EFI sensor trouble

I am thinking it is most likely to be a problem with a corroded or loose connector or wire.

Have a good look at the connectors mentioned above to see if there is any obvious problem.

Seagrass
 
dnktng said:
Since it is happening with the throttle partially open, that sounds more like the TPS or its wiring having an intermittent fault (either shorting to ground or opening) than a battery problem.

The wiring from the ECM to the TPS includes 3 wires..... the 5VDC, ground, and the 0-5VDC signal that returned to the ECM. All three wires are in a connector at the ECM, a 10 pin white connector near the throttle bodies, and the connector at the TPS. The ground wire is also connected to chassis ground.

Many post concerns about the 5 VDC supplied by the ECM, but that same 5VDC that is wired to the TPS is also wired to the MAP sensor.....and the MAP sensor has a low and high voltage failure mode but isn't giving a fault. It is possible the TPS is more sensitive to voltage changes or operates closer to a low or high voltage fault than the MAP sensor. But in your case, you were getting faults with the throttle in near mid position and it should have been a long ways from a low or high voltage fault. The same ECM sends the regulated 5vdc to the Water Temp and Inlet Air Temp Sensors also, but we never hear of high or low voltage faults from either of these sensors.

The TPS is a simple and reliable device. If it was mine, I would gently fully open and close the throttle with the engine off many times.....likes 20 or 30 ......just to exercise the TPS. Then I would disconnect and reconnect the 3 connectors between the ECM and TPS. A few gentle taps of the TPS with the plastic end of a screwdriver may help wake it up. Then ride and see if there is any improvement. If faults continue, I would have to bite the bullet and buy a used Throttle body.
Click to expand...

What "dnktng" says is right although I disagree with the last paragraph. Also, even an analogue meter is not really fast enough to pick up tiny dead spots in a resistive sensor unless you are incredibly slow with the sweep.
An analogue "scope" is the best way to find it as they react instantly to the fault without requiring digital conversion or physical conversion (to a needle) which can often hide the fault. This type of problem is a real pain in the .... for mechanics as they can be very hard to pin down and can take a lot of time to confirm.
The fact that it has happened at idle as well as open throttle makes it less likely that it is the TPS itself but it would be good to eliminate it.
It may be a loose or dirty connection between the ECU and Sensor and that too should be eliminated.
I had a bad connection on the Intake Air Temp Sensor that caused me trouble and it was because the connector wiring had been wound around the harness to take up the "slack" and had pulled one of the pins partly out of the connector.

Macka
 
How about slow idle speed being the culprit??
Nice day in Puget Sound region today, so I got the NT out for a short run and investigate the MIL problem.
1. Ignition on before starting: no MIL 2. Started engine: no MIL 3. Down the road about 10 min with couple stops for traffic signals: no MIL 4. Goosed throttle up hill: no MIL 5. idle in driveway to let temp gauge come up to half; then about 1 or 2 minutes later I hear the idle RPMs come up a little, the engine runs a bit more smooth and NOW I see the MIL at 8 flashes and I suspect the ECU has gone into a fail safe mode, set the code, and bumped up the RPMs.
Am I making any sense? I think the bonehead [me] who does the valve adjustments didn't do it quite right last time and was satisfied with idle speed at COLD idle when done recently [mileage wise]. I plan to check into this!
 
It sounds to me like the system voltage drops too low when the fan comes on (if it did) whilst idling. Otherwise it could still be dropping low at times if the reg/rec is faulty.
It would be good to use a "minimum/maximum holding" multimeter on the battery to see what the voltage is doing or a "recording scope" to watch the trace for downward spikes in the voltage. It REALLY sounds like you have an intermittant low voltage problem.

Macka
 
Did you ever replace the battery? I mentioned in an earlier post that these AGM batteries can do some strange things including huge voltage drops so that could be the cause of the problem.

Seagrass
 
Similar problems cured with a new battery. Yes it still started the bike.
 
