OK, my 95 355 Berlinetta 2.7 DME has never sounded right at open throttle. Today, I got around to trouble shooting it to this. 1. Apply 12V across the vacuum switch and the bypass valve will open. That circuit is verified to work 2. Drive around with a DVM taped to my door and floor the gas and no 12V appears across the bypass vacuum switch. There is continuity from the vacuum switch all the way back to the ECU pin 21, left hand side. The throttle position sensor works because I measured the resistance across the pins (at the ECU connector) while moving the accelerator cable by hand. It changes values from around 1K to 2o+ Kohms, both sides are like that. So, it appears the ECU is not putting out the signal asking for the valve to open. What would cause the ECU to not put out? 1) Bad ECU chip 2) Bad ECU board 3) Bad pins 4) Others?
I'm not sure how your DVM is hooked up. Looking at the wiring diagram for the 2.7, to activate the valve, the ECU provides an earth via pin 9 of the left hand ECU. The ignition key (in RUN) should provide a constant 12 volts to the valve. I'm not sure how you're getting continuity via pin 21. Check for volts on the yellow wire on the valve with the key in RUN.
Sorry, my mistake. I was looking at the rear connector pin number, not the ECU pin. Pin 21 on the LH ECU is correct, but as stated before, the ECU provides an earth.
Yes, I got that. I connect a DVM directly across the two pins to the vacuum switch. The DMV is battery driven so it can float across any voltages. I did verify that one of the pins was 12V when the key is on so there is power. The differential voltage between the two pins was zero and never changed. Which means pin 21 was never driven low by the ECU.
Mitchell did you try swapping engine ecu side to side? I never played much at all with the 2.7 bypass
That's a pity. It does sound like an ECU problem. Do you know if a 2.7 exhaust bypass valve thermocouple/TCU (fitted to USA and Australia cars) can inhibit valve operation if it's producing incorrect signals? I guess that would generate a CEL though.
o Actually they do not speak to each other. There is a connection between the left and right ECUs on the later 95s but it is really just a connection to the cam position sensor which is only on the 1-4 bank of the later 95s. On the early 95s there were cam position sensors on both banks. It is possible that the ECU has a current limiting resistor in series with the connection to ground which is much higher than the resistance of the coil in the solenoid. Otherwise grounding the solenoid would cause a short circuit except for the resistance of the solenoid coil. Measure the resistance across the two terminals of the solenoid (vacuum switch) with the wires disconnected. If it is very low you may not see and voltage drop across the switch when activated. In a solenoid the voltage, V, is related to the time variation of the current, I, through the solenoid as V = L dI/dt. L is the inductance of the coil and dI/dt is the change in current through the solenoid with time. When the solenoid is not grounded the current is zero and there is no voltage difference between the terminals, as you measured. However, when you ground one side with a DC voltage applied to the other terminal, after a short time, the current will reach a constant value and again there would be only a small voltage drop across the terminals if the resistance of the coil is very low and there is a current limiting resistance elsewhere in the circuit. You would only see a significant voltage difference at the moment you ground the switch. That is, there would be a spike which your DMV may be too slow to capture. When you applied 12 V directly across the switch the current may have been much higher than would normally occur, forcing the solenoid to activate where as if there is a current limiting resistor elsewhere in the circuit the current may not be sufficient to operate the valve. SO, it could be a bad vacuum switch. Again, you need to measure the resistance of the solenoid and maybe compare it to that of the air injection switch which I believe is identical.
If it's that high then the ECU probably just grounds the circuit. Might try disconnecting the 55 pin connector and grounding pin 21 of the connector and see if the solenoid operates. That would eliminate everything but the ECU.
JohnK, that is a lot of mental exercise for a switch. Although, I will measure the resistance to rule it out. There is one more thing I want to check out before declaring the ECU dead, its RPM signal. There is a spec in the book for looking at it with an oscilloscope.
If you are talking about the TDC/RPM sensor (also referred to as the crank sensor), I wouldn't bank on it. If either of those sensors on a 95 is bad, or the signal is bad, the engine won't run on the bank with the bad sensor/signal, or would run very badly with backfiring, etc.
OK, Problem identified sort of. I swapped the ECU left to right and the non-functioning pin 21 goes with the ECU to bank 1. Thankfully, it only controls the secondary airpump so it does not matter much. And this is OBD1 therefore there is no test of the secondary air pump like OBD2, no error code. So next is to disassemble the ECU to see if there is something really obvious.
I had an issue with my valve not opening. There was vacuum, hoses held pressure, no issues seemingly but it wouldn't open. Found the pins on on both ECU connections in the cabin needed to be cleaned up and added dialectic grease. It may be you have a bigger issue.
The pins on the cable at both ends have been tended to, new pins. I will open up the box today and see if some visual inspection will give a clue. I will also swap out the ECU chips to see if that helps.
