1993 accord lx auto trans.

I've been having a loping idle issue and have cleaned out the iacv and am likely going to end up buying a new iacv.
Recently, I was driving the car with the jumper shorted (blue obd-1 plug) and after the car warmed up, the d4 light started flashing the no.3 code (throttle position sensor).
After Turing the car off and driving it again a couple days later, everything was running fine and again, after it warmed up, the d4 light was giving the no.3 code.
It appears the tps needs replacing but everything else I've read states the tps rarely goes bad.
Side question: do you think this issue could be causing my loping idle?

 

The d4 light gives out tranny codes not check engine codes.

 

Is your check engine light flashing as well? If it is what code is it showing? Code 3 on the check engine light (MIL/CEL) is for the MAP while a 3 on the D4 is TPS related. You need to troubleshoot the check engine light (MIL) codes before the TCM (D4) codes.

If you are only seeing the D4 light flashing 3 times. Check the connector on your TPS. Make sure that all 3 wires are tight in the connector. If the wires are all clean and tight then you will need to check for a short in the wires or a fault in the TPS itself.

 

No cel, just d4 flashing 3 times, lather, rinse, repeat.

It appears I'll need to remove the tb to get access to the fasteners on the tps. Any pointers/things to look out for?

 

No need to remove the throttle body. You can test the electrical connections, wires and the TPS without removing it from the throttle body. Where this is only flashing a D4 code I would hazard a guess that it is more than likely a wiring or TCU issue than a sensor issue. if the TPS was faulty it would usually show up as an ECU error as well.

 

I can definitely unplug it but wow, not much space there to do any meter testing...

 

Do you have a CD5-7 Helms shop manual? They can be downloaded for free online!

 

I've got the oem svc manual from hondahookup...

 

OK so then you can follow the trouble shooting on 14-56. That should help you out!

 

avoid blox tps like the plague! f u blox! dorman is what worked for me.

 

in my 7 years at this board I have never heard of anyone had a tps go bad. Probably 2000 people have replaced them, but it has NEVER fixed any problem.

99.9% more likely either there is an open or the ecu is bad.

Its very easy to test em. You just backprobe the signal wire and it should go from .5v to 4.5v as you open it up very smoothly with NO dead spots just go very slowly as you open it and keep a close eye on the meter

 

I'm sorry to report that I haven't yet checked the TPS and continue to have the problem. Only after warm up and coasting will the d4 light begin to flash 3 times. I've noticed it only comes on if there is about 1-2 minutes of coasting. This is an automatic and I live in a hilly area (which is why there is so much coasting).

I am also having an issue with my idle (see prior post). It appears I'll need to just replace the IACV but am waiting on funds to build up so I can buy one (expensive part!!). The idle stays quite high while in Drive (about 1,400 RPM) and lopes while in park or neutral.

Since the idle is so high, it also seems to cause the motor to overheat if it's not moving, which is lots of fun at rush-hour!

Could the TPS issue and the other problems be related, you think?

 

Backprobe the TPS when the car is hot and note voltage. It may be fault with the sensor only when it is heat soaked.

Fluctuating idle is not from the IACV, common problem is the FITV screw backs out and not longer properly controls idle after warm up. Check the FAQ at the top of the forum on how to fix.

 

Thanks MadMike, et. al. - It seems like I have narrowed the issue down to the IACV though:
When I plug the lower hole inside the throttle body, on the side closest to the firewall, the idle returns to normal (no longer lopes) and comes down to a more reasonable idle speed.

Also, when I unplug the IACV while it's running, the idle stops loping and calms down to about 500-600 RPM.

I've disassembled the FITV and cleaned it all out and put it back after threading the plastic "screw" all the way down, then backed it off 1/4 turn.

So I'm guessing the TPS issue and the idle issue do not appear to be related?

 

Fix the TPS issue first. It is either a problem with the TPS or the wiring to it.
Verify your ground connections are clean, tight and undamaged. Faulty wiring/sensors can affect other functions.
TPS fault may be causing a feedback loop to the ECU which may be commanding the IACV to try to best correct a non-issue from a false signal(TPS).

Fix the known TPS issue first.

 

Results: ground check: neg probe to negative battery terminal, positive probe to green wire lead: not full continuity (010 measure instead of the expected 000).
Yellow: neg probe to neg battery terminal: 5.0VDC
Red: neg probe to neg battery terminal: 0.50vdc at full closed throttle, consistent measurements throughout throttle movement with a fully open reading of 4.39vdc. I was reading it was supposed to read 4.50vdc at full throttle, so I don't know what to make of the results...

 

When going from closed to WOT is the voltage change a smooth linear change?
If there are any gaps or jumps the TPS is most likely bad.

I am unsure of what you mean by 010 for the ground wire reading.

Verify all connections, even those to the ECU/TCU are clean and tight.
A faulty TCU may also cause a bad TPS reading.

 

Thanks again mm for the input.
1. Yes, smooth voltage with no hiccups from closed to WOT.
2. I can't recall the decimal point on my meter at the moment but when checking continuity, I usually get a .001 or .000 when I have full continuity. (1.000 with zero cont.) its appearing there may be an issue with continuity (grounding).
3. I will clean the contacts with some residue-free electronics cleaner ASAP and see what happens.

 

Continuity readings explained:
Simple Interpretation of Continuity Tester Readings

part-way down, it states if there is some resistance (10ohms or so), it can cause overheating, thereby causing issues.

I wonder if this is why the issue only shows up once the car is warmed up?

 

You are really overthinking this. Like mike said, it's very possible that the tcu is bad since if your tps was bad it would likely throw a regular obd1 code for tps. I would swap the tcu out. For the vacuum leak, even after screwing down the fitv it could still leak as the seals do wear over time. I would swap both out with one from the junkyard. The iacv's are harder to come by so if the vacuum leak is still there it might be wise to either get a new one from rockauto or get a used one off ebay.

 

According to the Honda service manual, when unplugging the IACV while at idle, if the engine speed decreases, replace the IACV with a known working one.

I tried this and while the idle seems like it's calmed down a little, it's still loping.

Is there a way to test the FITV before replacing it?

Also - about the TPS/TCU - where are the ground locations which would be within the electrical system for the TPS/TCU? Are there grounding locations that are notorious for causing issues?

 

The iacv closes when you unplug the connector so the idle is going to decrease. This is totally normal behavior and doesn't indicate an iacv is bad. I'm not saying that the iacv would be good but that is normal behavior. The TPS receives ground from the ecu. There are three pins for the connector going to the tps. One is ground, one is 5v reference and the other is the signal wire going from the sensor back to the ecu. With the ignition on but the car off, test for those. It's possible the ecu is not sending the 5v reference but it should be throwing a check engine code for it.

You should swap out the tcu and see if that fixes that code.

 

So new IACV, no dice.
New FITV, no dice, although it did seem to run slightly better, between the lopes!

However, I was inspecting each vacuum line trying to find any thing and unplugged the line going from the vacuum box to the throttle body, and the loping idle went away, and the car returned to a more normal idle. I took a video of it:
https://youtu.be/hk5jyi94VhI

 

Any takers on this one? Is this a sign of a bad map sensor?
I'm not sure what else to try short of pushing this blasted thing off a cliff.