odd issues with Kpro - please help

[garz]

This is an odd one - hoping someone on here may have seen it before - or can offer an explanation

after much wiring fault checking and sensor checking it seems my Kpro ECU ( PRA type) has some kind of hardware fault

the symptom is that the Intake Air Temperature reading was always high - around 20 degrees higher than the engine coolant - when the car has been standing a day or so.
but not only is it high - it actually continues to creep higher over time

sensor tests fine - and issue exists with other sensors
wiring tests fine for continuity between the sensor and the ECU - no odd resistances

my working theory is: the Kpro has some kind of fault causing it to send too much current to the sensor and causing it to self-heat - hence the high temperature and the temperature creep

would not be such a problem and would probably run fine on a N/A car.
however my engine is SC + CC and the extra 20+ degrees is causing the ECU to back off ignition timing quite a bit as it thinks the charge temperature is much hotter than it is.

anyone seen this - or have any other theory ?

[Hondata]

The KPro does not control the IAT current - the ECU does. With the ignition off there should be no voltage on the IAT signal line. Perhaps check this to see if it is heating via the 5V pull up resistance.

[garz]

yep - understood - not driven by the Kpro - just driven by the original OEM ECU hardware
asking here just because there are lots of honda tuners here using this ECU

you mean to check that there is no voltage at ECU pin B17 ( IAT sensor signal ) vs the battery ground ( or ECU signal ground ) with ignition off - not sure i checked that - will do - Thanks for that.

also checked the following
1, resistance between B17 (high side - IAT pin) and the share SG2 ground ECU connector A10 ( with the sensor connections bridged ) = 0.2 Ohms - ie about right for that length of wire .

2, that the 5v bus (Honda call it vcc) is actually 5v - it was ( ECU connector A - pins 20 and 21 are 5V vs signal grounds on Pins A10 and A11 with ignition on )

3, that the signal grounds on the ECU were at or v close to 0v vs battery - they were 0.06v - so OK

4, checked all the other ECU grounds are v close to 0v vs battery negative terminal
A4 and A5 were both 0.025V vs Battery negative terminal

[Hondata]

Check voltage B17 to signal ground with the ignition off.
An additional check would be to unplug the IAT sensor when you shut down the vehicle, and plug it back in immediately before starting it. This would verify if it was the IAT heating or not.

[garz]

i like your thinking - i will defo try that too

[garz]

OK
finally did those checks - here is what i found

1, pin B 17 on the ECU (IAT signal line ) was 0.0 volts vs battery negative terminal with ignition off

2, after the car has stood for 1 week and not been started - ambient temps around 16-24 degrees Celsius - connected ECU to laptop - disconnected the IAT sensor plug - started kmanager - turned on ignition - connected to ecu - and began reading IAT and ECT sensors - ECT was 17.5 degrees C
i, was surprised to see a IAT reading of 13.2 degrees C - even with the IAT sensor disconnected ....
ii, this temperature reading began to slowly climb - to 14.5 degrees C - over the next 30 seconds
iii, at this point i plugged the IAT back in to see what difference it made - temperature immediately registered - 35.6 degrees C
iv,IAT reading gradually decreased to 32.6 deg C over the next 30 seconds
v, at this point i unplugged it again and reading immediately fell back to 13.2 deg C

looks like something odd happening here - would have expected no reading - or minimum possible reading with the sensor unplugged ?

one other possible clue - while testing the B17 with ignition on / off - i have a battery isolator on the negative terminal of the battery. with this disconnected i then tested B17 to the negative battery terminal - ie the battery side of the isolator - i got 12 volts.

i guess the key question, if you can help me with it, is - does this establish there is a problem with the ECU - or the car wiring somewhere else ?

i realise this is not strictly Kpro driven - and very much appreciate the support

[garz]

here is the data log of the above in case it helps

[Hondata]

It is a little hard to say. The rational behind testing B17 to ground with the ignition off but battery connected was to determine if there was stray power on B17. Certainly is should not be higher than 5V, but I think with the battery isolated from ground it was not a valid test.
At this point try substitution for the ECU and IAT into a different vehicle.

[garz]

yep - thanks again - i agree - just have to find someone willing to exchange to do the next checks

[garz]

OK
got a hold of another working PRA ECU - plugged it in with ignition off and IAT unplugged
started K manager - then switched on ignition - then started log - temperature reading was 35 degrees C - even with IAT disconnected
plugged IAT back in and made absolutely no difference - no change in temperature reading at all ( surely this cannot be right)

am tempted to read this as there is an issue with my cars wiring - but the fact that we have what looks like both ECU/wiring combinations giving unexpected results makes me think this does not necessarily follow
could be wiring
or
could be issues with both ECU's

what do you think

unfortunately could not test my ECU in another car - only had the ECU for a day

[Hondata]

You cannot test by unplugging the IAT. The ECU will switch to a default value until everything is reset.
The IAT is controlled by hardware in the ECU. There's no software change we could make to change its behavior.

[garz]

ah - i did not know it could switch to some default reading - is there a second thermistor inside the ECU then or does it just use a safe mid/high ambient value to use as a default ?

still concerned we got 2 different readings with 2 different ECU@s - meaning more likely there is an issue with my actual ECU

not suggesting there is anything wrong with the hondata part here by the way - its very clear that this bit is purely the original OEM ECU circuitry

just grateful for any help i can get to track down whether its vehicle wiring side - and i need to dig into that deeper ( time consuming) - or ECU side - and i need a new ECU ( not cheap )

[garz]

may be able to explain the other ECU's behavior

it seems when the same test is done in another car with that ecu - then the system reads 35 deg C in a 15 degrees C ambient - with the IAT sensor disconnected - but does not change when the sensor is plugged back in

however it does change when the ignition is turned off and back on again - so it seem that that ECU is defaulting to 35 degrees C with IAT disconnected - and staying in that default state until the ignition is cycled

my ECU seems to re-recognize the IAT - without the ignition being cycled

most likely they are just different ECU revisions - will check that

i know its not the hondata part driving this - but does that sound right to you ?

[garz]

its been a while - but just to close this out in case anyone is having similar issues

after much testing with different sensors on the same ECU and different ECU's in the same car / sensors( was able to use a stock OEM ECU and wireless OBDII canner to see what its IAT reading was) it is 100% confirmed that the base ECU the Kpro is fitted with has a fault in its IAT reading circuit - which has been causing the above issues

i now need to get the kpro swapped onto a new OEM ECU and the thing remapped to rectify the issue fully as i believe the tuner was fighting the ECU correction for IAT when the dyno tuning was originally done

thanks hondata for the help getting to the bottom of it