Cranks, no start - theft deterrent issue...96 Seville

[rockfangd]

I know this procedure works on saturns because I have used it before, but the chip in the key is in the head rather than having a pellet.

You watch the security light and after 10 minutes it goes out. If you mess it up you must start all over

[fmw63]

$113.00 per hour for them to basically throw parts at it is obscene.

Exactly. I can do that myself. But after I had gotten back up there to pick up to the car, service advisor told me had left me message that the service manager would cover the previous costs and continue the diagnostics, to find out what was really going on, but by then I'd had enough of them.

[fmw63]

It just hit me that all this car may need is the "theft learn procedure." They say you can use something called a "T50" or "T60." This is probably a limited OBD scanner for locksmiths and such. Without such a tool, the process is (1997 FSM, p. 8C1-4) is to make sure that the battery is fully charged so it doesn't drop below 10.4 Volts if the key is left on for a half our, then turn the key ON for ten minutes and turn it OFF; do this three times. Then, clear the codes and see if the password problem is gone. This may also program the IPC to accept whatever key resistance is being used, if it is one of the correct ones. It's OK to use a trickle charger to keep the battery up while doing this.

the theft learn procedure has been tried at least 4, maybe 5 times. didn't work.

[Cadillac Jim]

OK, let's start over.

B1552 Keep Alive Memory (KAM) Error

This code is thrown whenever you disconnect the battery. If that doesn't explain this code, you may have low battery voltage or a problem with the wiring, or a cracked fuse.

P1631 Theft Deterrent Start Enable Signal Not Correct

This code (1997 FSM, page 6-598) is also called "Theft Deterrent Password Incorrect." The PCM throws it whenever there is a password mismatch between the IPC and the PCM. The theft learn procedure will remove this fault and the code will go to HISTORY. The password is the "fuel continue" password.

P1632 Theft Deterrent Fuel Disable Signal Received

This code (1997 FSM, page 6-599) is set when the PCM recieves Fuel Disable or Fuel Undecided passwords from the Vehicle Theft Deterrent (VTD) controller. It's intended as a service tool that shows that the VTD has cut off the gas "due to a worn or incorrect ignition key." There is a Diagnostic Aids section with this DTC writeup:

• If DTC P1632 is set the IPC has received the incorrect resistance value or an unrecognizable resistance value from the Ignition Key.
• Check the Ignition Key to make sure that it is the correct key for that vehicle and it is clean and dry.
• Wait 10 minutes between attempts to clean and retry key.
• This DTC may also set if a theft was attempted.

B2750 PASSKey Data Communication Failure

This code (1997 FSM, page 8C1-46) is thrown when communication between the IPC and the PCM doesn't work. The IPC displays the message "THEFT SYSTEM CAR MAY NOT RESTART." The description says that this code is thrown when the key pellet resistance measurement isn't in range. The Troubleshooting Hints are:

1. If U1016 is also set, diagnose this first.
2. If DTC 1604 is also set, diagnose this first.
3. This condition may be caused by a password miismatch between the IPC and the PCM. Refer to IPC Programming Following Replacement in this section and perform the procedure.

The section "IPC Programming Following Replacement" is in the 1997 FSM pages 8C1-3 and 8C1-4. Make sure that the battery is fully charged so that it remains above 10.5 Volts when the key is left on for a half hour, or put a trickle charger on it. Do NOT use a fast charger while doing the IPC learn procedure. Without the special scanning tools T50 or T60, you do this three times: you turn the key ON for ten minutes, then OFF. I don't think you want to leave it off for more than a second or two but it may not be critical so long as it isn't off for ten minutes or some such. When you are done, turn the key on again and see if B2750 has gone to HISTORY.

