removing "snot-like" goo from ABS circuit board- Bluebee pls. read!

Originally Posted by geobrick

I played with a few settings like the chirp noises and door locks. I wanted all doors to open on one key button press.

I need both of those set myself.

Let me know the procedure.

I just picked up a used laptop for this purpose & need to install the software first though.

Trebbia, Bluebee, To move the discussion of the car programming off of this topic, I searched for the topic where I figured out how to do it with a carsoft cable, IPNA, and NCS Expret. Start at post #64 in this thread. We can then start a new thread if there are other questions. The key to doing it was in a link to a polish forum (I used google translate). I followed those steps. It seems I never provided a detailed step-by-step but we can start a new thread on this forum to do that.

I'm not at home until the weekend so I won't be able to follow-up on the brake module progress until then.

Fixed the alternator using a Bosch regulator from pelican. Now measures 13.9V. I also put the brake module back temporarily (before I hooked up the battery and stated the car - I didn't want any new codes set). The car started, the alternator works but the 3 brake module related lights remained on.

As I showed in my early photos, there was no wire that looked loose but when I checked using a toothpick I found the loose wire on the 7th poke (starting in the upper left corner when oriented as it's shown in the photo or the corner closest to the engine heat shield when the module is installed). The loose wire is on the same connecting pad as the one shown by Joyism5.

Here’s some images. The first one is a close-up of one of my previous images but this one points out the loose wire. The second one is a newer image from a different angle. You really can’t tell where the disconnect is.

I just can't believe how fantastic these photos are!

I consider myself good with the camera - but - this is MUCH BETTER than I can get.

Let us know how you end up re-attaching that loose wire so we can all follow in your footsteps.

Note: Photos shrunk to 640x480 and attached for posterity.

I just tried to reattach the wire using the solder gun. First I used no solder, tried just to reheat the wire....NO LUCK. Then I tried to use solder...NO LUCK. The solder did not stick to the golden pad.
I will update a video shortly. Bad image quality, I can tell from now.

geobrick , I am very curious how you will stick the wire back. Now I'm switching back to the plan A, to use the e-poxy conductive.

Originally Posted by joyism5

I just tried to reattach the wire using the solder gun. First I used no solder, tried just to reheat the wire....NO LUCK. Then I tried to use solder...NO LUCK. The solder did not stick to the golden pad.
I will update a video shortly. Bad image quality, I can tell from now.

geobrick , I am very curious how you will stick the wire back. Now I'm switching back to the plan A, to use the e-poxy conductive.

Joyism5, My plan is to try to reshape the wires to form a bridge with a level area at the ends (like an omega shape) and solder them together. Since my eyes are 51 years old, I haven't been able to do this kind of delicate work but I'll try it with a magnifying glass and lots of light. Then I'll try to "tin" both pads with silver solder. I'd use a soldering iron with a pointed tip. If the solder bonds to the pads, I'll use tweezers to hold the two wires (that I previously soldered together) and try to "tack" solder it onto the pre-tinned pads.

The biggest hurdle is that the wires are so small. Trying to do anything with them is not easy. My back-up plan is to tin a thin piece of "solder wick" and shape that into an omega shape and try tacking that onto the pads.

I'm not sure how to clean up the resin after soldering. I've never done that before but it may be required here. Also, I planned to use some silicon dielectric grease to fill in the gap left in the goo. Does that seem reasonable?

Bluebee, For the photos, I'm using a canon S-95 (a high-end point and shoot elph). I set it to the highest resolution and used macro mode. I crop it as needed, add text and arrows and reduce the size to 650 pixels wide (only if it's still required after cropping). The other thing I've learned recently is to use the png format instead of jpg for the final saved image because jpg is a lossy format and since the original photo is already in the lossy jpg format (though the S95 has a raw option that would probably be better to use for this work). So when you open a jpg file which already has jpg artifacts (though hard to notice), re-saving the edited jpg as a another jpg creates a second level of lost detail. I learned that png compression is not a lossy format and if you use png 8 vs. 24, the file sizes are comparable with jpg. Png 8 has a smaller color pallet than png 24 but both retain more image detail than jpg. I am not a photographer or photo editor (just an engineer who wants to know everything about everything - except the gossip, called news, on TV).

