PA Trouble Code 26- On Board Diagnostics
lportersr
05-10-2005, 12:48 PM
My 1991 PA is showing an intermittent code 26 on the OBD. The fault relates to "Quad Driver Error." I Found the attached article for all to read. I would welcome anyone elses experience with this before I start troubleshooting.
Research Article:
Code 26 ... Not Again!
by Les Bentley
Mr. Smith just drove in the lot with his '89 Buick Park Avenue. He said that the Service Engine Soon (SES) light is coming on. Not again! This is the third time in the last 10 days he has come in complaining about that darn light. And each time, the code stored has been 26.
You begin flashing out trouble codes. And yes, there it is -- a code 26, quad driver error. You have checked everything. You even installed a new injector and EGR valve, and the problem just won't go away. The complaint can never be duplicated and there is no other driveability problem except for the light coming on. Worse yet, the car belongs to the shop owner's father-in-law!
Driving the car around the block revealed nothing and the scan tool was no help either. So you try looking up any service bulletins that apply. No bulletins are found for 1989, but a 1988 bulletin describes a quad driver as a device or driver inside the computer that can monitor and control anywhere from one solenoid up to four different solenoids. This can be in one external device or four different ones. It also explains that the computer monitors for correct operation by monitoring the voltage at the computer connector for each particular device; case in point: ECM disconnected and ignition key on, engine off. If you check voltage at terminals BC2, BC3 and BD2, which operate the EGR solenoids, they should all show 12 volts. This would also hold true for any device operated by the ECM on the other three quad drivers.
After determining that all circuits show battery voltage, you reconnect the ECM and road test the car. Still, the light does not come on and again the scan tool shows nothing wrong. In desperation, you break down and call the customer and ask him under what driving conditions the SES light comes on. He explains that the light usually comes on after about 10 miles of driving and remains on until he slows way down or comes to a complete stop. You also ask if any other lights come on at this time. He answers no.
Great! You finally have the information to duplicate the complaint. While road testing the car you finally get the SES light to come on and to store a code 26. Taking a snap shot on the scan tool still doesn't show anything abnormal.
Reading the bulletin again, you find that when the ECM turns on a device, (grounds the circuit) the voltage should be near zero. If this is true, then monitoring the circuit with a voltmeter should duplicate what the computer is seeing when the code is stored. The bulletin also indicates that Quad Driver 2 will not set a code 26. If this is true, test those circuits last. This leaves you Quad Driver 1, which controls the canister purge solenoid, the SES light and low speed coolant fan; Quad Driver 3, which controls the hot light and TCC solenoid; and Quad Driver 4, which controls all three EGR solenoids.
While looking for technical service bulletins that pertain to code 26, you find one about poor connections at the computer. Good point. If the computer is supposed to be seeing 12 volts at a device that it has not grounded and if there is a poor connection at the computer, it could set the code. All connections are tight and clean. This must indicate that something is happening to the circuit once it has been activated for a while. Start with the quad driver that has the least functions - Quad Driver 3. "Hot" is not coming on when the SES is so that shouldn't be a problem, but what about the TCC?
With a voltmeter, backprobe terminal YC10 on the computer. Of course, at this point, the car is just idling and the voltmeter shows charging system voltage. A strange thing begins to happen during a test drive. When the computer first grounds the TCC solenoid, the voltage on the meter drops to 0.3 volts. But the longer the TCC is engaged, the higher the voltage gets until the SES light comes on, and 1.0 volt is showing on the meter. Is it a bad ECM or a shorted TCC solenoid? The solenoid tests fine, with no excessive current flow. This indicates that the driver inside the ECM breaks down over time and is unable to hold the TCC solenoid to ground. When the computer sees the voltage level on the circuit increase, it knows the TCC should be on, so it sets a code 26. Replacing the ECM fixed the problem.
Now, somebody is going to say that a TCC problem should have set a code 39. Not in this case. There was no open TCC circuit or inoperative circuit inside the ECM. It was just that the quad driver was unable to hold ground level and started releasing the ground -- not enough to disable the TCC, but enough to turn on the SES light and store that nasty code.
I hope this helps the next time you see a code 26. Remember to verify the customer's complaint and find out what is happening when the code sets. Then try to duplicate the complaint. Be sure to test the component when it is failing -- not when it's working properly. If you have any questions, just give me a call!