B1026 Driver Deployment Loop Open

For some reason the 1997 FSM does not have a writeup on DTC B1026 but it does for B1117, "Passenger Deployment Loop Open" on pages 9J-30 through 9F-33. The SDM (Sensing and Diagnostic Module, what they call the air bag module) puts a little current through the air bag circuit and if more than 0.46 Volts is seen across the terminals for the air bag the code is thrown. The driver air bag is on terminals 9 (low side, DK GRN wire) and 10 (high side, WHT wire) of the SDM connector. There are shorting bars on both connectors for the SIR Coil Assembly (the helical ribbon cable that carries wires to and from the rotating steering wheel) that can be used with a special tool for this purpose to help diagnose this problem. Resistance from the SDM connector to the air bag connector in both the WHT and DK GRN wires must be lower than 0.5 Ohms.

Conclusion: It looks very much like the circuit from the IPC to the key has a break in it. If you can pull off the connector on the IPC and measure the resistance between pins A10 and B10, which connect to the WHT/BLK and PPL/WHT wires, it should measure the key pellet resistance. If the circuit is open, you need to find out where the circuit is broken. If it does match the key pellet resistance, the IPC is bad.

Problem: In your original post you say that all the codes are HISTORY. That shouldn't prevent the car from starting. However, when you turn the key off, all the codes go to HISTORY and only become CURRENT when the fault is detected again. So, try to start the car and then check the codes without turning the key off, and mark which ones are CURRENT, if any. If they are never CURRENT, reset them and see if they come back immediately. If they don't then we start looking for other problems that prevent starting without throwing a code, like a bad fuel pump.

[fmw63]

Conclusion: It looks very much like the circuit from the IPC to the key has a break in it. If you can pull off the connector on the IPC and measure the resistance between pins A10 and B10, which connect to the WHT/BLK and PPL/WHT wires, it should measure the key pellet resistance. If the circuit is open, you need to find out where the circuit is broken. If it does match the key pellet resistance, the IPC is bad.

I'll try this as soon as I can get a chance

Problem: In your original post you say that all the codes are HISTORY. That shouldn't prevent the car from starting. However, when you turn the key off, all the codes go to HISTORY and only become CURRENT when the fault is detected again. So, try to start the car and then check the codes without turning the key off, and mark which ones are CURRENT, if any. If they are never CURRENT, reset them and see if they come back immediately. If they don't then we start looking for other problems that prevent starting without throwing a code, like a bad fuel pump.

yes. at the time of first post I didn't see any current codes. it's possibly, i suppose, that i missed seeing them, but i don't think so.

currently the codes are PZ1558, IP2750.

[Cadillac Jim]

B1558 BCM EPROM Checksum Error
B2750 PASSKey Data Communication Failure

If the B1558 is CURRENT or comes back repeatedly, it probably means that the IPC is bad. Are there any other codes? A low battery or memory reset code may simply mean that the B1558 was thrown as the battery was disconnected or connected, or the IPC unplugged or plugged back in if it was done with the battery connected. I would still check the key resistor from the IPC connector because that will rule in, or rule out, the EITLR, continuity through the key's pellet resistor.

[KHE]

Exactly. I can do that myself. But after I had gotten back up there to pick up to the car, service advisor told me had left me message that the service manager would cover the previous costs and continue the diagnostics, to find out what was really going on, but by then I'd had enough of them.

That sounds like a good deal - I would have let them continue the diagnostics with a limit on the time. Now you're starting over.

[fmw63]

B1558 BCM EPROM Checksum Error

B2750 PASSKey Data Communication Failure

If the B1558 is CURRENT or comes back repeatedly, it probably means that the IPC is bad. Are there any other codes? A low battery or memory reset code may simply mean that the B1558 was thrown as the battery was disconnected or connected, or the IPC unplugged or plugged back in if it was done with the battery connected. I would still check the key resistor from the IPC connector because that will rule in, or rule out, the EITLR, continuity through the key's pellet resistor.

a bcm error could mean the ipc is bad?

no, no other codes right now...

[Cadillac Jim]

You're right, I was thinking IPM checksum error. Also, I did find B1558 in the 1997 FSM on pabe 8D2-22. The troubleshooting hints are grim: "If DTC B1558 is set, clear DTCs and cycle the ignition. If DTC B1558 is still current, replace the PZM."

[fmw63]

That's the BCM, right? Seen several on ebay. Think that would work?

