1990 Instrument Cluster

[Guest JohnW]

My cluster works and then goes blank without any consistenancy. The cluster may be working when the car is shut off and dead (only showing 00 and ERROR) when restarted. Also the cluster may be dead upon starting the car and then work fine for hours. No rhyme or reason. Bumps dont seem to affect the screen either. All indicators light up when pressing the test button. as do the turn signal indicator lights on the cluster. Is there a relay that my be at fault? No work was performed on the car except I changed oil (myself).

[Ronnie]

This may help:

Reatta Owners Journal - Dashboard Reads 00 & ERROR - '90 & '91 Models

Reatta Owners Journal - Digital Dash Panel/IPC Removal - '90 & '91 Models

[hinspect]

John, I have the exact same problem on my 1990, your description fits it to a "T". It just started a week or so ago. Sometimes it works and sometimes it does not. I have not taken Ronnie's advice yet and checked my grounds or connections on the display but it will be the next thing I do when I have the time. I have to drive it in a few minutes to Knoxville and have no Idea whether it will work or not by the time I get back. Maybe we will all learn something from this one.

[Guest JohnW]

Can you spray electrical cleaner directly into the ignition switch without causing a electrical short?

[Jim]

In the '88-89 Reattas the most common parts to go bad are the CRT/touch screen and the headlight switch. The IPCs were very good.

In the '90-91 Reattas the IPC is the most common part to go bad and when they do they will read ERROR and 00. They can be intermittent or go black and stay black. Cleaning the contacts on the rear might help but usually not.

I have some good tested working IPCs available and can be reached at jfinn@cpinternet.com

[harry yarnell]

Jim pretty much nailed it; bad IPC.

[Guest JohnW]

I do not necessarily dispute the fact that the IPC is in fact the cause; however, my engineering mind has some problems with this conclusion. For 33 years I helped build, repair and maintain locomotives. I have seen mechanical problems start and unexpectedly repair themselves like a leaking water pump that seals itself, strange noises that "just went away" or a smoking engine caused by bad fuel. Electrical problems seem to be a different animal all together. Rarely (if ever) have I found an intermittent electrical fault that wasn't directly or indirectly caused by a physical or mechanical connection such as a problem relay or a loose or shorted wire. It seems that true electrical problems caused by electrical components either work or don't work and the days of a "leaking capacitor" are long gone. Has any root cause analysis ever been conducted by the intermittent failing IPC's? I personally would like to know the results. Thanks to all on this board who helped me and others keep their Reatta running.

[Guest Mc_Reatta]

Solid state electronics usually fail intermittently from heat and cold, voltage and or current fluctuations due to the input sources and faulty connections supplying power and signal sources to the device.

Those that repair IPC's seem to be tight lipped about what they know. I'm sure there are some failures due to bad solder joints and cracked traces on the circuit boards from age, temperature and vibration. But I would expect hard failures to be due to some chip in the device like a display driver or bus controller. Since these were probably proprietary to Delco, good luck finding out how to test and what a suitable replacement would be. I expect those that can fix them have used knowledge of similar displays from that era and used some of the parts form them to replace items by trial and error to find some that work as replacements.

That said, I've never looked inside of one myself. Don't have a 90.

[KDirk]

I'll add my thoughts here for what they are worth.

The 90/91 cluster "00/Error" issue can result from two general problems:

1. External failure of the vehicle data bus, resulting in no, or corrupt, data to the IPC. This can be as simple as a open or short in the ALDL network wiring, a loose/poor contacting connector or a failed module (BCM is a likely suspect in this case since it controls the bus). If the IPC does not see the correct data stream, "00/error" is the automatic response, along with the electrical malfunction telltale. This is a fallback mode that the IPC initiates when it does not have the data it needs to work with.

2. An internal failure of the IPC electronics. Now, the most likely culprits in this case are either cold (or cracked) solder joints from vibration and thermal expansion/contraction or failed components on the circuit board. In the latter case, I suspect many such issues are in fact dried out capacitors. Even the very best MIL-spec caps (which were not used in the IPC) are not likely to last much beyond 20-25 years in the environmental extremes of a vehicle. Time, and many cycles of heat/cold will do this to passive components. Even resistors can drift off value over time. This makes them the most likely points of failure. IC's and transistors very rarely fail outright unless damaged by an external cause, such as a failed passive or a voltage spike/surge.