So far, I have only been able to successfully get the solder to flow and bond to the gold pad but not to the gold area on the circuit itself.

Even if/when I do get solder to bond with the gold near the circuit, I don't think I'll be able to reuse the original wire. It's too short and small to work with. I'll post pictures soon.

Originally Posted by geobrick

Does that seem reasonable?

Hell, yeah. From where did you get the silver solder? I'd thought so, that something is wrong with the regular solder. I will go with the 2nd attempt after I will see your pictures.

Here is my video:

I give up to use the DSLR camera which I have, the lens is not a mach for these photos.

Cool video. It looks like you're going about it correctly but when solder balls up like that, it's not flowing onto the surface. My biggest problem is tinning the gold surface on the circuit. Maybe 350c is not hot enough for gold. I'll have to look it up.

Originally Posted by geobrick

Cool video. It looks like you're going about it correctly but when solder balls up like that, it's not flowing onto the surface. My biggest problem is tinning the gold surface on the circuit. Maybe 350c is not hot enough for gold. I'll have to look it up.

I tried to heat it up with my soldering gun (pointing straight on the gold pad) but that didn't help. Here, a more experienced guy like you will step in with a resolution. I can feel it from now.
PS. when comes the time to shot some pic. , shoot one with your soldering gun please. Just to have an idea howfine the point it is.

I'm taking a break for an hour but here are some photos.

The set up.


The soldering equipment.






Progress (?).

Great, great pictures. I have the same mirror, but the soldering iron is far away from the one you have. I do have the conic tip but I need to buy the silver solder. Nice job with the upper gold pad!
I'm waiting to see how this will turn out for you. I will let the Master first...

That seventh silvered wire is a beatch!

Nice pictures though.

Makes me want to open mine just to see how BBA did it!

I didn't do much yesterday but I took the module to work to see if anyone else had any suggestions. Someone suggested I try bringing it to a jeweler who works with gold. Not a bad idea. Someone else said the wires are most likely ultrasonically bonded during the manufacturing process.

Here's a video of ultrasonic wire bonding I found on YouTube.


Today I tried to make use of the "nubs" left behind by the original bonds. These are well bonded to the gold pad. If I can bond solder to the two nubs, I can repair the circuit. If anyone has a better term than "nubs", I'm willing to change it.

As you will see from the photos below, I haven't made any real progress but I was able to confirm I'm heading in the right direction. I accidently created a solder bridge across the two pads and while it obviously didn't form a good bond, it did create an electrical connection. I decided to try it in the car for a temporary test. The brake and ABS lights were off but DSC light remained on (the switch didn't help). During my test drive of about 2 miles I saw the brake light come on for a second or so. I'll continue trying to bond something to those nubs. Hopefully a good bond will take care of the DSC light too (otherwise there may be a separate issue).

Photos.

I can't get a good solder bond on the gold pad.

I'm out of ideas and it's time for me to move on to other projects like "How to code a new ABS module". I'm going to order a new one and see how that goes. I'd send my old one to be repaired but it seems to be hit or miss and I don't know how much life is left in it (other wires etc).

Here's my last photo for this ABS module autopsy topic.

geobrick, thanks you for the detailed post and pictures. I still hope that there is something that we could use home to get a good weld on the gold pad. BTW the solder bridge looks great (in my opinion) and like you stated, I think is not enough contact for this circuit. I am wondering now , what did the Russians used to solder this circuit? I remember reading that thread in bluebee's archive couple months back. I will contact a friend , he worked for about 15 years in a jewelry, and ask him what could make a good weld on the gold pad. Also I will contact some friends in Europe, they work with circuits of all kind, maybe they have a clue.
I will post whenever I find something.

