help with error code 23 & 96

[TXHusker]

I've searched the threads but didn't find an answer.

I have a 93 Mustang GT EFI, C4, A9P, Ryan's harness, new Idle Air controller in my '74. Error codes 23 & 96 (plus the typical 81, 82, 84, & 85)

I will crank doesn't start right away after sitting for an hour or so. Once it does start, it will run for about 20 seconds, then die. I can crank it over and it starts right up, but then dies again after 20 seconds. After reading some threads, I disconnected the IAC and Mass Air, it would idle (a rough idle) for well over 5 minutes, then while running I plugged in the IAC, with Mass Air disconnected, and it would die within 5 seconds. Disconnect the IAC, start up again let it idle, plug in the Mass Air, with the IAC disconnected, it would die again. I replaced the IAC and it still does the same thing.

Tonight, I got the error codes (23 & 96) as mentioned above.

23 = the TPS voltage, I have set this to about .95 but it doesn't stay there. The TPS isn't loose, but when I check the voltage after running the engine, it's off from where I just set it.

96 = Fuel Pump Relay not sending power to fuel pump. When I pulled the relays, there was a white/blue reside on the relay by the connectors (looked like dry windshield washer fluid). So I am pretty sure this isn't normal. Have I fried the relays?

I have tried reserching this to come to the answer on my own, but I am going in circles.

Any suggestions where to start?

 

[Broncobowsher]

I am not going to tackle all this, but the one that catches my eye is the TPS voltage. If you set it with the engine off, then it is different with the engine running with no other changes. We have something funky going on. I am betting that the ground isn't good. When the Alternator fires up it provides an altered state of ground and this flakes out the readings. Try this, measure the voltage between different ground points (preferably with the engine running). There shouldn't be more then a couple hundredths of a volt difference between any of them. By the time you pull a tenth of a volt the computer is confused.

96 is the fuel pump moniter (computer watching the voltage at the fuel pump).
23 is the TPS problem you have already found and are working on.

 

[jurob]

take care of the relay 1st. clean contacts and connector then check tps.

with koeo ck voltage on the green wire. slowly go to wot watching voltage to make sure there are no dips or extreme jumps and that it goes to at least 4.5.

 

[jr]

Here is the procedure to adjust the TPS.

1) Loosen the two screws holding on the TPS, (The screws can be very stubborn, so make sure you have a good quality screw driver that fits snug and has a long handle for lots of torque, or you might strip the heads) just enough to able to swivel the TPS with force.
2) Pierce the "GREEN" wire with the pin and hook the positive (+) lead from the voltmeter to it.
3) Ground the negative (-) lead from the voltmeter. Use a good ground, like the engine or other grounded metal source.
4) Turn your ignition key to the run position, but do not start the vehicle
5) Rotate the TPS until you achieve .98 to .99 Volts (For stock or slightly modified engines).
6) Tighten the screws on the TPS and recheck.

Troubleshooting

If the maximum or minimum voltage you can obtain is close to the above recommended voltage setting (.98-.99), but the TPS just won't turn enough to get it right, you will have to modify the TPS. If you get a very low voltage reading, make sure you have an accurate voltmeter and your connections are solid, if so, you need to replace your TPS. If you have no voltage, you have a bad connection or a bad voltmeter. Rarely, if never, will a TPS be in such bad shape, especially if your car still runs good.

 

[jurob]

my experiences with a tps sensor, most come from the factory in the .7-.8 range. adjusting to .9x actually has made no difference in response or anything.

tps is important, but its got to read more than just idle. most important part of a tps sensor on the original eeciv mass air computers to me is being able to read wot. my memory is that wot is 4.5 volts or somewhere around there (don't quote me on that). if i cant detect wot when you are at wot you can easily blow a hg because the eec advaces timing much more in non wot conditions. same goes for the wot fuel multiplier.

 

[TXHusker]

I will check the ground and the relays tonight after work..and let you know

 

[TXHusker]

I didn't get the opportunity to look at the relays before Thanksgiving break

 

[TXHusker]

I posted this in another thread about RJM Harness...

I replaced my relays, and they seem to be working now, but I am still not hearing the fuel pump engage when the key is on..maybe a bad ground.

I have errors: 21,23,24 & 96 I may have other bad modules as well...

I've read many threads and they seem to point having a good ground. My question is why would the engine run if I disconnect the sensors (MAF, TPS)

So many questions, so few brain cells....

 

[Broncobowsher]

My question is why would the engine run if I disconnect the sensors (MAF, TPS)

So many questions, so few brain cells....

