[1979 Eldorado]

I read Dave's post #17 one more time "Two of my cars have the later version ECM(1225120) and one has the original version (1224760) .... Also the PROM chips are different from the E/K body to the C/D body cars. They are identified by two letters with KJ being one of the later versions used on the E/K cars."

I was thinking he had written E/K should have KJ but that code is simply "..one of the later versions.." Hopefully your "KC" is valid for E/K then. I think there are 2 very good signs, despite a lot of numbers missing from that list, that being you do have a service PN ECM which is one Dave identified and second the PROMs are sequential 7067KC and 7068KC. So at least we don't have any suspicious signs of anything obvious. I also like that your 2-digit code begins with "K" because I could see GM keeping the first letter common and "K" is the car letter designation for Seville.

I wonder how many total numbers they had and I wonder why they published only a partial list of even the OE ECM numbers? ... crazy.

 

[81 BENY Biarritz]

Hi Peter and Scott, if you really want to get into the PROM versions, have a look at this link from a previous Forum post. It does list the KC version on the second page of the fourth file but you have to figure out which model it applies to as the column headings do not line up. I think I figured it out once before so it should confirm whether this PROM version matches the original production code (the last six digits of the VIN #).
Maybe Peter you have already found this in your previous searches?

cadillacV864 - Google Drive

drive.google.com

 

[81 BENY Biarritz]

Peter, yes the same ECM was used for the C/D bodies (Coupe de Ville/Fleetwood) and the E/K bodies (Eldorado/Seville) with the 368 engine. It is only the PROMs that were used that are different. As far as I know they were used to program various performance and fuel control parameters to suit the different weights and dynamics of the two different cars configurations.
One thing that I did see when searching for the 1225120 ECM is the wide range in prices. This is the same Cardone remanufactured unit that would normally be available through a Canadian supplier I have used before so the cost is in Canadian dollars (vs US dollars!!)

Search Auto Parts by Part Number

Search and buy auto parts from a huge inventory of over 3 million auto parts in Canada. Find reliable OEM alternatives from top brands at lowest prices.

partsavatar.ca

Only issue is that it is Out of Stock like it seems to be at all the typical large parts distributors. Maybe you could contact Cardone and see if they plan to continue to offer it in the future?

Dave

 

[Lucky Peter]

Hello Dave,
I've just had a look at the link that You've send me, and there is a lot of interesting information. I've found my ECM production number along with its PROMs codes and they're both the same as in mine ECM, so that's some good news. There is a one thing that I'm a little worried about - What could have caused the damage to the current ECM in my Seville? There wasn't any overheat signs or something like that, maybe there is a possibility that it just broke down by itself? Maybe a little later today I'll send a message to Cardone and if they are not going to restock these ECMs, I'll just need to buy the one from AutoECMs, because that's the only option for now. You've also mentioned about the last six digits of the VIN number, but I don't really understand where to look. In this file I can see a six-digits numbers, even with the letter K, so i suppose it is what I need to look at, but I just can't figure it out. My last six VIN numbers are '700876'. I'm attaching the adequate part with my numbers highlited below. Could You help me figure it out?

Edit:
I've just opened another file called 'Serviceman Bulletin' and there was a clear explanation of these changes. It turns out that the mounting of the second generation ECM's began on cars with the last six VIN numbers '690461'. Since my Cadillac's last six VIN numbers are '700876' it means that it has a correct (second-gen) ECM in it.

- Peter​

 

[JayArr]

Hey Peter

This has been in the back of my mind for a while. You have solenoids that click when they aren't supposed to and you have injectors that don't come on when they should.

Is it possible that the wiring has been swapped? That the wires for your injectors have been accidentally connected to the solenoids and the solenoid wires are accidentally connected to the injectors?

That could conceivably result in solenoids that click when the key is turned and injectors that don't ever come on. This would result in a no start condition.

