Put on test pipe now car runs poorly

The test pipe on its own should cause no requirement that the FI be re-adjusted - the Kjet will do that on it own.
A place to start is checking the O2 sensor is installed well and that the electrical connectors are all snugly seated and clean. Next would check the FI fuses in the trunk (right hand side) on the FI ECU panel and relays.

best
rt

 

So, it appears the 'test' pipe worked and it gave you an answer that somethings wrong...

 

Frank, you'll be glad to know my 308 is now a proud owner of a hyperflow cat.

 

Buy a Hyperflow. Trust me, it's like a "clean" test pipe. I can't tell any difference between the two. Same sound, same power, no more eye-burning fumes.

 

removing the cats will cause problems if the EFI uses feedback from O2 sensors. becouse now the cats are gone the overly rich condition exists and the EFI will try and cut back on the fuel trim, leaning it out. until WOT then it's no longer in feedback but open mode. there are O2 sims for the very purpose, to make the EFI think you still have cats.
for the early CIS pre lambda EFI there is no O2 sensors so you'll get a bit better performance from removing the cats but a whole lot smellier.

 

On my 84 308 the O2 sensor is in front of the catalyst, (the thermal probe is after it) so I am assuming it has no idea if the car has a cat or not. I guess that is why it will run fine with or without one.

 

Bingo.

 

If the car does not run well without a cat there is something wrong with the car. You should consider getting the underlying problem fixed.

I do not advocate removing cats but something is wrong for it to react that way.

 

Really, toss the test pipe and go Hyperflow.

 

I'd like to bring up another issue that may be involved here (but I'm unsure so I'm just asking).

On these K-Jet with Lambda systems, there seems to be an "intelligent" electrical connection between the thermocouple ecu and the injection ecu (that is not present on the F KE-Jet with Lambda systems). If this signal is used to tell the injection ecu that "OK, I'm now hot enough that you can trust the O2 sensor output as a valid signal for A/F ratio control", it might be that without a functional cat present (to heat the thermocouple), the injection ecu never gets this signal so the system defaults to run in an open-loop fashion when warm rather than a closed-loop fashion (so the warm-running would be more sensitive to the mechanical adjustments).

So my question is: has anyone with a US '84-'85 308QV or US 328, that has a test pipe installed, actually confirmed that system runs closed-loop at warm idle (i.e., does the O2 sensor output voltage vary between 0V to +1V in a quasi-periodic manner at warm idle consistent with actual closed-loop operation)?

Copterjon -- one thing you might try, since you've reinstalled your cat, is to just remove the thermocouple from the hot exaust gas stream but leave it electrically plugged in (or maybe just unplug the O2 sensor) and see if you get the same poor warm-running symptoms that you had with the test pipe installed.

Just a thought -- and would be grateful for any information/data...

 

Thanks for volunteering John. If you have a voltmeter and would like to try measuring the O2 sensor output directly during operation from cold to warm, let me know, and I'd be glad to give some instructions, if needed (it doesn't take any real disassembly, and you don't have to be Dave Helms to do it ).

 

That's how I did mine. Worked fine.

 

Great post, Steve. Without the cat, the thermocouple might not reach the temp required for good closed loop running.

I would speculate, wildly, that one such test for a poor running no-cat car would be at wide open throttle when most systems are designed to run in open loop mode. If an otherwise poorly running no-cat car works fine at WOT, then the relatively cold thermocouple could easily be the cause.

 

Hey Brian, aren't you forgetting about the thermocouple? Will it really get hot enough without a cat to send the correct closed loop signal to the ECU?

 

No

It goes to the light on the dash only, it performs no other function.

And for the temp range it is designed to read the only way it can get high enough to reach the low end is for it to be on the inside of the pipe on the output side of a cat.


Again, if the car is in proper running order removing the cat (and the thermocouple) will have no negative effect.

 

Are you sure that's how the 3.2 Mondial's handle it?? On my 348, a thermocouple tells a cat ECU that the car is now hot enough to trust the O2 sensor, initiating closed loop mode when warm and when not at full throttle.

Remove the cats and you'd have to wonder if the cat ECU would signal the a/f ECU for closed loop lambda operation.

 

My car, with HyperFlow cats, doesn't have ECU's or thermocouples. Runs like a bat from hell.

 

Is it ever going in to closed loop mode, though (or have you just tuned your open loop well)?