[Cadillac Jim]

I wouldn't go to replace the BCM (PZM) until I verified that it was bad. The B1558 can be thrown by a scratchy connection while disconnecting or connecting the battery or possibly other things, which is why the FSM says to clear the code and ignore it if it isn't persistent. Try turning the key off and on and clearing the code first. If it keeps coming back after you clear the codes, you can get a PZM (BCM).

I don't see anything in the FSM about reprogramming the PZM, and it may be generic to the 1996-1997 or even a wider range of model years for Deville/Seville/Eldorado. But, I haven't changed one. Others will have information on compatibility of PZMs between cars and any programming requirements.

I will say that bad modules are rare. A module can be damaged by back-probing a connector at a terminal leading to a static-sensitive connection to a module. Really dumb things, like hooking up jumper cables backwards to a car with a dead battery, or using a boost fast charger on a battery while in the car and hooked up, can brick every module in the car along with the radio and nav system and OnStar.

[fmw63]

I wouldn't go to replace the BCM (PZM) until I verified that it was bad. The B1558 can be thrown by a scratchy connection while disconnecting or connecting the battery or possibly other things, which is why the FSM says to clear the code and ignore it if it isn't persistent. Try turning the key off and on and clearing the code first. If it keeps coming back after you clear the codes, you can get a PZM (BCM).

I don't see anything in the FSM about reprogramming the PZM, and it may be generic to the 1996-1997 or even a wider range of model years for Deville/Seville/Eldorado. But, I haven't changed one. Others will have information on compatibility of PZMs between cars and any programming requirements.

I will say that bad modules are rare. A module can be damaged by back-probing a connector at a terminal leading to a static-sensitive connection to a module. Really dumb things, like hooking up jumper cables backwards to a car with a dead battery, or using a boost fast charger on a battery while in the car and hooked up, can brick every module in the car along with the radio and nav system and OnStar.

Thanks for the help guys!

[Cadillac Jim]

You should really consider getting a factory shop manual (FSM) from eBay. I've been getting the procedures for the DTCs from my 1997 FSM. Since there was very little change between the 1996 and 1997 model years for Deville/Seville/Eldorado, that works. There were changes in the FSM format too, but not the cars. So, if you look, you can use either the 1996 or 1997 model year FSM for E/K platform.

I've got many years of experience with DIY with my own cars, and others here have a lot more experience than I do, in that they have had multiple Cadillacs or have worked on many Cadillacs besides their own, and some or techs or even business owners that deal with Cadillac maintenance and upgrades. My best contribution in this elite crowd is electrical systems because of my other background, and I know my way around the 1997 FSM and have the 2002-2011 FSMs on DVD. We are always here for you. But having your own FSM on hand is an enormous asset for the owner of an older car that needs do do work on his/her own car correctly in a timely fashion without relying on a dealer.

My personal belief is that a dealer is not necessarily the place to maintain an older car, because some dealers don't have techs familiar with older cars and may not have the manuals on hand anymore, but they still must charge the Goodwrench hourly rate when they assign someone to work on your car. Also, a lot of things don't need Goodwrench training, like brake jobs and a lot of undercar work, body and interior work, etc. So, a FSM and getting to know an ASE-certified mechanic that has a good OBD scan tool and that is comfortable with your Cadillac are your best friends with an older car.

[fmw63]

You should really consider getting a factory shop manual (FSM) from eBay. I

Oh, I agree. Just when I think about getting one, don't have the funds. They are a big help. Had the set before for the 93 Seville i had few years back.

[fmw63]

Ok. Here's the latest.

1. Pro Module removed.

2. Measured resistance at A10 & B10 - same as key - 7420 ohms

3. codes cleared, a few times.

4. IP2750 is current and won't go away. Bad IPC?

[Cadillac Jim]

It sounds like the IPC is programmed for a different resistance. The learning procedure may get it to recognize the existing key resistor, if you have a half hour or so to try it. BUT, the 1997 FSM page 8C1-3 has a this section, which it labels "[!] Important"

• All replacement IPC's received from an authorized service center are configured, so that the IPC can learn (store to memory) the vehicle's ignition key pellet resistance (key code) value (PASS-Key® II).
• The vehicle will not start until the key code value is learned by the IPC.
• The theft deterrent module re-learn procedure must be performed, refer to IPC Programming After Replacement in this section.