 

[1979 Eldorado]

Jay,
Dave posted the wire colors and junction numbers under another topic and I created a chart out of it that identifies the wire colors by deactivation solenoid. Possibly a quick check would be to check all of the wire colors by position. Here's the table I made:

 

[Lucky Peter]

Hi,
Jay, this is an interesting idea, but I doubt that someone swapped wiring, because recently I was testing active cylinders solenoids circuits, and all the wires were going in the right places as shown in one of the wiring diagrams sent by Dave. Some time ago I've also checked both injectors wires, and they're going in the ECM and in the Fusebox panel as shown in the wiring diagram too. Also I've decided to check the cylinders deactivation solenoids one more time, just to be sure they're alright. I've used this very nice table made by Scott, and all the wires in all four connectors are in the right place, and they are going to the right pins of the MDA connector, so I don't suspect any problems with them. The only thing that bothers me is the color of these wires. Some of them are faded and dark, so it is hard to name it, for example tan wire in my opinion isn't tan anymore, but it is still going to the right MDA pin, same as the all others. I also want to remind You that this car was running fine before, but then after a couple of days, when I started it up there was rough idle, then eventually leading to stalling and no start, so that's why I don't think that the problem is with swapped wiring, because it wouldn't be running from the start. Latest tests I've made using multiple charts sent by Dave are pointing out to a either bad ECM (which is more likely according to these charts) or PROMs, and it would explain so many strange issues happening with this car. Only question that I've already asked before is what could happened to this ECM, since it don't have any signs of damage in it? It is sure an interesting case, but I hope that We're gonna figure it out sooner or later and make this Cadillac running again.

- Peter​

 

[1979 Eldorado]

Sometimes diagnostic instructions either mention "verify all electrical connections" before they get into the specific diagnostic tree. I wonder if oxidation on a terminal(s) could be an issue. Maybe you mentioned it but, if it's one injector not opening did you try tracing the wires to each junction; cleaning each connection? You could test from the injector to as close to the source of the supply and ground as possible but if you do I strongly recommend using a test-light as opposed to a multi-meter. The reason is, with any connection, even a poor connection, you will still measure voltage but a poor connection cannot support the load. So if you use a test light you can see if the light is bright or dim.

Does the ECM trigger a relay for the injectors? How does this system know when to fire the injectors? In the older system we have a speed sensor with magnetic reed sensors on the distributor.

Sometimes old electronic components just go bad over time for no particular reason other than age but because you don't see anything obvious it seems like some of the basics should be double checked. My thought being a bad connection, in the wrong place, may test like a bad ECM. I've seen a few times where a multi-meter makes a problem more difficult to find due to lack of load during the test.

 

[Lucky Peter]

Hello Scott,
Lately while I was doing some tests, I've of course looked at every connector that I've unplugged and I haven't found any signs of oxidation in any of the terminals of many different connectors, including the ECM ones, MDA ones including active cylinders solenoids ones too, also injectors and many more.

One more thing according the injectors - there isn't a problem with one of them not opening. As I've said before, there are indeed opening and injecting fuel, but definitely too much, so it just stalls. After latest researches We know that the two of active cylinders solenoids are energized all the time just after turning the key, so this car is running only on six cylinders, which also increases its rough idle state. When I disconnect active cylinders solenoids it won't stall anymore, but it is still barely running, of course still very, very rich, but at least it isn't stalling.

Another thing worth mentioning one more time is the fact that this car is running in 'Hard back-up' mode, so there must be some kind of problem that's engaging this mode always after turning the key to ON position, You don't have to start the engine and You see this mode's symptoms: coolant temperature indicator won't light up, MPG display doesn't display anything and the diagnostic mode is unavailable. For me it just looks like some kind of internal damage in the ECM. Some messages above, Dave has sent a Google Drive link with multiple useful V8-6-4 information posted by Gdouaire, and in a one text file there is an explanation of this mode, which differs from standard fail-soft mode that is triggered by "smaller" problems. I'll attach a photo with appropriate part of it below.

There are many issues, but there is one common point that connects them - the ECM, which may be sending wrong signals that are messing everything up. Still there is a possibility that there may be a problem with PROMs, because in a file attached below there is a mention that 'hard back-up' mode can be a symptom of poorly seated PROMs, but maybe it isn't problem with PROMs directly. Is there a possibility that the ECM gone bad, and it just can't "read" all the information from PROMs? Maybe there are some other internal ECM problems that can cause this mode to activate? Tell me what do You all think about it, and of course don't stop asking and giving me some new ideas.