Originally Posted by joyism5

geobrick, thanks you for the detailed post and pictures. I still hope that there is something that we could use home to get a good weld on the gold pad. BTW the solder bridge looks great (in my opinion) and like you stated, I think is not enough contact for this circuit. I am wondering now , what did the Russians used to solder this circuit? I remember reading that thread in bluebee's archive couple months back. I will contact a friend , he worked for about 15 years in a jewelry, and ask him what could make a good weld on the gold pad. Also I will contact some friends in Europe, they work with circuits of all kind, maybe they have a clue.
I will post whenever I find something.

I'm not sure what the Russians have used but the one thread I saw with a video of the repair only shows the repair being done on the larger (copper?) pad. The wire was still well connected to the gold side. I'll be curious to see what you come up with but in the mean time I'm going to focus on learning how to code the new module so when it gets here I'll be ready.

I saw the Russian video earlier in this thread.

Some of the rebuilders won't charge you if they can't fix it.

All of them will take opened abs computers.

Got few answers back, the guy with the jewelry could only tell me the gold temp point is around 1400F. My friend told me, that he cleans very well the area with a pencil eraser if possible, after that washes the area with boric acid, and uses a solder that contains a high alloy percentage . Very important is to have the area cleaned well. And also a regular iron soldering gun . After reading a little bit on the internet I have found some info :
Wikipedia: Tin-based solders readily dissolve gold, forming brittle intermetallics; for Sn-Pb alloys the critical concentration of gold to embrittle the joint is about 4%. Indium-rich solders (usually indium-lead) are more suitable for soldering thicker gold layer as the dissolution rate of gold in indium is much slower. Tin-rich solders also readily dissolve silver; for soldering silver metallization or surfaces, alloys with addition of silvers are suitable; tin-free alloys are also a choice, though their wettability is poorer. If the soldering time is long enough to form the intermetallics, the tin surface of a joint soldered to gold is very dull.

I will stop by Radioshack soon to see what they have for soldering gold plated circuits. I also found the pdf attached somewhere on the internet. Credit goes to....whoever posted it.

Joyism5,
That's good information about the gold melting point. 1400F = 760C. The highest I had my soldering iron set to was 390C. But when we solder copper, we don't take it to its melting point. We just heat it enough for the solder to melt and flow onto it. The problem I think we're having with the gold pad is that gold is a very poor conductor of heat and the heat from the soldering iron isn't getting to the gold in a way that allows it to get hot enough for the solder to flow onto it. If you look at my last photo, you can see the solder curves under (like a ball) when it hits the gold pad. A good solder joint would flow outward.

So how do we get the gold pad hot enough to let the solder flow. We don't need (or want) to heat the gold pad to its melting point because we don't want it to melt. I tried putting the tip of the soldering iron firmly on that gold pad for 30 seconds at 390C and still couldn't get the surrounding area of gold to get hot enough to take solder. A higher temperature or maybe a different soldering tip material would help. My equipment maxes out at 450C and the soldering gun I used to melt open the cover doesn't have a precision enough tip to work in that area.

The solder I used, while called silver solder, was still mostly tin (Sn62 - I assume means 62% tin). I'm not sure how much silver is in there. The rest is probably lead. It also has flux and the problem with the flux is that while the flux is liquid, it does a great job of cleaning the surface but if it dries up before you get a good bond, it crystallizes and actually prevents bonding. You may need a solder without flux but I'm not really sure about that.

Let us know how it goes with your fix and I'll get back to everyone next week about how to self code an ABS module (I hope).

The new brake module arrived earlier this week and I’m going to need some help coding it. This is probably not the best thread to post the details of my progress since this thread is focused on the autopsy and repair of the module itself. If any of you have a recommendation on the best thread to post the details about getting NCS Expert to code the module, let me know your suggestions.

For this thread let me add my experience with how the warning lights behaved with a new, un-coded module. In my case, all 3 warning lights remained on. You might ask, “How do you know the new module is not defective?” The answer is both carsoft and INPA can communicate with it. They couldn’t with the bad module. The software couldn’t find the old module but the new module responded. See the limited details below (I’ll provide much more in the proper thread). (side note: I’m still forced to use the carsoft interface because I haven’t been able to get my modified VAGCOM interface working yet).

