PicoScope 7 Automotive
Available for Windows, Mac, and Linux, the next evolution of our diagnostic scope software is now available.
What sensors would kill an engine if they dropped out?
What sensors would kill an engine if they dropped out?
This is related to a topic on triggering the 4423 scope when the vehicle is unattended and stalls, intermittently. topic22306.html Now that the trigger has been fixed (by avoiding the use of the level dropout trigger), I thought I'd start a new thread about the stalling problem itself.
1997 Audi A8 quattro 4.2 L V8. The engine has 8 fuel injectors and 8 coils on plug. Chasing an intermittent stall at idle. It just drops dead after running for 5 to 30 minutes. No few second stumbling as happens when I pull the fuel pump fuse. The fuel pump is only several months old. The car can be started right up after stalling, every single time. I've caught it in the act of stalling several times now. The crankshaft sensor does not drop out. (That had been my major suspect, and I've had the wires break or short against the exhaust pipe, killing the engine.) The fuel pump voltage and current seem to behave normally until the ECU shuts off the fuel pump at 1 second after the rpm has dropped to zero. Unplugging the Mass Air Flow sensor does not kill the engine. Neither of the camshaft position sensors should kill the engine, according to the service manual.
The only clue I've found so far is that just before stalling, the 3 fuel injectors I've been monitoring (2 on one bank, 1 on the other) start staying open 4 to 5x longer for 2 squirts, then no longer open. Primary ignition coil pulses are normal; these are also killed when the injectors are. (But I haven't checked every coil and every injector.) The engine starts slowing down only after the fuel injector pulses have already started to lengthen. The engine comes to zero rpm within a half second of starting to slow down.
I'm planning to scope the O2 sensors to see if there's a clue there. (Is the engine being killed by too much fuel, for some reason?) What else should I scope?
1997 Audi A8 quattro 4.2 L V8. The engine has 8 fuel injectors and 8 coils on plug. Chasing an intermittent stall at idle. It just drops dead after running for 5 to 30 minutes. No few second stumbling as happens when I pull the fuel pump fuse. The fuel pump is only several months old. The car can be started right up after stalling, every single time. I've caught it in the act of stalling several times now. The crankshaft sensor does not drop out. (That had been my major suspect, and I've had the wires break or short against the exhaust pipe, killing the engine.) The fuel pump voltage and current seem to behave normally until the ECU shuts off the fuel pump at 1 second after the rpm has dropped to zero. Unplugging the Mass Air Flow sensor does not kill the engine. Neither of the camshaft position sensors should kill the engine, according to the service manual.
The only clue I've found so far is that just before stalling, the 3 fuel injectors I've been monitoring (2 on one bank, 1 on the other) start staying open 4 to 5x longer for 2 squirts, then no longer open. Primary ignition coil pulses are normal; these are also killed when the injectors are. (But I haven't checked every coil and every injector.) The engine starts slowing down only after the fuel injector pulses have already started to lengthen. The engine comes to zero rpm within a half second of starting to slow down.
I'm planning to scope the O2 sensors to see if there's a clue there. (Is the engine being killed by too much fuel, for some reason?) What else should I scope?
Re: What sensors would kill an engine if they dropped out?
Hi.
1.ECU crankshaft input
2.ECU camshaft input
3.ECU tacho output
4.ECU to final Output Stage Unit coil control signal output
5.ECU to injector output
6.Unit to coil primary ignition output
.
On the old MOTRONIC, the TACHO-signal output is not on the CAN BUS, but on a separate wire (C10 yellow).
In this way,you can track what caused the engine to stop,a problem with the processor (no input signals), or a problem with the mixture (rich or poor mixture).
All this matters, if the simplest ECU power and ground checks are performed.
If we are talking about the ABZ 4.2 l engine, the optimal connection looks like this:
1.ECU crankshaft input
2.ECU camshaft input
3.ECU tacho output
4.ECU to final Output Stage Unit coil control signal output
5.ECU to injector output
6.Unit to coil primary ignition output
.
On the old MOTRONIC, the TACHO-signal output is not on the CAN BUS, but on a separate wire (C10 yellow).
In this way,you can track what caused the engine to stop,a problem with the processor (no input signals), or a problem with the mixture (rich or poor mixture).
All this matters, if the simplest ECU power and ground checks are performed.
Re: What sensors would kill an engine if they dropped out?
Yes, it's ABZ 4.2 l engine. I only have a 4 channel scope, so I have to try combinations of traces at different times. On the list you sent (thank you!), a few questions and comments -
1.ECU crankshaft input - behaves as expected before, during, and after a stall. The signal isn't dropping out.
2.ECU camshaft input - there are 2, but neither should cause the engine to shut down, according to the service manual. A fault causes the ECU to squirt fuel every revolution instead of once every other rev. What's the purpose of monitoring this signal in this particular intermittent stall problem?
3.ECU tacho output - a new one on me. I see the C10 terminal on the wiring diagram. Is this to see if the ECU is killing its outputs or going dead for just a short time? I actually did get a trace of the engine almost dying but it kicked back in before it died.
4.ECU to final Output Stage Unit coil control signal output - I have monitored the output stage signal that fires the coils (on 1 cylinder so far). I assume this is what you mean by 6. below. If that is working, this (upstream) one must be working, no? But I can see your point, if this one (4.), the input to the power output stage, is working but the output from the POS stage is not, it tells me something. But the injector (5.) also shuts down around the same time, so I suspect that the ECU is shutting both 4. and 5. down, (for some as yet unknown reason).
5.ECU to injector output - monitored 3 at once. As with 6. just below, this signal disappears very shortly after the engine starts to slow down. It first lengthens, the engine starts to slow down, it stays lenthened for 1 more squirt, and then off.
