[teal95]

1987 Thunderbird TurboCoupe, 2.3t, T5 swap from A4LD, 8UA processor

I put a different engine in the car (old one idled fine but bores were trashed due to leaking headgasket that then sat for an extended period) and it won't idle. It idles fine for 30 seconds to a couple of minutes then it starts climbing up to 2k. Occassionally it will pause for a while at 1500 before continuing. I have a twEECer on it and have datalogged it. It's not the TPS because I can watch it. The TPS is fine, RATCH (min learned throttle) is fine and the EEC knows when it's in idle mode. IAC DC is ~40% when it's idling correctly and climbs in proportion to the idle speed. When it's at 2k it's around 60%. I've pulled codes (15 due to twEECer, 28 due to cold out, 81 due to unplugged boost sensor, 85 and 88 since it has a T5 now, 27 due to old VSS and 72 because the only way I've found to get the idle down is to quickly shut the ignition off and on) and got nothing useful. What would cause the EEC to raise the idle but wouldn't set a fault code and is cleared by an ignition cycle?

Any ideas?

steve

[decipha]

sounds like the tweecer is corrupting some code associated with the idle air logic

it also wouldn't hurt to check for the rpm idle adder based on idle time, this would be easily noticeable by the dsdrpm payload increasing with the iscdc

i would remedy this by disabling idle air correction by setting both the ipsib scalars to a max correction of 0, then dial in your air and enjoy

[teal95]

I'll take a peek but you have to remember the L series is pretty primitive. I don't believe it has a DSDRPM, it doesn't control idle the way the newer stuff does. It only has a desired airflow. Spark is fixed at 25 during idle (it's not using spark to fine tune speed). And I haven't changed any of this stuff. It's running a stock binary. I don't believe it's a tuning problem, I believe it's mechanical. I was just hoping since I had data I could logically figure it out.

It's very common for the exact same idle problems to appear in the turbo cars and it's almost always a failing TPS. RATCH gets learned as 0.0v and then it never goes into idle mode. But since I have data I can see that isn't the problem this time.

tanx,
steve

[decipha]

sounds like its time for an ecu swap

[teal95]

I don't think it's the EEC, but I'll try that tomorrow (I have a spare in the trunk). I can't upgrade the EEC without changing the low impedance injectors. And I'm pretty sure it's a mechanical issue, I just can't figure it out.

steve

[donnysey]

You will have to solve the neutral safety/clutch switch codes if you want a turbo coupe to stop doing a high idle. It has to see that switch or it will just keep doing that. Used to work on lots of these. look at the wiring diagram for the original transmission neutral safety circuit. The EEC is basically looking to complete a ground to the starter solenoid when in park or neutral on automatic cars. Manual cars the EEC is looking for a complete circuit to ground on the solenoid between shifts and at start up. If it begins seeing big rpm swings (shifting gears) without that switch cycling it will go to the default high idle. Most 80s Fords worked this way. Not sure if you have a switch in tweecer to throw between automatic and manual like you do in BE and the GUFx strategies but that would be a good start and then make the clutch switch work.

Sometimes a vane air flow meter will cause this on its way out. Generally the engine will sort of seem to hesitate for a second and then idle up. It will show up on a labscope right before the the problem occurs if you watch it.

I would go for the first problem though with the codes you have and be sure that when you run the self-test you follow the directions regarding pushing the clutch pedal, turning the steering etc so you don't set false codes.
Cheers!

[teal95]

Some idiot didn't plug the vacuum port when they removed the A4LD. I can't believe I didn't hear it the first 15 times I went hunting.

steve

[decipha]

ahh idiots can be quite clever in their stupidity