyanate, and mostly found only datasheets on why it's great for potting circuit boards :p.

I have no idea what you could use to chemically remove the stuff. I'd be a little worried that anything strong enough to remove the isocyanate might damage the board or any components it comes in contact with. On top of that, you'd definately want to be careful about any fumes coming off of it. I read the wiki link you posted. They use phosgene in some part of the process to make it. Isn't that nerve gas?:eek:

[Whorse]

Yeah I wonder...I'm looking at these two SBECs on my desk and staring at my hands hoping I washed them well enough...

So far, the isopropal is helping, but it doesn't eat away at it, just softens it a bit and dipping the tools in it helps out a lot.

[Whorse]

Here are some pics of a TBI SBEC I worked on today. Fried board, so just for practice.

[Tony Hanna]

Looks good man!

[Whorse]

Thanks! I'm gonna try to find a stiff bristled rubber brush to clean a bit more of that stuff off so it doesn't burn when I solder it. I think when I get around to it I will try using solder wick to remove the solder.

[Tony Hanna]

Thanks! I'm gonna try to find a stiff bristled rubber brush to clean a bit more of that stuff off so it doesn't burn when I solder it. I think when I get around to it I will try using solder wick to remove the solder.

I'm still trying to find the one I have apart. I backed the Sundance out of the garage a few minutes ago to make some room to drag out boxes...:)

[csxtra]

Not to hijack your thread Jason, but does anybody make a remote socket/ribbon cable setup that a person could use to to mount the chip externally? I'm cleaning the potting on one myself (if I can find the box it's packed away in) and figured it'd be neat to do an external socket mounted inside a project case for easy access when swapping chips.

I have been running a setup like this on my CSX since early last year. If I remember correctly, I'm using about 5 feet of ribbon cable, enough to reach from my logic module to the glove compartment door. Works fine, no issues encountered so far, even after driving back and forth to Topeka last year (110degF outside temps).

I used a DIP plug (http://www.jdr.com/interact/item.asp?itemno=gr-idp08) to connect the ribbon cable inside the LM, and then soldered the other end onto an Aries ZIF socket on a mini breadboard inside a Radio Shaft project box. A little tedious doing the soldering, but after an hour or so of work, I routed the ribbon cable, plugged it into the LM, velcroed the box to the inside of the glove compartment door, dropped a chip into it, and the car started right up.

I thought I might need to shield the ribbon cable, but as it turned out, it wasn't necessary. But since this was for a LM that is inside the car, I didn't have to worry about the underhood environment (EMI, RFI, moisture, excess heat/cold).

Hope this helps,
Warren

[Tony Hanna]

I have been running a setup like this on my CSX since early last year. If I remember correctly, I'm using about 5 feet of ribbon cable, enough to reach from my logic module to the glove compartment door. Works fine, no issues encountered so far, even after driving back and forth to Topeka last year (110degF outside temps).

I used a DIP plug (http://www.jdr.com/interact/item.asp?itemno=gr-idp08) to connect the ribbon cable inside the LM, and then soldered the other end onto an Aries ZIF socket on a mini breadboard inside a Radio Shaft project box. A little tedious doing the soldering, but after an hour or so of work, I routed the ribbon cable, plugged it into the LM, velcroed the box to the inside of the glove compartment door, dropped a chip into it, and the car started right up.

I thought I might need to shield the ribbon cable, but as it turned out, it wasn't necessary. But since this was for a LM that is inside the car, I didn't have to worry about the underhood environment (EMI, RFI, moisture, excess heat/cold).

Hope this helps,
Warren

Nice! So, if you didn't have any problem with a 5' cable then an 18" should be good to go.:)
Thanks!

[Mike M]

Well, its nice to know that the chip extension CAN work. As soon as you first posted it Tony I went and did a bit of research (before finishing reading the post) on whether this is a good idea or not. I am using a Moates burner and chips already so I browsed their forum and sent Craig Moates an email. One thread I ran into on the forum was asking this same thing, Craig had replied "anything over a couple of inches doesn't work dependably. Either interference or lack of signal strength.".

So we have an argument one way and the other now, maybe the LMs being mounted inside the vehicle would be less prone to issues? Perhaps, I don't know. I'm waiting for Craig to email me back although I'm pretty sure he'll reply that its not a good idea. Unless we can find a real short cable, in the 6" range. I will look into it more.

I started to get into the D-cal/chip burning last year but I couldn't get it to work or get the car to run, so I gave up on it last year over winter. I tried it again yesterday, still not working, as a last ditch effort I installed my chip upsidedown and the $#%$W%^$^ing car started right up! :p So now I'm back into the chip burning game, and I'd really like to get into the chip burning thing (without R&Ring the ECU all the time!).

[Mike M]

Wow, looked right past it on Moates - http://www.moates.net/product_info.php?cPath=50&products_id=133

I'm just going to wait for Craig to reply, I'll probably order it whether he thinks its a good idea or not ;)

[gkcooper]

OK guys, I used to work in the Chrysler (Accustar) plant that re-manufactured everything electronic. In fact, I was the manager of the Mopar Tech Hotline. We used a water jet to remove the potting. It was the best way after much trial and error. Just use a pressure washer or carwash. Be careful not to leave any moisture under any of the surface mount packages and you will be perfectly OK. We did it in the thousands. I would just place it in 140 degree oven for an hour or two afterwards.

Remember that potting is isocyanate based resin, so IT IS TOXIC. Don't leave it in your eyes, under your nails, and don't make a sandwich out of it.

Greg K. Cooper
Huntsville, AL

[Tony Hanna]

Wow, looked right past it on Moates - http://www.moates.net/product_info.php?cPath=50&products_id=133

I'm just going to wait for Craig to reply, I'll probably order it whether he thinks its a good idea or not ;)

I must have overlooked it as well. Yeah, it can't hurt to try, and you're not really out much if it doesn't work for some reason. With what Warren posted about using the 5' cable though, I really can't see any reason why it wouldn't work.

My project has sort of stalled. I never could find the logic board I started on, so I ended up getting another smec from John. I got it apart and the potting taken care of, and then couldn't find a set of cutters small enough to cut the pins on the eprom. I'll buy a set and get back to work on it eventually, but this Holset project has me sidetracked for now.:)