Les Bentley is ASE master and L1 certified. He is a GM specialist with 16 years experience.
Research Article:
Code 26 ... Not Again!
by Les Bentley
Mr. Smith just drove in the lot with his '89 Buick Park Avenue. He said that the Service Engine Soon (SES) light is coming on. Not again! This is the third time in the last 10 days he has come in complaining about that darn light. And each time, the code stored has been 26.
You begin flashing out trouble codes. And yes, there it is -- a code 26, quad driver error. You have checked everything. You even installed a new injector and EGR valve, and the problem just won't go away. The complaint can never be duplicated and there is no other driveability problem except for the light coming on. Worse yet, the car belongs to the shop owner's father-in-law!
Driving the car around the block revealed nothing and the scan tool was no help either. So you try looking up any service bulletins that apply. No bulletins are found for 1989, but a 1988 bulletin describes a quad driver as a device or driver inside the computer that can monitor and control anywhere from one solenoid up to four different solenoids. This can be in one external device or four different ones. It also explains that the computer monitors for correct operation by monitoring the voltage at the computer connector for each particular device; case in point: ECM disconnected and ignition key on, engine off. If you check voltage at terminals BC2, BC3 and BD2, which operate the EGR solenoids, they should all show 12 volts. This would also hold true for any device operated by the ECM on the other three quad drivers.
After determining that all circuits show battery voltage, you reconnect the ECM and road test the car. Still, the light does not come on and again the scan tool shows nothing wrong. In desperation, you break down and call the customer and ask him under what driving conditions the SES light comes on. He explains that the light usually comes on after about 10 miles of driving and remains on until he slows way down or comes to a complete stop. You also ask if any other lights come on at this time. He answers no.
Great! You finally have the information to duplicate the complaint. While road testing the car you finally get the SES light to come on and to store a code 26. Taking a snap shot on the scan tool still doesn't show anything abnormal.
Reading the bulletin again, you find that when the ECM turns on a device, (grounds the circuit) the voltage should be near zero. If this is true, then monitoring the circuit with a voltmeter should duplicate what the computer is seeing when the code is stored. The bulletin also indicates that Quad Driver 2 will not set a code 26. If this is true, test those circuits last. This leaves you Quad Driver 1, which controls the canister purge solenoid, the SES light and low speed coolant fan; Quad Driver 3, which controls the hot light and TCC solenoid; and Quad Driver 4, which controls all three EGR solenoids.
While looking for technical service bulletins that pertain to code 26, you find one about poor connections at the computer. Good point. If the computer is supposed to be seeing 12 volts at a device that it has not grounded and if there is a poor connection at the computer, it could set the code. All connections are tight and clean. This must indicate that something is happening to the circuit once it has been activated for a while. Start with the quad driver that has the least functions - Quad Driver 3. "Hot" is not coming on when the SES is so that shouldn't be a problem, but what about the TCC?
With a voltmeter, backprobe terminal YC10 on the computer. Of course, at this point, the car is just idling and the voltmeter shows charging system voltage. A strange thing begins to happen during a test drive. When the computer first grounds the TCC solenoid, the voltage on the meter drops to 0.3 volts. But the longer the TCC is engaged, the higher the voltage gets until the SES light comes on, and 1.0 volt is showing on the meter. Is it a bad ECM or a shorted TCC solenoid? The solenoid tests fine, with no excessive current flow. This indicates that the driver inside the ECM breaks down over time and is unable to hold the TCC solenoid to ground. When the computer sees the voltage level on the circuit increase, it knows the TCC should be on, so it sets a code 26. Replacing the ECM fixed the problem.
Now, somebody is going to say that a TCC problem should have set a code 39. Not in this case. There was no open TCC circuit or inoperative circuit inside the ECM. It was just that the quad driver was unable to hold ground level and started releasing the ground -- not enough to disable the TCC, but enough to turn on the SES light and store that nasty code.
I hope this helps the next time you see a code 26. Remember to verify the customer's complaint and find out what is happening when the code sets. Then try to duplicate the complaint. Be sure to test the component when it is failing -- not when it's working properly. If you have any questions, just give me a call!
Les Bentley is ASE master and L1 certified. He is a GM specialist with 16 years experience.
Automotive Network, Inc., Copyright ©2026