There are also power supply quality issues to take into account. Voltage can drift a lot based on the condition of the battery, alternator, other loads on the system and environment. While this is filtered and regulated in the IPC (and CPS for 88-90 models, not present on the 91's) poor regulation and filtering also will also do damage that may take time to be apparent.

MC-Reatta is right in that many parts (CPU's and peripheral chips) were custom made and thus hard (or impossible) to find as replacements except from other complete units. It is also frustrating that in this day of almost everything worth knowing being online, that there is scant data available for these units. I have to give credit, tbe outfits that repair these clusters have done a masterful job of keeping the repair data close to the vest.

That said, I don't think it should be impossible to figure out how to repair many of these, and I have worked on the issue intermittently myself. I have no solid answers yet, but I am determined to get some insight on the issue eventually, as I do not like to get beat by a problem that isn't all that difficult.

The biggest difficulty in troubleshooting is that the circuit boards are completely covered in a conformal coating. This was done to protect the circuits against corrosion and stress cracking. It does both those things, but makes the boards much more labor intensive to work on, unless all coating is removed. It is hard to do component tests with a meter and scope unless the coating is removed first, as the lead probes cannot make good contact with pins/solder joints through this clear coating.

The second big problem is the lack of a bench testing rig, that would supply a proper data stream to test the cluster under simulated operation. I have to think there was some kind of tool made for this purpose (like the Kent-Moore CRT tester) but it was only issued to Delco service facilities, so it is likely a very rare item. This would be an invaluable device to have, but I doubt one can be found "in the wild".

So, I will continue to plug away at the several clusters I have hoping that I will finally have a flash of brilliance that will solve most IPC failures of this type. Of course, there is still the blacked-out cluster issue which is different again. That is another battle to be fought another time.

KDirk

[Ronnie]

KDirk, You seem to be knowledgeable about the electronics. Each time the 00/Error on the IPC questions comes up here on the forum I offer the link to my website about that problem. Someone us usually quick to try to shoot down the idea of cleaning the contacts as being any help. In your opinion should I stop offering that suggesting and maybe even remove the instructions I have on my website since some seem to think they are not helpful? I really don't know if the instructions are helping or not since I don't know much about the IPC electronics. I just try to provide helpful information about Reatta repairs in an easy to find format.

Here is a link to the instructions I'm referring to: - Dashboard Reads 00 & ERROR - '90 & '91 Models

[KDirk]

Ronnie,

I don't think the suggestion (or your tutorial) on cleaning contacts is harmful, or useless. This fix costs little other than time and some spray cleaner to try, so if it doesn't work nothing is really lost. Truth is, there is a chance (probably less than 1 in 10) that cleaning the connection points could fix the problem. Of course, this only works if the issue is corrosion or oxidation on the connectors or the pins on the IPC itself.

I can see this occurring for cars that were stored in poor conditions (high humidity/salty coastal areas) for long periods of time, or for salvage yard clusters that may have been removed from cars that were not sealed against the weather.

The thing is, most (90%+) of the cluster failures that result in "00/error" are internal failures of the cluster, and this percentage has been rising as the parts inside get older. This is the problem I have been trying to crack, hoping that I can establish a common cause (specific failed components in the cluster) that will make it possible to fix without dropping $150-200 for a shop to repair it. So far, no dice - but I am trying.

So, to answer your original question, I think you can leave your tutorial, but perhaps with the caveat that it will only be effective in the case of connection quality issues caused by oxidation or corrosion of the pins/contacts. As said above, if someone does this and it still doesn't work there is really no harm done. May as well be tried before resorting to more extensive/expensive repairs.

KDirk

[Ronnie]

Thanks Kevin, Your comments are appreciated and I will take your advice. I refer people to my website if I think the information they are seeking is there. However, I like to think the information they find there is going to be useful and as accurate as possible. I don't consider myself an expert on the Reatta so I rely on the members of this forum to give me feedback about information they find on my website in order for me to keep it as accurate and helpful as possible. That is why from time to time I ask questions like this to try to verify that the information on my website is helpful.