It runs in a failed sensor stratitgy. It uses other sensors and guesses what value the missing sensor should be around. Some things like the TPS have multiple duties, some more important then others. TPS mainly acts like an accelerator pump on a carb. sudden changes in throttle postion need sudden changes in fuel delivery or there are hesitation/bog/backfire/stall. So that is the main function, secondary is the transition from idle to cruise to power settings in the computer. So losing the TPS gives a drivability issue that feels like a bad accelerator pump and the computer now guesses what stratigy to use off MAF which may or not be rightbut will be close enough to keep the engine running until it sees a dealer for repair. So by eliminating bad signals the computer can actually be happier.

For the MAF example, if there is a bad MAF that is giving eronous airflow numbers then there are bad calculations for just about everything. Unplug the MAF and the computer ditches that signal and goes to a limp mode looking at RPM and throttle position, there is a little table burried in the computer that is a rough calculation for MAF that the computer uses. This isn't computely right but it is close enough to get the car running and to a dealer for repairs.

The only signal the computer can't live without is PIP. If it doesn't see the engine moving, it won't even try and squirt the injectors. No fuel, no fire, no engine run. The TFI will make spark as long as it is powered up, just won't have any advance to it. Result, you can run an EFI engine without a computer providing you find a way to feed it some fuel. It won't run right, but shootig a can of (flammable, not the non-flamable) brake clean or carb cleaner into an engine will feed let it run as long as there is spark.

Some of the other sensors it just guesses at as well. If it sees no signal it knows to ignore it and use a guess. No coolant temp, look at air temp. It's main purpose is to help with cold start anyway. So it doesn't start very good when cold, but it will still start. As soon as it is running for a few minutes the O2 sensors kick in and correct for the missing sensor and away you go.

 

[TXHusker]

I replaced the TPS and that took care of error 23, I replaced all the relays but still have error 96. I also had errors 21 & 24 (not previously mentioned) which relate to the ECT and ACT. I replaced both of them but the errors are still there.

I disconnected the battery while I replaced these sensors which was about 30 minutes. That should be long enough to erase the codes.

I tested the fuel pump and it is working fine, I now hear the relays click when the key is turned on. As I understand, the fuel pump only engages when it senses a drop in fuel pressure. Correct me if I'm wrong there.

Any other suggestions to check out? Anyone in OKC with this setup?

 

[ransil]

I tested the fuel pump and it is working fine, I now hear the relays click when the key is turned on. As I understand, the fuel pump only engages when it senses a drop in fuel pressure. Correct me if I'm wrong there.

Any other suggestions to check out? Anyone in OKC with this setup?

You are wrong.
Turn the key and the fuel pumps should com on for a second or 2 when running it should be running all the time.

There is a wire that tells the computer that the fuel pump is running, find it and make sure it connected, you will need to find where it's at, first place I would look is on the output of the relay and ohm it out back to the computer.

Very unlike you have more than one sensor bad, they just don't do that, ECT & ACT need a warm engine before testing, its possible the computer is bad water dripping on it or welding on the bronco with it connected, or just its time, however for the most part they are pretty reliable.

Check your ground wire(s).
my standard answer here "sheet metal screws make crappy grounds".

 

[Huskerbronco]

I got code 96 when I had a bad ground for the fuel pump. The fuel pump would not come on for the 2 sec. like normal when the key was turned to the run position. Once I got my frame properly grounded the code went away and the pump started to work.

I think the computer looks for a voltage drop or something to tell if the pump is running. If there is no ground there can be no voltage drop, or something like that.

Just my 2 cents.

 

[TXHusker]

I appreciate the help and I am working on better grounding.

As I read the instructions, the step that discusses the inertia switch keeps glaring at me. I haven't purchased one yet. The instructions lead me to think that it was optional. I understand the benefits and will get one, but the instructions don't tell me it is vital part of the harness (i.e. completes the circuit for the fuel pump) any thoughts?

 

[Huskerbronco]

I appreciate the help and I am working on better grounding.

As I read the instructions, the step that discusses the inertia switch keeps glaring at me. I haven't purchased one yet. The instructions lead me to think that it was optional. I understand the benefits and will get one, but the instructions don't tell me it is vital part of the harness (i.e. completes the circuit for the fuel pump) any thoughts?

Yeah the inertia switch completes the circuit on the power side of the fuel pump. The pump will not run without one installed. Try jumping the two pink wires in the connector for the inertia switch to see if the code goes away. I had to do that before I got a inertia switch installed.

 

[TXHusker]

that did the trick!!!

I created a jumper for the inertia plug, turn the key on I heard the fuel pump kick on and it fired up on the first try. ;D

Now I know what everyone else has felt after they completed the swap!

thanks for the assistance!

 

[1971sport]

Just for the fun of it you should be checking signal return (computer ground) since many sensers are tied together .having that many codes at one time is caused by disconnecting things while running or just flat out shorts .start by having codes erased by a good diag tool and not just disconnecting the battery. then run engine and check codes also sounds like its not going into open loop for start up a good tool will let you read voltage to each sensor as the computer sees it