- Peter​

 

[1979 Eldorado]

I don't have experience on the system which came after 1979 but I do know that people often jump to the conclusion it must be the fault of the thing which is "different" on their vehicle. I'm not saying that you are doing that but we still need to verify the basics. So normally too rich could be an injector failing to seal when closed, a bad temperature sensor telling the car the temp is colder than actual, a bad MAP reading. Again, I don't know your car well, but are any of the inputs going to the ECM or any output device, like the injector not closing, a possibility? As well what about an EGR valve stuck open? All of the things I mentioned could cause an issue... I just thought maybe those things should be mentioned. The topic has been over some amount of time so maybe you mentioned checking those items. It seems like possibly the arrow is being pointed at the ECM because symptoms/checks have lead that way... It makes sense, but I still wonder if it could be something else.

 

[81 BENY Biarritz]

Hi Peter, you likely have already checked or maybe have noticed this but one very typical thing that this engine has had issues with is the coolant temperature sensor wiring. I had read of numerous others posting on this Forum and others that had cars that were running rich due to the coolant sensor wiring. When I checked my Eldo and then the Seville after I got it they both had faulty insulation on the wiring just behind the sensor until it goes into the protective plastic shealthing further back. My Fleetwood is still sending an intermitent fault code for it as I haven't had a chance to correct it yet. In the cases that I have read about the deteriorated wiring it send a false very cold coolant reading all the time to the ECM which in turn makes it think the engine needs a much richer mixture resulting in heavy black smoke from the tailpipe when the car is running. Have a close look at your wiring and connector to the coolant sensor to see if it has already been replaced or might be in questionable condition still. Not sure why this happens but the original insulation where it is exposed just crumbles especially after this length of time. Dave

 

[JayArr]

Is there a list of all of the conditions that would cause a vehicle to enter the "Hard back up" condition anywhere?

If we assume that the Proms are OK and seated correctly then what other condition could cause this mode to occur?

 

[Lucky Peter]

Hi,
Scott, You're absolutely right that there are multiple things that can make a rich running condition happen in normal circumstances, but for me this case looks a little more complicated than a normal rich running condition. I'm just not sure if I'm describing my thoughts clearly enough.

For example, let's say that the MAP sensor gone bad, so I think in this case it should trigger an error code, eventually car should be running in fail-soft mode, which differs from the mode that I've been experiencing. Now the main problem isn't just rich running condition anymore, the new problem is why the car is always in "Hard back-up" mode, as I've said before, You don't even need to start the engine to see this mode symptoms present, so in my opinion there must be something that is engaging this mode either after turning the key or even before You turn it, or maybe some component broke (maybe some kind of short?) the ECM, and that's why this mode is always active? I think that this scenario would make some sense, but what do You think?

Now I must admit that I've screwed up... Dave, You've just mentioned about these coolant temperature sensors wirings going bad, and I just reminded myself that maybe a year ago I've read some threads about this problems and gathered some information. I've forgotten about the coolant temperature wiring on my car! When rich running condition first appeared I was searching through all the internet, and I've found a thread, which described that a bad wiring of this sensor can cause a rich running condition. Some days later I found out, that in my Seville these wires were in fatal condition, because the insulation was falling apart just after touching it. Then of course I've insulated these wires, so now they're alright.... But maybe not everything is right??? Now I think that I have some theory about what could've happened in this Cadillac (I may be wrong of course).

My new theory:
There is a big possibility that after all these years, when the wiring of coolant temperature sensor started to deteriorating, it could cause these two wires to make a short circuit! If this sensor doesn't have any protection when going into the ECM, then it could cause a internal failure in the ECM, so that's why it is always running in 'hard back-up' mode.

Now after reading this theory several times, I think that if there was a short, wouldn't the coolant temperature sensor be damaged too? Another thing is with a short, wouldn't there be any signs inside of the ECM after this kind of failure??