I'm glad that you have an interest in being able to repair a part when something goes wrong. I always get a lot of satisfaction out of being able to repair something instead of just going out and buying another one. There is a growing number of people on this forum who are scanning the threads for an opportunity to jump in and say they can sell you a part to fix your problem. I see nothing wrong with that if it is obvious the person who started the thread is looking to buy a part but I feel it is out of line if the thread is about diagnosing a problem or seeking help in making a repair. We need parts vendors who have parts available when we need them. I appreciate them taking the time and effort to make the parts available but I feel the Reatta Buy/Sell sub-forum is the appropriate place to peddle them.

Edited April 25, 2010 by Ronnie (see edit history)

[harry yarnell]

I'm a retired electronic tech who pride myself on my troubleshooting skills, but these '90-91 panels have left me scratching my head. First off, yes, many intermittents are due to cold solder joints, but I've never found any. Believe me, I've looked. As a previous poster said, the problem maybe an intermittent in an IC or chip, of which replacements arn't available to the peasents.

And yes, service (schematics) can't be found.

Wasn't there a movement by someone on this list to come up with schematics on these panels/ Or was it the crt unit?

The folks that service these do NOT release thier troubleshooting techniques. I had a fellow in Baltimore who would repair these for $90 a pop, and he wouldn't tell me a thing as to what he did. He has since dissappeared.

I have never had any success cleaning/spraying the connector pins. That said, I HAVE had success cleaning the HVAC control module pins when it won't take a button command.

Edited April 25, 2010 by harry yarnell (see edit history)

[KDirk]

Harry,

I suspect that there aren't all that many that fail due to bad IC's. I also suspect replacements for many of these chips are very difficult to find even for "real" service centers anymore, as application specific chips have a relatively short shelf life before obsolescence (say 10-15 years tops, and that is being generous) due to rapid changes in technology.

Since most failures are either the dreaded "00/error" or complete blackout, and most shops will flat rate repair these clusters for either symptom, the time and parts involved cannot be that much or the flat rate pricing would be much greater. Save for the occasional IPC that is really badly screwed, these are probably major profit generators since they have the leg work done and already know the common points of failure. So, 150-200 apiece w/warranty is probably making them pretty good money on average. This is another reason I deduce that the repairs are usually fairly simple (just not obvious to us non-anointed ones).

I can understand why none of the shops has volunteered their service data, they would be cutting their own face off. What surprises me is that no one with the knowledge and tools has taken time to figure this out on their own and post the information openly just for the satisfaction of "beating the system". I suppose that the 90-91 (thru 93 for Riviera) cluster just isn't a common enough item for anyone to have gotten serious about it.

There was a writeup on a mid-late 90's Cadillac digital IPC that someone did, was online somewhere a few years ago, and in that case it was almost all bad solder joints. I agree that the Reatta IPC is not too likely to be bad solder connections, as it's design and build quality was more substantial than many later digital clusters assemblies GM used. The Delco influence seems to be a factor here, many later designs were made under contract by Denso and are not as well put together judging by those I have seen disassembled.

KDirk

[Guest Mc_Reatta]

What also lends credence to the idea that it is a relatively simple repair is that posts by some folks who have had these repairs done and looked inside the unit couldn't see any area which looked like it had been worked on. If these boards are conformal coated, it would be virtually impossible to replace a chip, cap, or repair a solder joint or broken trace without leaving an obvious footprint of where the coating was removed and replaced. Also a new chip or cap would probably stand out in some way from the 20 year old ones. Brighter markings, blacker glossier black color etc.

There was a member, Goggi, that was attempting to get information on Delco / Delphi products (radio, crt, ipc etc.) He supposedly received some info and scanned it and tried to get it posted, by I haven't seen anything show up on reatta net or Ronnie's site.

[Ronnie]

There was a member, Goggi, that was attempting to get information on Delco / Delphi products (radio, crt, ipc etc.) He supposedly received some info and scanned it and tried to get it posted, by I haven't seen anything show up on reatta net or Ronnie's site.

Nothing was ever offered to me for posting. Maybe he offered it to reatta.net. Below is the last post about it that I could find.