Below are some pictures of the new module with some screen shots indicating it’s alive and well but still un-coded.

For Bluebee: While I was working with NCS Expert, I created a screen shot of the car settings page showing how to set the speed controlled door locks. I posted it below but if you can recommend the best thread to post the detailed "how to", I"ll put more information there.

Images of the new module.






What are these?




Here's how I know the new module alive (but un-coded).
The old module would not respond to carsoft or INPA. INPA reported "module not found" and carsoft had something similar. The new module however showed these results:


INPA says there's an ID mismatch. That seems right since I haven't coded it yet.


Carsoft seems to report even more detail. It shows that the transmission type, steering angle sensor, and ASC intake valve information need to be coded. I think that's a good sign.


A quick off topic note for bluebee and anyone else interested. Here is the NCS Expert screenshot for the auto-lock speed settings. I'll provide a step by step within the appropriate thread.


When you're on the car programming page the word "Verriegelungsschwelle" means "locking threshold". The way NCS Expert works is it reads all the current settings first (no need to click "Read" again. Those current settings are indicated with the "S" or "I" in the right column (among the other setting choices). My understanding is that the "S" indicates the setting and the "I" indicates the individual setting per user's key.

The way you choose a new setting is to select the item you want to change in the center section. Choose the setting on the right side. Then select "F2 - Take Over". This puts the new setting in the bottom section (in the photo I am about to change 16km/hr to 8km/hr). After you have all the settings the way you want them, click "F4 Broadcast". That codes the new setting to the car. To understand how to get things set the way you want them will take some translating to figure out how some settings are dependent on others. The example here shows how to pick the speed the doors will automatically lock, but what if you don't want it at any speed? Then you'll have to set the function "Verriegeln Aut AB X KM/H" to "inactive". I tried translating most of it but some of them include German acronyms. I can't pretend to know what it all means but you can experiment some yourself.

Done with the 2nd attempt for soldering the module....no luck. Just like geobrick stated : I couldn't get a bond on the gold pad. I have bought the silver solder (4%), bought a butane soldering gun and still the gold pad did not weld with the solder. I have been trying couple times, last time I kept the tip to heat couple minutes, put it on the gold pad for another minute or two and still did not heat well enough for the silver solder to bond on. They said it will heat up to 2400F but...flame maybe, not the tip. I used it at maximum temp. (soldering tip) but the gold pad did not heat well enough. Or the soldering silver is not the right one since it makes a reaction like crystallization/white paste. Again,like geobrick stated above. At this point I am looking around to send it to somebody to fix it. Or should I buy another one? I'm still debating.

I think bonding to gold takes some very special tools (like the ultrasonic bonding in the video above). Gold does not conduct heat so the solder won't melt unless it hits the soldering tip but that's not where we want it to flow.

I wish I had some tips but as you know, I gave up and bought a new one (still trying to code it though).

It's done. Somehow I coded the module.

I used a carsoft interface with DIS.

I'm not sure what I did specifically because when I tried coding using NCS Expert, it seemed to work but there was no positive feedback that it really coded plus the infamous trifecta lights remained lit.

So I spent several hours over several days learning to install DIS so that I could do the steering angle sensor calibration. I followed the excellent installation instructions from Randomy on the bimmerforums sight (I used to think bimmerfest and bimmerforums where the same web site).

After some hurdles getting DIS working, I did the steering angle sensor calibration (DIS is very cool). Then I tried to use DIS to code the module but got the following error:

Pay no attention to that part number because even when I came up with the right part number for the old DSC module, "6755741", I got the same result.

Assuming I didn't accomplish the coding, I continued researching until today when I started the car and the lights were gone. I checked for any error codes and there were none. Carsoft no longer shows the errors shown in my earlier post.

I took the car for a test and did some hard stops with the right side of the car on gravel and the left side on the road and the ABS kicked in fine. Then I did some quick starts with the back wheels on gravel and the DSC kicked in. So it's working but I'm not sure if all I needed to do was the steering angle sensor calibration or that along with the coding in NCS Expert.

Nice job. I'm glad you did some progress. How much you spent in total?