I've decided to check voltage on both coolant temperature sensor and manifold temperature one. While I was doing this test, ambient temperature was somewhere around 18° C (65° F). I've checked voltage in the red ECM connector, when it was plugged in the ECM. I've also tested it on multiple grounds, one main of the ECM (bolted to dash) and on two pins in the same connector, and there were no changes, so ground is not any problem. The coolant temperature sensor voltage was around 4,1 V, and a manifold temperature one was 4,8 V. I'm not sure, but when I was testing this a little earlier (maybe an hour earlier, so no big temperature difference) , there was somehow more volts, definitely more that 5 (maybe even 6,7 V? I don't remember exactly), which I think is too high with that kind of temperature. If I'm not wrong, if the temperature is lower, the voltage should be higher? I think that I'll test these sensors voltages again tomorrow morning when temperature will be a little bit different to see what it will be showing. Do anyone know, what is the right range of a voltage on these particular sensors at a different temperatures? I've read somewhere that is should not exceed 5 volts on cold engine, is that correct?

Jay, You asked straight to the point question... Unfortunately I haven't found that kind of list anywhere, so for now it is a mystery... But maybe not for a long time?

So that's all of my thoughts for now. What do You all think about it? Is there a possibility of my new theory being true or judging by the sensors readings it is a dead end? Sorry by inconvenience caused by my memory, I should've said about this coolant sensor wiring at the start. Have a nice Thursday.

- Peter​

 

[1979 Eldorado]

"Jay, You asked straight to the point question... Unfortunately I haven't found that kind of list anywhere, so for now it is a mystery... But maybe not for a long time? "

I gave you my short list of common things which would cause the issue; one of them I mentioned was even bad coolant sensor, one was MAP, one was injector not closing, and one was EGR stuck open I didn't add it but a large vacuum leak could also fool the MAP.

It sounds like the coolant sensor must be a two wire sensor. With it unplugged can you read the resistance and can you get an accurate temperature reading? It's either PTC type, like for the older system, or it's NTC type like a lot of the new sensors. If you can find a day when it's exactly 25C that's the standard reference temperature for NTC. With NTC, if you test at a colder temp the resistance will be higher and warmer it will be lower. Those NTC sensors have huge resistance range so even a small temperature measurement could lead to an incorrect assessment but the good news is it takes a pretty big difference in resistance to make an actual difference. So if you provide some reference points I could assess what it likely is.

If the temp sensor didn't "pop" like a fuse then I suspect it didn't damage the ECM. What do the temperature sensors and connectors look like?

I do understand that some of the assumptions are based on this "Hard Back-up" but I have no idea if something like confusing information going to the ECM could also cause this. I've never had a 1981 so I'm trying to help with general knowledge.

 

[Lucky Peter]

Hello Scott,
Yes, You gave me a very helpful list with many different things that would cause rich running condition, but as replying to Jay, I was thinking about a list that would contain all the things that can make this car run in 'hard back-up' mode, not the rich running condition. But I think that You're right, and I've just gone too far with diagnosing this strange ECM mode, so as You've said - let's go check the basics.

The coolant temperature sensor in my car probably wasn't ever changed, because there are still some traces of original blue paint on it. It is a TX1 type, so You can find it on the Internet and see how it looks like (I'll attach a photo of it below). I've got a new one laying around, but I've never had a chance to replace it, because there was always something else to do (and I didn't even knew if the original one was broken). The connector doesn't look too bad, and the wires were insulated some time ago. Same thing goes with the intake manifold temperature sensor, both wires and connectors are looking fine. There was in fact one small place where they are insulated by me, because they're started to crumble in that one place.

Today I've managed to check these both sensors, when the ambient temperature was somewhere around 12° C (55° F). First I've checked these both sensors' voltage from the the right pins in of the red ECM connector, and then I've unplugged both sensors and checked their resistance. Here are the results of this test:

Name	Voltage	Resistance
Coolant Temperature Sensor	4,33 Volts	4,85 kΩ
Intake Manifold Temp. Sensor	4,95 Volts	16,68 kΩ

I don't know what is the right range of voltage and resistance for these sensors, but for me resistance in the intake manifold one looks somehow odd. Maybe it is causing some kind of troubles? I think that the manifold sensor that is installed in my car must be some old type one, because it has a different connector shape that the ones available to buy today. That isn't a big problem, because there are multiple sensors with connectors and wires included, and because of its shape I think that it will be better to change everything. Now let's try to check if this measured values of voltage and resistance are in the right range, or maybe there's something wrong with them? Maybe Scott is right, and some confusing information going to the ECM can trigger a 'hard back-up' mode? I think that with all Your help We'll be able to figure it out.