January 13th, 2010

The Goggi

Member

Re: Reatta radio schematics

I have a full set of schematics being scanned in full color large format PDF. I am expecting them any day now from being scanned (not cheap). I will forward them to Reatta.net and I am sure they will post them under their docs section. It only requires a login.

Just a little more time and I'll have them for y'all.

[Guest Richard D]

Are there a lot of surface mounted devices on the IP circuit board? I remember that the early CD players and radios that used the first SMDs would fail due to the hot cold cycling cars are subject to would cause micro cracks where the device was soldered to the board. I would fix a Sony CD player and it would play every day on my bench for a week. When reinstalled in the car it would fail sometimes in a couple of days, sometimes a month. Of course here in south Florida the interior gets over 150 degrees on a daily basis. Just my 2 cents.

[Roadster90]

Interesting Thread .... anyone able to contact The Goggi ?

Edited April 27, 2010 by nic walker
typing error (see edit history)

[padgett]

My CD player must be a Florida native then - will not read a CD unless the temperature is over 80F.

[Guest JohnW]

Interesting Failure. My Instrument Cluster now works about 50% of the time. I just happened to look down at the cluster and watched the RPM indicator go down to zero just before the dreaded 00 ERROR appeared. I keep by GPS in the car so at least I have a speedometer.

[Roadster90]

My IPC was intermittent (1990 - display just went black several times) several months ago and then started acting OK up until yesterday.

Now everytime I go somewhere or start the car the display is black - I RAP ON THE PADDED DASH SMARTLY -or hard 3-5 TIMES right above the IPC, and the darned thing lights up and stays lit until I cut the car off....

I know this indicates an open circuit, but is it provable that it it the main gang plug connection or a circuit board maybe????

Thanks,

Nic

[padgett]

Prolly the same thing that makes the CD in the white car only work above 80F.

[bwarren]

I've been working on a 91 IP and have made some progress and will report back as soon as I can get my info organized. As others have mentioned, component sourcing is the hang up. Maybe with a group effort, we can find these parts. I will get back to this as soon as I can.<O:p</O:p

[Guest Richard D]

My IPC was intermittent (1990 - display just went black several times) several months ago and then started acting OK up until yesterday.

Now everytime I go somewhere or start the car the display is black - I RAP ON THE PADDED DASH SMARTLY -or hard 3-5 TIMES right above the IPC, and the darned thing lights up and stays lit until I cut the car off....

I know this indicates an open circuit, but is it provable that it it the main gang plug connection or a circuit board maybe????

Thanks,

Nic

If you have a very small soldering pencil and GM did not use conformal coating on the boards (can someone tell me if there is coating on the board? I have some spray to remove and reapply the coating that was used on aircraft radios, I will share it)you could reflow the solder on the surface mount devices, however since it stays on after you show it YOU are the alpha owner you might try cleaning the connector and and checking it's tightness.

Edited April 29, 2010 by Richard D (see edit history)

[KDirk]

Richard-

There is conformal coating on both sides of each circuit board in the IPC (total of three boards stacked together). I have not been able to determine with complete confidence which type of conformal coating was used, as there are different removal solvents for various formulations of conformal coatings. Using the wrong remover can be a mess, so I have been reluctant to try anything yet. The stuff is also not cheap, so I don't want a $20+ can of stuff I'll never use if it turns out to be the wrong solvent.

FYI, the 90/91 IPCS were mostly thru-hold components, except for a few of the proprietary interface chips (ALDL communications interface/CPU which is a surface mount PLCC type chip for example)

As far as the data The Goggi received, it was mostly regarding the cassette and RIM module for the 88/89 models. He scanned and posted schematics at Reatta.net which I downloaded. Useful data to have, but not anything for the 90/91 IPC problem unfortunately.

KDirk

Edited April 29, 2010 by KDirk (see edit history)

[Guest Mc_Reatta]

That gives a starting point for someone to try after cleaning connectors and other obvious failure inspections turn up negative. Common failure mode in a suspect circuit for an 00 Error display. Coating can be removed mechanically too. Nylon or brass brush in a Dremel at low speed, and/or scraping with an Xacto knife blade. Re-flowing the solder joints on these surface mount chips if done carefully shouldn't hurt and might be worth a try.