Have not replaced battery yet, but will shortly. My charger indicates the battery at 95% every time it's been connected, but it IS 6 yrs old now.
So more to the story lately. I'm trying the idea of pursuing an emissions warranty with Honda because: As noted here, the warranty should go 5 yrs and about 18k miles. My bike has 13.2K AND was into the dealer for a diagnosis of MIL code 8-1 on 7/3/14 at which time the report stated: "if the MIL comes back, replacing the TP/Throttle body would solve the issue." Seems reasonable to me that if the problem was documented within the warranty period it can be honored outside the time period. Of course this is a judgement call.
Honda customer service says they will give further consideration of the question if I have another, current assessment done by dealer [for another hundo$ of course]. I've made an appointment, but will try adjusting idle RPMs and replace battery before that time.
 
Seeking advice about clearing codes to get the MIL light to go off. I've seen some postings about this but would like more clarity. I don't need to clear everything [?] but want to use the on bike system to make a current [both voltage & timely] test after I make changes to battery and idle speed, to verify if problem solved by those means. In the past, when I shut down the ignition, the light would stay off, but I tried that last time and it was back on again soon as ignition came on. I do note that there is a process to short out terminals to clear the codes but there may be other ways such as disconnecting battery [at battery change].
Thanks to everyone posting in my time of need!! :)
 
Tourguy said:
Seeking advice about clearing codes to get the MIL light to go off. I've seen some postings about this but would like more clarity. I don't need to clear everything [?] but want to use the on bike system to make a current [both voltage & timely] test after I make changes to battery and idle speed, to verify if problem solved by those means. In the past, when I shut down the ignition, the light would stay off, but I tried that last time and it was back on again soon as ignition came on. I do note that there is a process to short out terminals to clear the codes but there may be other ways such as disconnecting battery [at battery change].
Thanks to everyone posting in my time of need!! :)
Click to expand...

1. If the dash MIL is on and you are able to do the kickstand down & clutch release trick to make it flash an 8 code, it is an active fault.

2. Anytime a fault is received, regardless if you see the dash MIL when it was active, it will store the code in the ECM. That code can be later read using a shorting wire even after the intermittent fault has cleared. If you have had 53 fault codes 8's come in and clear on the dash, there will one code 8 stored in the ECM. My first guess is that the stored ECM code does not get reset by disconnecting the battery....not 100% on that though.

It is worth noting, and would make troubleshooting your problem MUCH EASIER, that you have an active fault with the engine off and parked now, not just when riding. It also says that if your fault was initially aggravated by engine vibes that it has progressed and is now giving faults even when off.

Suggestions:
-With the active fault with the engine off would be an ideal time to take the tank off to expose the TPS and connectors, confirm the code 8 is still on with the engine off and just the ignition on, and systematically tap and giggle connectors and the TPS to see if the dash MIL code 8 active fault clears.
-If you are getting a the active fault with the engine off check your battery voltage and connect a charger to raise that voltage to see if your active fault clears. It it doesn't clear with a charger on, there isn't much reason to believe a new battery would fix the problem.
-Turn your ignition on and verify the active fault is still on and is a TPS code 8 fault. Without starting the engine, rotate the throttle to see if it clears. If it does clear it would nearly 100% verify that it is the TPS because it is the only thing changing, and not a connector or ECM. Unfortunately if the MIL doesn't clear it doesn't rule out the TPS.
 
dnktng said:
1. If the dash MIL is on and you are able to do the kickstand down & clutch release trick to make it flash an 8 code, it is an active fault.

2. Anytime a fault is received, regardless if you see the dash MIL when it was active, it will store the code in the ECM. That code can be later read using a shorting wire even after the intermittent fault has cleared. If you have had 53 fault codes 8's come in and clear on the dash, there will one code 8 stored in the ECM. My first guess is that the stored ECM code does not get reset by disconnecting the battery....not 100% on that though.

It is worth noting, and would make troubleshooting your problem MUCH EASIER, that you have an active fault with the engine off and parked now, not just when riding. It also says that if your fault was initially aggravated by engine vibes that it has progressed and is now giving faults even when off.