The "IPC Programming Following Replacement" immediately follows this section and gives two ways, one with a T50 or T60 scan tool and the other turning the key on for 10 minutes, then off three times.

The problem I am having, on careful reading, is that the IPC may not accept the learning procedure unless that function is enabled with a Tech II, and perhaps it can do it only once.

The 1997 FSM, page 8A-133-13, says in the ignition key description that there are 15 different resistor values. I have a table of these values somewhere but can't locate it, nor can I locate such a table in the FSM.

I do see something important. On page 9D-2 of the 1997 FSM, Figure 2, they show that the wire to the ignition cylinder for the pellet resistor must have a wire loop routed through the opening in the turn signal switch to allow it to flex when the key is turned. If this wire is dressed wrong when the turn signals are serviced, which pulls the wire pulled out of position when the turn signal switch is pulled out, it will probably break eventually. This is probably what happened to you.

I have one more question. The dash has a "Security" warning light. This light normally flashes when you turn off the key and open the door and keeps flashing until you close the door. If the resistor is wrong, the Security light should come on and stay on while you are trying to start the car.

I would think that if you have B2750 CURRENT and the Security light is ON, that you may be able to get the dealer to reprogram the IPC for the key resistor in a few minutes. If the Security light doesn't come on when you try to start the car and the bulb is good, then perhaps the IPC is bad.

Replacing the IPC is a big job and a dealer is required to program the new one. The IPM, the computer inside the IPC, is the master computer for the whole car and it has your VIN and all your options programmed into it, and the dealer needs your old one to program the replacement. So, you should be VERY sure that the IPC is bad before you replace it. So far, all we know for sure is that it isn't accepting your key resistance and the three-cycle half-hour process doesn't seem to work.

[Göran W]

Don´t know the tolerance but pellet no 13 is supposed to be 7,500 Ohm.

[Logan]

Pass Key table.

[Cadillac Jim]

Thanks, Logan, it was probably a post of yours that I recall, not my own computer. fmw63 is measuring 7420 Ohms directly from the connector to the IPC, and that is clearly a valid value for number 13. He has a B2750 CURRENT and a theft learning procedure doesn't work. I'm reluctant to declare the IPC bad because that's rare, and a difficult, expensive repair, and I don't have direct experience with bad IPCs. What do you think?

[fmw63]

Thanks, Logan, it was probably a post of yours that I recall, not my own computer. fmw63 is measuring 7420 Ohms directly from the connector to the IPC, and that is clearly a valid value for number 13. He has a B2750 CURRENT and a theft learning procedure doesn't work. I'm reluctant to declare the IPC bad because that's rare, and a difficult, expensive repair, and I don't have direct experience with bad IPCs. What do you think?

Any more ideas?

[Cadillac Jim]

We need to hear from someone with experience with the theft learning procedure, and has experience with fixing B2750, and preferably some experience with replacing IPC's and with the Tech II in addressing B2750. In particular, I have some questions:

• Is the ten minutes wait with the key on critical? What do we really need to set the oven timer at? Is 12 minutes a sure bet?
• Is the theft learn procedure locked out on IPCs once it's done on a reconditioned unit? Can it be done on any IPC if you get a new key and don't know the old pellet number? If not, how do you get the pellet number from the VIN?

Until I hear something definitive on these points I'm not giving up on the existing IPC.

[Cadillac Jim]

In the absence of anyone else stepping in, I will give an opinion here. The IPM, the module inside the IPC, is the main computer for the car, and encodes all the car's options and calibrations, the odometer, etc. and must be programmed for each individual car. If it can be made to accept a key, or if the key pellet that it will accept can be changed, a dealer with a knowledgeable technician and a Tech II can do it.