[Guest Richard D]

Hi Kdirk, The spray we used is made by GC electronics, There is one type of remover and two types of coatings, one is silicone based, left a rubbery feel to the boards and components. The other is called Specialty Acrylic Plastic and leaves a hard, shiny coating. They recommened three light coats. All three have mil-spec numbers and were about $20-$25 each.

[bwarren]

This is what I have so far. Bad display has no illumination at all. Not the 00 ERROR symptom. Side by side comparisons with a bad display and a good display. Probing circuit until an abnormality is found and removing the questionable components from each unit and testing out of circuit. Continued to fallow traces and removing those components displaying a measured difference until all circuit probing is equal between the 2 units. All defective and questionable parts were found on the middle PCB. CR1 (diode) shorted. I suspect this may have been the component that caused the cascading affect. Next C1 and C2 are questionable. My capacitor tester is not of great quality, but tends to be accurate when values are spot on. Q4, Q5 questionable. Q6, Q7 were visibly damages by heat. I suspect there was some smoke involved as the circuit board itself has some thermal damage. C23 sits next to Q6/Q7, was visibly swollen and makes a physical connection to Q6/Q7. Its value is way off. Obviously having the proper equipment would be beneficial. My only tool is a substandard DVM that has the ability to test transistors and capacitors. I have tried to label the components and indicate the makings. As we have surmised, tracing the numbers is proving to be difficult.

Edited April 30, 2010 by Rawja
font (see edit history)

[KDirk]

Brian-

For having substandard test gear, you have done well. May I ask, did you manage to source replacement components for the transistors? They are obviously Motorola (now days=ON semiconductor) devices, but I am unsure if they are stock part #'s or Delco "re-marks" with house numbers. Somewhere I have a mid 90's Motorola device cross reference buried. I need to find it to see what I can determine about these parts.

The rest (capacitors, diodes) are easy, as they have standard markings. I am curious if you got the blacked-out cluster working again, or if you are not that far along yet. Please post back, as it would be helpful to know if this resulted in a successful repair. The blackout condition is not as common as the 00/error, but if your findings save a few clusters, it was worth the effort.

KDirk

[bwarren]

The repair has not been made yet. Still trying to source parts. Caps should not be a problem. I sent an email to Motorola and got this reply, "Thanks for reaching out to Motorola. I’ve reviewed your e-mail and I’m ready to help. We advise you to call the Symbol Product Department number at 866-416-8545 for proper assistance. Rest assured that we will take care everything for you." I thought the diode would be easy too, but I'm striking out with that one as well. Life is busy right now and this is on the back burner at the moment. I believe if I used the parts from my working unit, the bad one would work, but am not certain of the longevity of the repair. If my attempt was not thorough enough, I might end up with 2 dead units. Not ready to take that jump yet.

Edited April 30, 2010 by Rawja
font (see edit history)

[KDirk]

Richard -

Thanks for passing that along about the coatings. I have suspected that the coating on the IPC's is an acrylic formulation, as it is hard and fairly shiny. My looking around has revealed about 5 different formulas from various vendors and this is what has complicated my choice of removers.

I do not want to use mechanical/abrasive removal techniques as these boards have some very thin traces and do have some surface mount components. The rear most board has a lot of SMT stuff actually, mostly chip resistors and capacitors, and some PLCC's and a couple of SOIC chips. I can desolder/reflow joints even with the coating intact (it seems to be a reworkable type coating) as it just melts off. This gets messy though, as it burns and leaves discolorations and "stickiness" on the surface of the PCB, so removal really needs to be done for decent results.

Brian -

Thanks you for that additional info. When you get anything further, please pass it along. The more minds we have working at this issue, the more likely it will be solved. I have been concentrating more on the issue with "00/error" as it is more frequent, and has impacted a lot more owners than the full blackout issue. I currently have 4 IPCs with the 00/error failure mode sitting, and one with full blackout. I am hoping to get them all up and working again.