Suggestions:
-With the active fault with the engine off would be an ideal time to take the tank off to expose the TPS and connectors, confirm the code 8 is still on with the engine off and just the ignition on, and systematically tap and giggle connectors and the TPS to see if the dash MIL code 8 active fault clears.
-If you are getting a the active fault with the engine off check your battery voltage and connect a charger to raise that voltage to see if your active fault clears. It it doesn't clear with a charger on, there isn't much reason to believe a new battery would fix the problem.
-Turn your ignition on and verify the active fault is still on and is a TPS code 8 fault. Without starting the engine, rotate the throttle to see if it clears. If it does clear it would nearly 100% verify that it is the TPS because it is the only thing changing, and not a connector or ECM. Unfortunately if the MIL doesn't clear it doesn't rule out the TPS.
Click to expand...

Not sure this trouble shooting procedure is valid.

If bikes are the same as cars, the fault will not clear until the ECM runs a complete cycle without error. This means you need to start the bike and go for a ride of at least 20 minutes or so to give the system time to warm up properly etc.

Seagrass
 
The EFI system on the NT is pretty basic (which is good) and I don't think that it is "smart enough" to reset after X number of cycles like many modern car systems will.
I also don't remember if disconnecting the battery will clear any or all memorised codes. I have never tried it myself. If memory serves me I think someone here may have answered that question before.
ASAIK the codes will stay until manually erased.

As I mentioned earlier, TPS failures are VERY rare. Many Throttle Bodies have been replaced unnecessarily.

Macka
 
Last edited:
seagrass said:
Not sure this trouble shooting procedure is valid.

If bikes are the same as cars, the fault will not clear until the ECM runs a complete cycle without error. This means you need to start the bike and go for a ride of at least 20 minutes or so to give the system time to warm up properly etc.

Seagrass
Click to expand...
Interesting, care to make a wager. Loser has to let the winner use his NT for 2 weeks of touring the other's country. The MAP sensor can be unplugged to test the idea and is easily accessible from the right middle cowl cover (the vented black plastic cover near your knee).
 
dnktng said:
Interesting, care to make a wager. Loser has to let the winner use his NT for 2 weeks of touring the other's country. The MAP sensor can be unplugged to test the idea and is easily accessible from the right middle cowl cover (the vented black plastic cover near your knee).
Click to expand...
Since it is easy to test, I did it. Here is what I found.
-Disconnect MAP Sensor, turn on ignition, and get a MIL. Do the kickstand/clutch trick and it flashes 1 for MAP sensor Hi or Lo volts as expected.
-Reconnect sensor and MIL stays on. This is all done without cycling the ignition. I waited about 30 seconds and then.....
-Cycle the ignition, MIL cleared.
-Read ECM, has a 1 stored, reset it. done. No riding, No warm up, but the ignition would need to be cycled to clear the MIL on the dash after each component you check or the MIL would stay locked in. I don't know if I left it with the ignition on and not started if the MIL would ever clear.....I suspect it wouldn't unless I start it. After all, MIL do come on and go off while riding without cycling the ignition, I've seen it happen.
 
Macka said:
I also don't remember if disconnecting the battery will clear any or all memorised codes. I have never tried it myself. If memory serves me I think someone here may have answered that question before. ASAIK the codes will stay until manually erased. Macka
Click to expand...
I don't see that there is any 12Vdc applied to the ECM with the ignition switch off. If true, disconnecting the battery would be no different than turning the ignition off, which doesn't clear the ECM.
 
dnktng said:
I don't see that there is any 12Vdc applied to the ECM with the ignition switch off. If true, disconnecting the battery would be no different than turning the ignition off, which doesn't clear the ECM.
Click to expand...

Thanks dnkyng, that agrees with what I believed to be correct as it seems logical. Manual erasure is necessary to clear the codes.
Your experiment also confirms it. It is handy to know.
Macka
 
Last edited:
Great info!!! Details important here. Did you reconnect the MAP sensor "hot" or ignition off? And when you "cycled" the ignition, you were just turning it on to let it go thru the test mode and NOT a complete off then on cycle? Perhaps you did reconnect hot and cycle by then turning ignition off and back on to see MIL was off.
Already have the new battery and will get the new one in today and fully charged. Then will do the tests and report back. The $$ [dealer] appointment is this Friday.
 