If the IPC must be changed, the procedure is to exchange it with a refurbished unit, but the new IPC is programmed using data from the existing IPC as part of that process. Because of anti-theft and security features, only a dealer can do this for you. You may be able to bring a new, refurbished, or used IPC to the dealer with your car and have them reprogram the new one and install it for you; I would check with the dealer about that before I did it.

[Logan]

I have zero 1996 or 1997 info. I have 1995 and earlier....and 1998 and latter. So any info I have would not apply.

That said, chances are any serious IPC programming would have to be performed at a Delco radio speedometer repair facility. They have a IPC terminal for doing such repairs.

A GM dealership or a GM Tech2 cannot change the mileage in a IPC cluster.

If a dealer could program the mileage...we would all be driving low mileage cream puffs.

[Cadillac Jim]

Yeah, in 1986 tried to buy my wife a car at at Chrysler dealer in Santa Ana, CA, a suburb East of L.A. All the cars in the used car lot had 31,000 to 34,000 miles on them. In these days of odometers, close inspection revealed scratches on the number of all of them. I used tire wear to gauge mileage, and was interested in one car with about 45,000 miles on it - but it didn't work out and we got a car elsewhere. The car market in L.A. back then was rapacious for American cars, which is, I believe, the reason that the imports dominated so heavily at the time.

I have read recently where some criminal elements have figured out how to change the mileage in IPC's of a variety of makes, and there were some arrests a few years ago in auction houses regarding large numbers of cars that were moved to different States where the VIN mileage tracking didn't exist, but were tripped up by the new States checking with the old title States as part of the registration process.

I drove a 1997 ETC for fifteen years and have the FSM including the transmission overhaul manual and the OnStar supplements, if you ever need something on the 1996-1997 model years. But the only 1996-1997 Cadillac that I ever worked on was my own, and it never gave much trouble.

But, the issue in question, why the 1996 Seville IPC won't take the three-ten-minute-reset, is beyond me, and I would defer to a dealer on that. Since 1996 was a long time ago, the tech would probably have to go to GM Service Information (SI) to find out, and a bonded tech would have access to locksmith information.

[fmw63]

Yeah, in 1986 tried to buy my wife a car at at Chrysler dealer in Santa Ana, CA, a suburb East of L.A. All the cars in the used car lot had 31,000 to 34,000 miles on them. In these days of odometers, close inspection revealed scratches on the number of all of them. I used tire wear to gauge mileage, and was interested in one car with about 45,000 miles on it - but it didn't work out and we got a car elsewhere. The car market in L.A. back then was rapacious for American cars, which is, I believe, the reason that the imports dominated so heavily at the time.

I have read recently where some criminal elements have figured out how to change the mileage in IPC's of a variety of makes, and there were some arrests a few years ago in auction houses regarding large numbers of cars that were moved to different States where the VIN mileage tracking didn't exist, but were tripped up by the new States checking with the old title States as part of the registration process.

I drove a 1997 ETC for fifteen years and have the FSM including the transmission overhaul manual and the OnStar supplements, if you ever need something on the 1996-1997 model years. But the only 1996-1997 Cadillac that I ever worked on was my own, and it never gave much trouble.

But, the issue in question, why the 1996 Seville IPC won't take the three-ten-minute-reset, is beyond me, and I would defer to a dealer on that. Since 1996 was a long time ago, the tech would probably have to go to GM Service Information (SI) to find out, and a bonded tech would have access to locksmith information.

In the absence of anyone else stepping in, I will give an opinion here. The IPM, the module inside the IPC, is the main computer for the car, and encodes all the car's options and calibrations, the odometer, etc. and must be programmed for each individual car. If it can be made to accept a key, or if the key pellet that it will accept can be changed, a dealer with a knowledgeable technician and a Tech II can do it.

If the IPC must be changed, the procedure is to exchange it with a refurbished unit, but the new IPC is programmed using data from the existing IPC as part of that process. Because of anti-theft and security features, only a dealer can do this for you. You may be able to bring a new, refurbished, or used IPC to the dealer with your car and have them reprogram the new one and install it for you; I would check with the dealer about that before I did it.

Going to the local Caddy dealer Thursday...