chucky420

Hello everyone, this is my first new thread. I'm hoping the brains on Honda-Tech have some insight and can point me in the right direction. About a month ago my daily driver (1997 civic ex 2dr 5mt w/ 2000 b16a swap obd2a P72 ecu) started refusing to idle normally. It would start fine, and warm up to operating temperature, but would idle at 1500rpm indefinitely. If I allowed the engine to warm up, turned it off and restarted, it would idle properly around 750rpm. However, when the car was moving, (clutch engaged, in gear, throttle closed) the idle would return to 1500rpm. From a stop, if the clutch pedal was released slowly, in first gear, without touching the throttle, the idle would jump from 750rpm to 1500rpm as soon as the car started moving. When the car was in motion there were problems with decelerating, the car wouldn't really slow down very much at all when the throttle was closed. With the throttle closed, in any gear, clutch engaged, car in motion, the fuel injectors would continue to run. Under those conditions the injectors are supposed to turn off creating engine braking until the the engine rpms drop to 1200rpm. I did some checking and found the TPS to be at fault, it read 1.6V with the throttle closed. I popped off the throttle body and the IAC valve and cleaned them up. Then I cut a groove in the two bolts/screws that hold the TPS on the throttle body and removed them with a flathead screwdriver. I went to the hardware store and bought two bolts that take a hex key and using them I reinstalled the TPS. Then I reinstalled the clean throttle body and IAC valve with new gaskets and worm gear type hose clamps for the coolant hoses (don't trust spring type clamps, compared to worm gear type clamps they suck at sealing and are a pain to deal with in tight spaces). I adjusted the, now adjustable, TPS by loosening the new hex key bolts and rotating the TPS. Unfortunately, at the end of it's adjustment range, it still read .9V with the throttle closed. I found a new replacement TPS online, even came with a new gasket and, once it arrived, I installed it. And, with plenty of room left for adjustment up or down, set it to read .5V with the throttle closed. Fired it up and it idled at 1800rpm, and progressively dropped to 900rpm as the engine warmed up. I adjusted the idle in the correct manner and when I fired it back up it stayed solid at 750rpm. Now, intermentantly, the car has the deceleration problem again. So I check the TPS again and it's a perfect .5V closed with a clean sweep to 4.8V wide open. I decide to backprobe the TPS wire at the ECU connector and drive it around the block. With the engine running, throttle closed, the TPS dropped to .4V, then I turned on the headlights and it dropped to .3V. I turned on the blower motor and it dropped to .2V, weird. I hook up another DVOM to the battery terminals and with the engine running, it's a solid 14V. Even with the headlights, blower motor, rear defrost, whatever, still 14V at the battery but almost 0V at the TPS. So, I move the backprobe to the ECT wire at the ECU connector and it's doing the same thing. As I turn on electrical components the voltage drops even though engine temperature isn't changing. I think the voltage drop in the ECT circuit is causing the ECU to interpret the input as a cold engine. Because of that the ECU is telling the engine to idle high. That's my best conclusion on the intermentant deceleration problem. I would suspect other sensors might be doing the same thing and I'm suprised that the MIL hasn't illuminated. Any one experienced a similar problem? I'm not totally stumped, but I am not sure what would be the quickest way to diagnose and solve the problem. Any ideas? Other than this the car has been solid for many years, 170K miles, about 120K miles on the swap, car had 25K miles when I bought it, 90K miles when I did the swap, swap had about 40K miles when I got it. Thanks for any good advice/information in advance.

94EG8

Did you bleed the coolant after you reinstalled the throttlebody? (your post was pretty detailed and I didn't see that mentioned so I figured i'd ask.

Former User

When you back probed the TPS and ECT sensor, were you measuring voltages output to the ECU from the sensor or voltages input from the ECU to the sensors? If you were measuring output voltages from the sensors, then do the input reference voltages from the ECU similarly vary? If the latter is true, then you may have an ECU problem.

chucky420

Thanks for the quick replies people. And for reading my long posts.
94EG8: I ran the car for about fifteen minutes with the radiator cap off, on a slight incline, with the front end lifted a little, while I checked for leaks and topped off the radiator. Then I put the cap back on and double checked everything, shut it off and let it cool down. After it was cool I filled the overflow up to the lower level, since it dropped a little when it cooled down. The radiator was full. For some reason, I never thought to check the factory service manual, but there it was. Almost identical procedure. Very small section on the cooling system, only 14 pages, hidden between the engine section and the fuel and emissions section. Yeah, I'm confident the cooling system is free of air.
RonJ@HT: When I backprobed the TPS wire at the ECM connector, I probed ECM connector D terminal 1 TPS red/black with the DVOM positive lead, and connected the negative lead to the chassis. The other two wires in the TPS circuit are ECM connector D terminal 10 VCC2 yellow/blue (power) and ECM connector D terminal 11 SG2 green/black (ground). I think that I did it right because the other two wires provide power and ground to other sensors so I figured that was the way to go. I should check to see if those wires provide good power and ground to the TPS. I checked the ECT by backprobing ECM connector D terminal 2 ECT red/white. The other wire in the ECT circuit is ECM connector D terminal 11 SG2 green/black (ground). Both the ECT and the TPS share this wire where the ECM provides a ground for the sensors. I'm leaning towards this being the problem, but I'm not 100% sure how to check it. Should I connect the positive lead of my DVOM to the battery and backprobe the negative lead to that wire and check for the voltage drop? Probably should find a source for lower voltage for that one. And for the (power) wire in the TPS circuit, I figure I should backprobe that one with the DVOM positive lead and connect the negative lead to the chassis to check the voltage drop on that end, right? I'll check that stuff out and get back to y'all. I still find it a little strange how it's fine with the key on engine off, but wrong when the engine is running. I read a post where another person had a very similar problem with an Integra, however his voltage was increasing. I don't know if it got figured out or not. The last post said the alternator got replaced but the problem still existed. Oh well, thanks again. Will update soon.

Filthypanda86

Hey Chuck did you ever get this problem resolved?
This TPS issue is driving me nuts! I replaced mine with a new unit and calibrated it correctly but I get different voltage readings when the car is driven as opposed to just sitting on the driveway with the key in-engine off.
My suspicion is that the ground wire from the TPS is not properly grounded, does anyone happen to know the best method of checking this? I read somewhere that setting the multimeter to continuity and probing the ground wire and the negative battery post would work, though I'm not entirely sure what readout I should get on my multimeter, (i.e. what I should be looking for).
On my vehicle after setting the TPS at .5 volts and starting the car virtually every accessory drops the voltage. Lights drop it down to .47, the cooling fan down to .42, and who knows what else. It bucks when driving, shifts extremely weird, goes hard into gear (auto), and idles very rough with a rising surge on occasion.

This is the best thread I've found so far on testing.