Tourguy said:
Great info!!! Details important here. Did you reconnect the MAP sensor "hot" or ignition off? And when you "cycled" the ignition, you were just turning it on to let it go thru the test mode and NOT a complete off then on cycle? Perhaps you did reconnect hot and cycle by then turning ignition off and back on to see MIL was off.
Already have the new battery and will get the new one in today and fully charged. Then will do the tests and report back. The $$ [dealer] appointment is this Friday.
Click to expand...
The MAP Sensor was reconnected with the ignition on.
When I cycled the ignition, it started ON, went of OFF, back to ON and let the self-check run.
Do you still get the MIL and Code 8 by turning the ignition ON without starting the engine? If so, that would make it so much easier to determine the cause than the intermittent MIL only when riding.
 
Yes, thanks, MIL code 8 still active now without engine running. I got the new battery installed and turned on ignition with charger attached . . . made no difference in MIL light. [It stays on now.]
AND i can fiddle tomorrow with connections and such as you smartly suggested earlier. Evidently, I don't need to clear any codes because the MIL light will indicate if a fault still active. I have seen info about a "DLC" in the manual [not DTC] and will be looking for that and the white connector you mentioned. I have some "bulb grease" which must be dielectric for putting on contacts. Hopefully I'll have time to check voltage out of the TPS and maybe even resistance thru the connectors.
I'll have more to say tomorrow.
Thanks again!
 
Tourguy said:
Yes, thanks, MIL code 8 still active now without engine running. I got the new battery installed and turned on ignition with charger attached . . . made no difference in MIL light. [It stays on now.]
AND i can fiddle tomorrow with connections and such as you smartly suggested earlier. Evidently, I don't need to clear any codes because the MIL light will indicate if a fault still active. I have seen info about a "DLC" in the manual [not DTC] and will be looking for that and the white connector you mentioned. I have some "bulb grease" which must be dielectric for putting on contacts. Hopefully I'll have time to check voltage out of the TPS and maybe even resistance thru the connectors.
I'll have more to say tomorrow.
Thanks again!
Click to expand...
The DLC is the red capped connector in front of the ECM.
The white 10 pin connector is best found by following the TPS wiring back from the TPS. The wiring will go thru the small connector for the TPS only, then keep following the wiring back to the 10 pin white connector. IIRC it is on the left side frame near the Throttle bodies/rear cylinder head cover.
 
I believe we have the answer to the mystery in my case. I decided to start at the blue connector under/at the TPS. I disconnected it and put bulb grease into the wire side [female] and reconnected it back to the TPS. Then with the ignition on again, the MIL light was off . . . and it stayed off while I wiggled/pulled on the wires and tapped the sensor!! I tested a few times more [just ignition cycling] while replacing parts back onto the bike and then ran it until achieved full temp . . . NO MIL!!
I want to acknowledge that failing connections was a cause mentioned, possibly more than once, in previous posts. I didn't take the time to do the other tests as planned . . . didn't see a need for it. I'm cancelling my service visit, Honda isn't replacing my throttle body when a connection is the problem. Seems a bit of letdown with all the possibilities and tests, etc. then to have it be the very first thing I tried. Sure good to see it works!!
I did note, when looking up at the TPS wire, that it sits on top of the engine so gets considerable heat, but there are other wires in that condition so is probably not a reason this connector suffers a recurring failure. Perhaps when the connection grease gets hot it flows out/down from the connector and needs regular replacement. I hope to do real test ride tomorrow.
I did find the DLC, but skipped searching the white connector.
Thank you all for comments and especially dnktng who was so very consistent with ideas and information!!! Gave me some confidence about how to proceed AND hopefully saved me SOME MONEY!!! :smile: [until the next thing :wink: ]
 
You must log in or register to reply here.
Back
Top Bottom