KDirk

[Guest Mc_Reatta]

Brian, could cross your diode 1N761-1 with an NTE 5009A which according to the data sheet is a 1/2 watt 4.7 volt zener diode which makes sense in a logic circuit.

http://www.nteinc.com/specs/5000to5099/pdf/nte5000a.pdf

the 349-1 crosses to an NTE128 (NPN) Silicon Transistor Audio Output, Video, Driver. The case is the wrong style but may still fit but won't shed heat like the IEM. Can probably cross reference these specs to another brand that you can get in the proper case.

http://www.nteinc.com/specs/100to199/pdf/nte128.pdf

The 266-1 crosses to an NTE 177 General Purpose Silicone Rectifier (diode).

In your photo these look like 3 pin devices not 2 pin diodes. Are these devices diodes or transistors? I'll keep searching this one.

Good source for parts is DigiKey DigiKey Corp. | Electronic Components Distributor | United States Home Page

[bwarren]

Mc, the 761-1 diode does not resemble any traditional zener that I have seen. Working from experience, it must be a power or schottky rectifier. Unless the designers went totally against all traditional standards of design, the 266-1 has to be some form of transistor and not a diode. As time permits, I plan on contacting the Motorola phone number for more info. I have no real clue on the 349-1 transistors. Unfortunately, transistors come in as many flavors and spec as......... OK I can't think of a good analogy right now, but there are likely several hundred (thousand).

[KDirk]

Mc,

Be careful of the NTE (or old ECG/SK series) cross ref manuals. When dealing with parts that have been re-stamped with house numbers, often times a matching number will be found, but will have no bearing on the part you are researching. The numbers listed by NTE and the like as OEM part #'s are generally industry standard parts numbers. For most transistors of US provenance, they have prefixes like 2N and TIP, Japanese designed/made parts usually used 2SA, 2SB, 2SC, 2SD and so forth. I believe this standardized numbering system was developed jointly by the EIA in the US, and the JEIA in Japan many years ago.

Anyway, when dealing with house or custom part numbers, the cross ref's miss most of these. Some will be listed with a company name next to the part number; looking at my old NTE book many references are made to Wurlitzer, Sears, Montgomery Ward and other historical major suppliers of consumer electronics who had house numbers on many of the component parts used in their products. I see no references to Delco or GM in a quick flip through, so the part numbers used on the devices Brian pulled out of his cluster are probably not in the NTE database at all since they were arbitrarily assigned under direction from GM/Delco.

This is what happens when custom numbering systems are used, numbers often duplicate one on an unrelated device from another manufacturer since there is no central authority checking to avoid reuse of existing numbers. This leads to confusion and difficulty in servicing things down the road.

Just another big pain to deal with, as if there weren't enough already.

KDirk

[Guest JohnW]

The warmer the weather gets, the longer the cluster works. Maybe I should move to the southwest or Florida.

[padgett]

My CD player only works over 80F.

[KDirk]

Funny thing is that this (working in warmer weather) seems to support cold/fractured solder joints. When they get warm, they expand, closing the circuit. When thermally cold, the opposite happens - contraction of the faulty joint, and voila - open circuit. Have had this problem (and corrected many of the same) with all manner of electronic devices - computers, TV's, stereos, and vehicle modules among others.

I was looking closely at one of my failed IPC's (error/00 fault) the other night and traced the circuit path of where the ALDL line enters the cluster all the way to the microprocessor. Can't be certain yet, but I am pretty sure there is at least one fractured joint (under the conformal coating) where the line passes through what looks to be a pull-up resistor just ahead of the input pin on the CPU.

I need to order some conformal coating remover, then I can try to re-solder the questionable points on the board and retest.

KDirk

[Guest joereatta]

My instrument cluster went dark this past winter. (I don't have a garage so it is left out in the cold of the Pa. Pocono mountains) It was fine until I went to check a fuse for the cruise control. I also found that the engine began to run rough at the exact same time. In searching old posts, there was a woman who described the exact same circumstances for her instrument cluster going out (fixing a cruise control fuse).

Any ideas/suggestions as to cause/cure? BCM gone south? Relation of rough running engine? Coincidence of 2 clusters going bad when checking cruise control fuse? Any help is appreciated. joe

[Roadster90]

Any comments to Joe's post please?

[Guest mongeonman]

The ipc and climate controle freak out at the same time intermitently and the fan goes to high speed and hot air come out of the defrost grill.????To fix it i turn the engine off and restart it.What could be the problem.thanks