6.Unit to coil primary ignition output disussed under 4.
I'm guessing the O2 sensor signal is too slow to be of much use, but I'm considering monitoring them as well to see what I can learn. I did monitor one by itself, and it had a one time dropout of the usual cycle at one point in 200 seconds. Strangely, the voltages didn't seem right (they were higher than expected), but I had the ground lead of the scope for that channel on an underhood ground instead of on the O2 sensor wiring return itself. I may try that to see if the voltage comes into the expected range. I did not have injectors monitored at that time, though.
Thanks for your comments.
1.ECU crankshaft input - behaves as expected before, during, and after a stall. The signal isn't dropping out.
2.ECU camshaft input - there are 2, but neither should cause the engine to shut down, according to the service manual. A fault causes the ECU to squirt fuel every revolution instead of once every other rev. What's the purpose of monitoring this signal in this particular intermittent stall problem?
3.ECU tacho output - a new one on me. I see the C10 terminal on the wiring diagram. Is this to see if the ECU is killing its outputs or going dead for just a short time? I actually did get a trace of the engine almost dying but it kicked back in before it died.
4.ECU to final Output Stage Unit coil control signal output - I have monitored the output stage signal that fires the coils (on 1 cylinder so far). I assume this is what you mean by 6. below. If that is working, this (upstream) one must be working, no? But I can see your point, if this one (4.), the input to the power output stage, is working but the output from the POS stage is not, it tells me something. But the injector (5.) also shuts down around the same time, so I suspect that the ECU is shutting both 4. and 5. down, (for some as yet unknown reason).
5.ECU to injector output - monitored 3 at once. As with 6. just below, this signal disappears very shortly after the engine starts to slow down. It first lengthens, the engine starts to slow down, it stays lenthened for 1 more squirt, and then off.
6.Unit to coil primary ignition output disussed under 4.
I'm guessing the O2 sensor signal is too slow to be of much use, but I'm considering monitoring them as well to see what I can learn. I did monitor one by itself, and it had a one time dropout of the usual cycle at one point in 200 seconds. Strangely, the voltages didn't seem right (they were higher than expected), but I had the ground lead of the scope for that channel on an underhood ground instead of on the O2 sensor wiring return itself. I may try that to see if the voltage comes into the expected range. I did not have injectors monitored at that time, though.
Thanks for your comments.
Re: What sensors would kill an engine if they dropped out?
1. Input crankshaft sensor should be measured directly on the connector of the ECU.
2. According to the diagram in the manual, the camshaft sensor signal is one (G40),for example, as on an AHC engine .
Maybe two in Your case.
The signal from the sensor must also be measured on the ECU contact.
MOTRONIC reacts differently to situations where there is no sensor signal at all ,or when the signal from the sensor becomes chaotic.
3. The Tacho signal provides information about the work of the processor in the ECU (whether it accepts all inputs and whether they are correct).
4. By connecting to the output signal of the ECU to the Final Output Stage, you control the operation of the intermediate stage in the ECU to the coil control driver .
5. By connecting to the injector signal, you not only see whether the injector is controlled by the ECU driver ,but also see the power supply voltage on the injector.
You can see - injection mode (serial,parallel,paired), time Ti impulse.
6. By Connecting to the ignition coil, you control not only the signal from the UNIT, but also the power supply of the coils .
......................................................................................................................................
And - about fuel.
Turning off the fuel pump power, is not equivalent to emptying the fuel delivery unit.
When replacing the fuel pump, a common error is incorrect assembly of the fuel module.
If this error occurs, the ejection pump does not work.
Fuel stops being pumped from the second section of the fuel tank ,or the pump unit is not filled with fuel from the first section.
Thus, the pump stops feeding fuel as soon as the unit is empty.
This must be checked.
.
Good luck.
2. According to the diagram in the manual, the camshaft sensor signal is one (G40),for example, as on an AHC engine .
Maybe two in Your case.
The signal from the sensor must also be measured on the ECU contact.
MOTRONIC reacts differently to situations where there is no sensor signal at all ,or when the signal from the sensor becomes chaotic.
3. The Tacho signal provides information about the work of the processor in the ECU (whether it accepts all inputs and whether they are correct).
4. By connecting to the output signal of the ECU to the Final Output Stage, you control the operation of the intermediate stage in the ECU to the coil control driver .
5. By connecting to the injector signal, you not only see whether the injector is controlled by the ECU driver ,but also see the power supply voltage on the injector.
You can see - injection mode (serial,parallel,paired), time Ti impulse.
6. By Connecting to the ignition coil, you control not only the signal from the UNIT, but also the power supply of the coils .
......................................................................................................................................
And - about fuel.
Turning off the fuel pump power, is not equivalent to emptying the fuel delivery unit.
When replacing the fuel pump, a common error is incorrect assembly of the fuel module.
If this error occurs, the ejection pump does not work.
Fuel stops being pumped from the second section of the fuel tank ,or the pump unit is not filled with fuel from the first section.
Thus, the pump stops feeding fuel as soon as the unit is empty.
This must be checked.
.
Good luck.
- Attachments
-
Last edited by vasek on Mon Sep 14, 2020 8:32 pm, edited 2 times in total.
Re: What sensors would kill an engine if they dropped out?
Generic knowledge, if ecu increases injection time, it senses drop in rpm which is not expected. Is something else on the engine taking torque off? Alternator, steering pump, ac compressor, water pump, anything to lock engine out? Gearbox? Think outside the box.
Re: What sensors would kill an engine if they dropped out?
I installed the fuel pump correctly, after a couple of attempts. It pumps the tanks down well.
Thanks, vasek and liviu2004. More info, which is just what I was looking for!
Thanks, vasek and liviu2004. More info, which is just what I was looking for!