- Peter​

 

[1979 Eldorado]

Piotr,
The coolant has a big thermal mass so it's possible the temperature change could lag an air temp change but that said the engine block itself will also slow the change because, even though the air sensor is in the intake, it's surrounded by metal. If everything sat over night it does seem like they should read close to one another. I am assuming that they are the same PN sensor though; is that true?

The most accurate way to check the exact temperature would be to use one of the quick-read meat thermometers inserted in the coolant. You would first measure your sensor resistance and then you would remove the sensor and place the quick-read thermometer into the coolant. Of course only do this when it safe to do so; meaning when the system has sat over night and is not pressurized. You could alternatively use a thermocouple or and infrared pyrometer to measure the surface temperture. I just really want to be certain that the two readings are not due to an actual difference. It seems lke if the car sat over night though, then yes, one of them is bad. The slope of the curve can be drastic if it's the NTC type though.

If it is NTC type and assuming that the actual temp they are trying to read is, in fact, the same then the higher reading sensor "thinks" the temperature is colder than the lower reading sensor. If they are in fact the same PN can you swap the positions? If you do that and then wait long enough for the coolant sensor to equalize we can test the mat the same temperature. I would suggest testing them in the coolant because I believe that's where the temperature will stay stable for the longest time with the car off and not having been running over night.

I'm not familiar with that specific connector but I started making sensors for Holley Sniper and the Delphi, now Aptiv, 2-pin connectors are clinched in the same manner.

 

[Lucky Peter]

Hi,
Scott, this is an interesting idea, but I can't swap these sensors' connectors, because they differ from each other. The coolant temperature sensor has this strange oval one connector, while the intake manifold one has a standard looking connector. I'll attach a photos of both connectors, so You can see how they look like.

It would be good to know what is the right voltage and resistance of these sensors at a different temperatures, and some hours ago, I've finally received service books for my Seville, so I think that maybe there will be some information about resistance or voltage of these sensors, which would really help to check them out. Maybe Dave or Jay have some knowledge about these particular sensors in this cars and what is its right reading. For now I'm going to check out these new service books and see if I can find anything about these particular sensors written in them.

- Peter​

 

[1979 Eldorado]

The harness connector in the photo ending in 160629 is the interface for the newer "Delphi" (Aptiv) sensors that I'm making for Sniper. I would almost be shocked if that connector was period-correct for 1981. Anyway there are two scenarios and I think we need to figure out which one it is...

Scenario 1: They should be the same sensor and someone replaced the one (probably the one which currently has the Delphi connector) with what they could find. As a word of caution RockAuto lists a temp sensor for my 1979 which is as useful as a ham sandwich. It's actually worse. Not only are the interface and the resistance curve totally wrong, the thread size is wrong. So if an owner modified their car, assuming the curve was correct, they would have a damaged/altered intake. Once they finally found the correct sensor they would either need to fix the intake, change the intake or use a bushing. A bushing is a terrible option because it can create an air pocket and localized boiling (bad for a lot of reasons).

Scenario 2: The connection is different because the two sensors should not be the same PN and the plug type was used for error proofing. That could be the case even if the Delphi interface was not the correct "other" interface.

Dave has some great parts sheets. Maybe he can see or maybe either Jay or Dave can look at the interface type on their cars. If the interface, on their cars, is the same for both sensors then I think we need to be suspicious about what you see in your car.

One more question for now... Which sensor is the coolant sensor? I'm asking because very likely the ECM adjustments are more significant based on overall engine temp than air temp. If they should be the same PN, Dave or Jay to verify, then you should try placing the "funky" looking connector in the coolant position. It would also be very helpful if someone could test a known good sensor and even better if the test was at 25C.

