2007 XC70 ECM121B code when below freezing, 30711554 Topic is solved

[abscate]

The more I think about this, a dangerous thing, the less temperature compensation makes sense.

The MAF measures mass which is how fuel is metered. I don't think the MAF tells the ECU the temperature of the air, just the mass. The temperature compensation is internal to the MAF so that theheatimg current can be converted to a mass via the amount of cooling of the hot film.

Humidity correction is not done. Humid air will will make theMAF over report mass flow by 2% worst case.

[XC70Rider]

I installed a new Bosch 17137 Upstream O2 sensor yesterday. Took her for a 25 mile drive and the it still ran lean with code P0101 appearing every other mile. The MAF rate is still low at 10kg/ hour.

I'm going to install a Bosch 0280750131 ETM I got delivered today. I've read on FCP and other sites that's it's not actually a module so no SW download is needed. Will it still need calibration using VIDA or will the ECM re-learn the ETM while driving?

[mrbrian200]

The calibration it needs is performed by the ECU automatically by simply turning the key to KPII/Run without starting the engine. Leave in in KPII for 60-90 seconds. After about 20 or so seconds in KPII if you listen closely under the hood in that area you'll hear a faint whining sound that comes and goes for ~15-20 seconds and then stops. This is the ECU running it's functional check and calibration sequence. From what I have observed, the ECU appears to perform a 'short/momentary' calibration and functional test every time you start the engine that simply accounts for day to day tolerance/drift of the TPS restive element between fully closed and part throttle. If you leave it in KPII longer without starting the engine it performs a more complex sweep/functional/calibration all the way out to WOT, which is what you want to let it do when installing a new TB.

[XC70Rider]

After installing the new Bosch ETM I did a full calibration as Brian suggested. Had a friend turn it to IGNII for ~2min so could I hear and feel the butterfly valve moving. I then took her for a 48 mile trip and the code didn't appear on this frigid 20f night! She also felt more responsive.

To my suspension it's the nylon gear that was frozen and holding the butterfly valve open for too long. Appears to of been an internal air leakage.

Should I open the OEM ETM and try to repair it?

[mrbrian200]

There is no source for those internal gears that wear. I looked. Once they're worn the TB is not repairable by a DIY. Note, they're not a simple plastic/nylon. They're a teflon infused self lubricating special material (licensed formula) that merely looks like plastic at a glance. The only thing that is potentially replaceable by a DIY inside there is the throttle position sensor (black plastic cover held on by clips that the electrical connector plugs into). And maybe the stepper motor. But that's not what goes bad.

Thinking about it, the high pitched whine that the stepper motor makes suggests a high frequency modulated control signal is being sent to the stepper motor, not a strict 'variable DC'. Could be more complex PWM signal with or without amplitude modulation and the capability to reverse polarity that I have observed when the stepper motor pushes against spring loaded idle stop mechanism to fully close the butterfly valve during the calibration sequences. I would be interested to see what that output to the stepper motor looks like on an oscilloscope.

[abscate]

Added model and TB part number to title for search

$300 for TB , $500 for OEM Volvo Jan2019

Pin assignments on the TB. Looks like the motor is controlled via ECU ground pin 3

[XC70Rider]

I left the OEM ETM outside last night so I could test it while frozen. I'll attach my mutlimeter to pins 1 & 2 and note the resistance readings from closed to WOT. I'll try it again when at room temperature of 70f and see if there's any difference in readings.

If no resistance difference then I'll plug it onto the ETM pigtail and watch the valve opening during calibration. I'll do this also in both freezing and warm temperatures.

[mrbrian200]

In my experience on my own car when the ETA failed, something about having the air load through the throttle plate on a running engine 'aggravates' the worn gears. On a bench test/sweep test with the engine off the throttle plate appears to work perfectly. Problems aren't apparent until you run the car and record live data on a tool. Vida's graphic function where you are effectively saving the data then stop, go back, zoom in and look at at the throttle position under varying conditions in a nice easy time relative graphical format where you'll more likely to spot an obvious problem.
If the TPS were bad, as in dead spots or contamination in the potentiometer sweep, that is one of the times the ECU actually displays a DTC specific to the throttle body and/or TPS. If you haven't seen those DTCs then the TPS is good. The TPS is essentially just like a volume control on an old school stereo before electronic volume controls and will exhibit a completely normal tolerance 'drift' affected by temperature and humidity. It's the nature of the beast. The ECU is an advanced computer system able to track and compensate for this normal tolerance 'drift'.
Normal fixed resistors can drift 1-5% or more depending on the grade/quality of the resistor. Carbon film potentiometers aren't even that precise generally with tolerances starting at around 3-5% for the 'precision' ones with 'standard grade' pots closer to 10% variance, or more. Tolerance is a variation both between one example to the next and also a function of temperature/conditions and electrical current loading. Hence the ECU momentarily flips the the throttle plate completely closed each time pass KPII to start the engine to see where 'zero' is today, because yesterday the TPS might have read a slightly different value at that same throttle plate position. This is completely normal and nothing to worry about.
Contactless sensors like what xemodex is using in their updated ETAs for these cars also have a tolerance, but it's much tighter than physical carbon film pots, usually some small fraction of/less then 1 percent, which depending on the application is tight enough to no longer need to perform constant checks on the thing. And as a general rule the solid state contactless sensors either work perfectly or not at all making diagnostic question 'is the TPS bad?' easy to answer without any grey area.

[EngineeringBloke]

Based on abscates Alldata sheet, the stepper motor is pins 1 and 2 (input only), no point in measuring this input as it is not an output.

There are two independent outputs pin 6 and pin 4 that provide a throttle position dependent voltage when 5v is on pin 5 and 0v ground is on pin 3.

You can test for resistance between pins with no voltages applied to other pins - no need for the 5v on pin 5):
3 and 5 - should be constant - resistance across both pots
3 and 6 - should vary with throttle plate position - output 1
3 and 4 - should vary with throttle plate position - output 2
5 and 6 - alternate, inverted output 1
5 and 4 - alternate, inverted output 2

Move the plate and check but do not touch the ETM or plate with your warm hands as that will affect its temperature and may hide the issue. Gloves?

Checking that the output is progressive with the position of the trottle plate will show that the monitoring system is working as expected, but will not test the gears.

But I would not expect testing of the pots alone to show any temperature effects. Nor do I think this will show the problem. The issue is that as the ETM provides a drive to the stepper motor, the motor moves, gear move, the plate moves, but with slippage in the gears, the plate is not at the expected position. The ETM may be able to provide an adjustment but this will be nonlinear and the ETM will rightly report a CEL.

Also, I agree with mrbrian200 that changing air pressure in the system may be needed, too.

[abscate]

The nice part of the P2 ETB or ETA is the lack of software intervention via VIDA - so no dealer tax to get the new one done. A $300 part does hurt, but its nice when you nail it and know thats the last time you will see that one.