So I have a mystery. After idling for 20 minutes, a 97 355 throws a flashing slow down light. All three TCUs are green, about 1 year old. All three signals at the output of the TCUs are about 500mV Cat temperature at the body of the cats are 250C or less. Yet, the SDL is flashing.... What can be the reason?
Bad thermocouple/s would be my first guess. If that doesn't fix it then I would check the female connector pins & wiring. And if THAT doesn't do it, then you may need a new fresh EPROM. Do you know what side is throwing the code?
OK, All three TCU outputs to the computer ECE says 500mV, no where high enough to throw a SDL. Bad thermocouple would have caused one of the three TCUs output to go to 4V. But, they don't do that.
If you are only getting .5v out of the SDECUs and the cats are hot the TC are shot or your new SDECUs are all bad. Hard to imagine all three of either would be like that. As you know .5v is the baseline reading at cold start. If it doesn't move up from there after warm up then something's not working. I have known good and working TC that you can borrow if you want to troubleshoot by swapping known working components. Have you watched the outputs real time? If they are intermittent you won't see the spikes on a static check.
I measured 0.4 volt baseline (cold Thermocouple) at the TCU output. At 250F (temp recorded by Mitchell at idle), the Thermocouple output is 4.96 mV which, amplified 100 times by the TCU, gives 0.496 V at the TCU output. So it appears that Mitchell's Thermocouples and the TCU-s are good, unless one of them has intermittent brief spikes which cannot be sensed by a multimeter. On the other hand, I don't think such momentary spikes would trigger the flashing SDL and, even if they do, I think the SDL should go off when the spike is gone. To further test the Thermocouples and the TCU-s, I would increase the engine rpm to, say, 2000 rpm and measure the TCU outputs then. If they all go higher than the measured 0.5 volt at idle, that would re-confirm their correct functioning.
I am guessing that the fault lies in the wiring harness or the car ECU itself.... I guess I need to break into the wiring at the ECU to see what voltage they see. Strange !!!
An SD2 or Leonardo would be nice as it would at least narrow down to exactly which one is causing this. Are you getting any codes at all, even pending to assist with narrowing down which is causing this? I've been fortunate in the past that my slow down light would be accompanied with a code so I knew which of the three was the cause. I also had a 97 355 spider
It is possibly just a bad contact, perhaps at the ECU connector. In the case of M5.2, the SDL is controlled by the ECU and the ECU will trigger it in both cases, incoming TCU voltage below 0.4 volt and above 3.7 volt. Your 0.5 volt at the TCU at idle may be, after things heat up, just 0.2 by the time it reaches the ECU through a couple of weak contacts.
When I first got into figuring out my sdl issue I drove around with the SDECU outputs tapped and leads into the cabin, checking voltages real time. That's how I saw the intermittent spikes on one output. The sdl would come on and stay on even if the output returned to within normal range. In my case it was a faulty thermocouple lead. A slight movement of it in a specific way would create the spike. Even standing at the back of the car and wiggling it wouldn't recreate the problem routinely. I agree Hugh to check OBD codes. You will sometimes have a pending code listed for cat temps, bank 1 or 2.
Mitchell, this is a very good lead for you. When the Thermocouple is disconnected, the TCU output will shoot-up to 4.9 volt (see my bench test on the pic). You possibly have a Thermocouple on the way out or intermittent connection Thermocouple to the TCU. Image Unavailable, Please Login
A small update. I did unplug the TC and the output shoots up to 4.75V I read the pending codes and 1445, 1448, and 1449 are pending. One of them is TC activation for bank 2. I pulled the ECU 88 pin connectors and checked all the grounds and they are all 0.5 ohms to chassis ground on the cable. Three grounds exist on the cable but they are not all connected on the cable. Two of them are tied together on the ECU board, And the flashing SDL comes on immediate when the engine is fired up this morning. Meaning the condition is fake. The engine was stone cold. So what's left is possible bad grounding at the TCU connector and a bad TC causing a spike before it settles down. I am looking for a TC but ... Bob has to ship some to Ricambi first.
I had to replace the thermocouple ECU once. But:: Using a $600 module to protect a $45 thermocouple is S T U P I D ! Some might say you are using a $600 computer to save a $500 CAT, but the math still does not work out.
From what I observed during my bench test, the logic functions are: 1. Send 0.4 volt to the ECU if the TC's voltage is below 4 mV 2. Send 4.9 volt to the ECU (bank shut down) if the TC is open circuit (disconnected) In between, the TCU is a 100 times linear voltage amplifier.
Taking very low resistance measurements with a standard multimeter can be rather inaccurate. To get better readings, you need to first connect the multimeter probes to each other and take the residual Ohms of the multimeter leads and their connections. Then, when taking some low resistance reading, deduct the residual Ohms from the reading recorded. Your 0.5 Ohms readings could actually be zero Ohms as the 0.5 you saw could just be the mentioned residual resistance. For very low resistance measurements, I use a Fluke (17B) multimeter which has built-in compensation for the resistance of its leads so it measures accurately down to 0.1 Ohm.
Do not rule out the O2 sensors. I know it is not widely accepted that they are connected to the SDL problems, but with my car everything was tested and found working (.6v cold, rising when the car warmed up+ restistance tested etc). And I had a SDL from cold as well. So I knew it was not the cat temp. I found out that my front and rear O2 sensors were switched on both sides by an incompetent previous mechanic or so. I switched them back, and apart from some getting going again problems , the SDL went away after driving 10 minutes and since has not returned... So if you have tested all (you tried the 2v battery trick?) , and you still have it from cold, look into the function or grounding of the O2 sensors.
Cat temp/sdl system is a stand alone thing. O2 sensors are not part of the system electrically. It is possible I suppose if you had a faulty o2 sensor that was leading to a rich condition and a hot cat as a result, it might contribute to a sdl. But the sdl is just a response to the Thermocouples and the voltage amplifiers , aka SDECUs.
I know that the cat ecu's is a stand alone system in itself, but I studied and checked my complete system, and everything worked as it should, but still I got a SDL from cold. So when the SDL would have been real, it would not have shown from cold in my case. And I had no codes, stored nor pending. Otherwise I would have found my problem much sooner. Are we 100% sure that the main ECU does not take other parameters into account as well when it comes to the SDL? And only the 0.5-5.0 V signal from the cat ecu's?
Robbe, I will know this answer soon with the SD2. According to the WSM, the SD2 can see CAT temps, CAT protection faults, etc. Once I get familiar with the tool, I will be posting screen shots of the various ECU software displays.
I'd also add that I don't believe it to be a stand alone system. I believe the information is sent to the Motronic and the logic within the Motronic makes the decision to report a fault or shut a bank down.
Well of course that's what happens. By stand alone I meant that there is not involvement of other sensing parameters beyond the Thermocouples It is common knowledge that the main ecu reacts to the SDECU and sends sdl or bank shut down.