Regarding the voltage/resistance readings. The way I suspect it works is you have a variable resistor in the sensor and the ECM has a reference resistor. The ECM designers designed a voltage divider circuit. That type of circuit allows the ECM to read resistance because the voltage changes based on the variable resistance.

Back to the Hard-back-up mode. The cars just one generation before had no computer reporting or error sensing at all. You car must have the very beginning of this technology. My thought is, if the car sensed something which seemed impossible to decipher, it may just default to the hard back-up. Possibly something as simple as a temp sensor being way out of range would be enough to do that??

 

[Lucky Peter]

Scott,
After reading Your message I must say that it is very interesting. I've just had a quick look at the manifold temperature sensor, and I don't see any signs of modifications of the intake manifold, but I see that someone cut the wires and installed this new connector, so it isn't original one for sure. It turns out that my memory is struggling again, because someone who was changing the connector, insulated wires using black tape, but I thought that was me some time ago. So now We know that definitely this connector isn't original, but maybe as You've said, if Dave or Jay see these messages, maybe they will check out this sensors in their cars and send a photo to compare them.

As far as I remember, these two sensors with its connectors are available at RockAuto. I've bought this TX1 type coolant sensor, but I didn't see any available connectors for this type. There are multiple newer "Delphi" type sensors, and they come with a new connectors. Is there a possibility to change the old one on the coolant sensors to the new "Delphi" style like one in the intake manifold? By the way the coolant temperature sensor is the one with "funky" connector.

Regarding measuring a good sensor resistance - as I've said before, I've got a new coolant temperature sensor, so I'll check its resistance in a moment. At ambient temperature around 22° C (71° F) the sensor resistance was somewhere around 3,7kΩ, when it was held in hand for a while, the resistance was slowly going down to around 2,4 kΩ. Unfortunately this test isn't the best I've made, because unintentionally from the way I was holding it, the heat from hand changed the temperature. Still, after this test, if the manifold sensor works the same way as this is working, I don't think that resistance exceeding 16kΩ is the right one. I haven't look at my books yet, so for now I still don't know if there is some mention about these sensors resistance.

Also I think that You may be right about car getting in 'hard back-up' mode because of some very out of the range sensor signal, because it would make some sense. It would be interesting if it was all because of the faulty temperature sensor. I think that's it for today, I'll check my books in a moment, and write to You all tomorrow. Have a nice Friday, and stay safe.

- Peter​

 

[1979 Eldorado]

Piotr,
If we can identify the sensing element then I have the ability to clinch that Delphi connector in. I never bothered making 3/8inch NPT because the standard Sniper part is offered in that but I currently have 1/4 NPT and 1/2 NPT. I could also have 3/8 NPT made. Visually, that new "Standard" brand part you have appears to be 1/2NPT. So if it is you could even carefully extract the sensing element and I could probably move it to a Delphi connector since your harness was modified.

Before doing that though we need to verify what the car should be seeing. As you've already noticed, accidentally holding it by hand, they change by a lot with small temperature increments. That means guessing at the actual temperature can give us a wildly different conclusion. I have an enclosure where I let all the parts sit together over night. I measure them in the morning, carefully picking them up by only the connector, and I have the Ohm meter turned on and I measure quickly. The enclosure has a non reflective surface, an uncoated paper plate, and I take a few Infrared pyrometer readings of the surface of the plate. One morning I was lucky enough to read almost exactly 25C for one part I was working on.

After doing the ambient temp, the most critical is at thermostat temp, so I heat everything I want to measure in a toaster oven. I let it equalize for about 4 hours. I then quickly measure and again grabbing only the plastic connector. I typically measure at 65C because I can still handle the part, with a nitrile glove, at that temp.

I'm going into detail because we don't know how well the aftermarket did with a part like this and second because you could spend next to forever trying to find another sensor with exactly the same curve but with a Delphi connector. Instead, if we can properly identify what needs to go inside, I can probably find a replacement element and then it's the same work as I've done for the Sniper sensors.

NPT is tapered thread but if you roughly measure the thread outside diameter we can tell if your car takes 1/2NPT. Possibly Jay or Dave knows already. They are named for the pipe ID they support so the OD is larger than the name implies.

Here's one of mine: