Need help, Strange problem, 1999 S70 T5

[Fiending]

Apologies if there is an existing thread, please direct me. However I can't find any mention of this issue anywhere.

The car: 1999 S70 T5, b5234t3, M56. Drives great, no problems that I'm aware of, no ETM symptoms or vacuum leaks. odo @ 330,000 km

The problem: The car starts, runs, idles absolutely fine. One day I was warming up the car and getting ready to drive off. Put the handbrake down, step on the brake pedal, clutch in, going for reverse gear, when all of a sudden the car is shaking so badly I shut it off. Key out, back in, start the engine, everything is fine again, until I press the brake pedal.

The problem only begins when the brake pedal is depressed, and will not stop until the car is shut off.

What I know: It certainly feels like a misfire, it makes the engine and the whole car shake violently. The tachometer sometimes stays alive during the fit, you will see it jump to around 1500RPM and then it will either drop to 0, or drop back to idle. It does not reflect the misfiring. It usually doesn't move at all, it just hovers 850RPM like nothing is happening.

I don't have a dice, the only clues the car gives me are an illuminated CEL, P1633 code, and an ETS light that only came about after SEVERAL attempts to recreate the issue, this was initially presenting with no warning lamps.

I borrowed a OBDlink CX from a BMW friend, and managed to get the default app to run. The live monitor for 'absolute throttle position' hovers around 8% while the car is idling, drops to 2% if the car is shut off, displays changes in throttle position if I tap the throttle while the car is running. BUT when the brake is pressed, and the fit starts, it also displays that quick jump to a higher % and then a drop to 0% throttle position, where it stays until the car is restarted.

I don't know if the ETM has been replaced, but if you peek under the intake manifold you see "2001 V70 XC turbo" written on that silver plate, so I'm assuming its been junkyard swapped, which I didn't even know was possible across model years. I haven't pulled it to inspect it yet, I'm in disbelief that it would be the chief problem, I don't have any drivability issues like surging, or loss of power under load. Never once stalled for no reason.

I also plugged the vacuum hose and nipple going from the manifold to the booster, no change.

Anyone had this exact problem before? My next step would be to tackle that xemodex flow chart but from my understanding it sends me straight to "replace the ETM" and while I absolutely love throwing parts at cars and seeing what happens, but from my understanding of ETM replacement options it's a pretty expensive part. Again, if junkyard swaps are possible across a certain spread of model years and engine codes, please direct me to an authoritative source on compatibility, but from what I've read on the forums people disagree vastly on possible options.

[BernieceReynolds]

If the ETM has not been replaced, or the origin is unknown, you should remove the throttle body to check for signs of wear or failure. Replacing the throttle body may be necessary if there is a problem with the internal electronics. If you are unsure about the compatibility of the parts from the junkyard, it is best to buy a genuine throttle body or a throttle body compatible with your S70 T5.

[abscate]

Ok, bot.

The ETS light is telling you there is a throttle body fault , and it is tied to the brake pedal sensor for various control functions. One of which is shutting the throttle off on braking if brake/accelerator pedal conflict. That sensor lives on the brake booster front face On a 1999, and is a 99-00 only part

The new style, contactless ETM is on eBay for <$200 US, see ETM room forum below for part number

[wizechatmgr]

Having a DICE and running the ETM sweep test is likely your best bet in combination with it giving you far more information. The other thing is I believe Volvo had a TSB/Update for the software on some of the ETMs as to which I am unsure.

I don't think I've heard of any ETM failures with the upgraded/hall sensor that weren't the motor on the module giving out on it after oil/gunk intrusion.

Really basic question - are you sure that battery is charged and the voltages are in spec? I know the automatics of that year will refuse to move forward when the voltage goes low - and when it gets really low, refuse to increase engine RPM beyond idle. Then they die while idling.

[Fiending]

Well I got VIDA working on the car and started digging deeper.

Clearing the codes with VIDA instead of 850-OBD2 app seemed to fare better, instead of a salad of codes and an easily repeatable problem we had... only a handful of codes coming back, and a car that was largely drivable. In about 5 hours of driving and looking at live data, we only managed to reproduce the fit twice.

Codes that wont stay away are
ECM-902A communication with throttle unit signal missing
ECM-903F throttle position signal too low
ECM-912A communication with the engine control module ECM signal missing
ECM-E000 Control module communication - internal fault
CDM-DF13 Can-bus h-signal Signal too high

Live data looks normal, everything within reference ranges, voltages uniform. The two times we got the car to go into its fit we noticed the canbus stops communicating in a systemic fashion, its not just the ETM fighting against the car, I have seen about 20 different codes come and go to never return, the theme is always communication. Clutch pedal sensor signal missing, ABS module communication fault, voltage this and that over and under, you name it. Also the car is most definitely not misfiring during the fit, we see the ETM values shoot all over the place before live data stops updating, and its like theres a tug of war for the throttle as the car fights not to stall, and eventually stalls.

ETM test procedure looked great with a max potentiometer deviation of a fraction of a degree. Tested it on a friends 2001 S60 T5 just to see, and his was drawing us pictures on the graph LOL max deviation of 90 degrees.

Drove the 1999 S70 T5 around the neighbourhood, got gas, could not reproduce the problem for about an hour of trying. Go figure. If anyone has some wisdom I would love to hear it, otherwise I'll be studying VIDA and getting really handy with a multimeter.

[matthew1]

I don't know the solution, but ECM-902A comes up a bunch in search.

search.php?keywords=902A&terms=all&auth ... mit=Search

[Fiending]

The service bulletin for ETM issues seems to explain whats going on.

The ETM will remain in a fixed position during communication interruptions, and then the ECM will try to control idle via fuel trim to stay running. It also explains that 903f will "be generated again when brake and accelerator pedal are activated at the same time, e.g. during a test-drive" and will keep happening because "if the ETM detects a loss of communication with the ECM, fault flags will be set in the ETM, however the DTC's in the ECM won't be stored until the communication between ETM and ECM is restored," and that the freeze frame data wont be recorded until communication is restored as well.

This explains the misfire-like running condition, the repeatability with the brake pedal, the rare pedal-box related codes, the freeze frame data saying 0rpm, 12.6v, and -100% fuel trim, and the fact that letting the car sit overnight with cleared codes made a difference.

Everything points towards intermittent interruption of power, and previous threads with similar problems have been solved with 1.)cleaning battery positive terminal, or replacing B+ cable 2.)reseating ECM 3.)replacing main engine relay 4.)replacing either the ECM or the ETM in some cases 5.)wiring issues along CAN being fixed

I'll go through it one by one until the car seems to notice. I have another car to get around in atm, I really want to verify what the fix is as opposed to doing everything at once and then guessing as to what the fix was.