PGMFI Logo
Library> Ecu Troubleshooting >TroubleshootingSolidCEL Destinations: Home | Library | Change Log | Index
Search | Go

Solid CELs are arguably one of the easier problems to troubleshoot because there are relatively few things that cause them to happen. A "solid" CEL occurs when there is a serious hardware failure on the ECU. The Ecu Backup Processor will kick into gear and (minimally) try to run the car. This generally results in the car running really rough if at all.

In OBD0 for instance, there are only 12 things that the ECU checks: SFRs of MCU, Checksum, internal RAM of MCU, external RAM, Further checksums, Analog to digital converter failure, 6260 Error, ... All of these things are very low level. Rarely are any of these things the problem with a solid CEL on a chipped ECU, however.

95% of the time, solid CELs on a chipped ECU come from one of two problems

  • Improper soldering of components necessary to chip the ECU
  • Inappropriate or bad program burned in chip installed in ECU

To troubleshoot a solid CEL on a chipped ECU:

  1. Burn a "stock" program for the ECU. If the ECU originally was a P06, use an unmodified P06 program. If the ECU was a P72, use an unmodified P72 program. If the ECU is a non-vtec ECU converted to vtec, try the original non-vtec program first and a stock P28 program second. You can find stock programs here.

  2. Install the ECU in a car (or power it up on a bench) with the stock program installed. Check to see that the dash CEL clicks on, the fuel pump pressurizes the rail for approximately 2 seconds and then the pump and CEL kick off. If the ECU behaves normally with a stock chip, your problem was with the program you were trying to use. Verify that the Check Sum is correct and that the ROM has the appropriate hardware features disabled, if necessary. See OBD1CodeCompatibility and OBD0CodeCompatibility for more information.

  3. If the ECU still has a solid CEL and identical behavior with the stock program intended for it, 99% of the time there are poor solder connections on the EPROM or latch used to chip it. For OBD1 ECUs, cut J1 for an additional test. With J1 cut, the ECU should be returned to stock and no longer have a solid CEL. Rubbing alcohol is a good solvent to use to clean up flux and other residues from soldering in order to inspect parts for poor connections. Whenever possible, inspect the top side of the board to ensure adaquate penetration for connections. (someone please add more here - maybe pictures of bad solder joints?)

  4. If the OBD1 ECU will not work with J1 cut, in all likehood it is fried and should be used for parts and / or replaced.

This is intended to be a basic introduction and nothing more. Please add your own experience and wisdom to it.

  1. If you have added capacitors to any of the non-power signals (C91 & C92 on a JDM P30 for example), make sure they are not too large. I put 0.1uF decoupling caps there, and got a solid CEL because the low reactance pulled the PSEN signal down to a solid 3V and held the ALE signal at ground. Per info on the General Hardware Assistance Forum, C91 & C92 should be 0.00001uF, should you choose to use them.
  2. If you want to check your soldering for opens/shorts/cold solder joints, you can check for continuity using the Ecu Chipping Wirelist. Others will have to check to see if this wirelist is appropriate for other ECUs -- markolson - 27 Dec 2004

Parents: Web Home > Ecu Troubleshooting Revision: r1.2 - 27 Dec 2004 - 18:16 GMT - markolson { Edit | Attach | History | More }
Copyright © 2002-present by the contributing authors. All material on this collaboration platform is the property of the
contributing authors, and is covered by the Non-Commercial Share-Alike License unless explicitly stated otherwise.
Ideas, requests, problems regarding the PGMFI TWiki?? E-Mail the WikiAdmin
Site Designed By: Digital Fusion   Need a website?