I need to pull off the connectors on the ECU to make sure they aren't corroded and making poor contact before I'm completely sure the ECU itself is the culprit. Here's what's going on....

90 NA vert

The TPS check connector (passenger side near pressure sensor--3 pin) has 12vdc on one pin (main relay connection) and about 4vdc on the others (goes to pin 2O and pin 2P on the ECU). These two pins also connect to the "solenoid relief" and "solenoid switch" coils. The TPS cannot be set properly because I have voltage on all three pins.

I have the following ECU check engine codes coming up....
17 feedback system, possible O2 sensor
18 TPS fixed full open
31 solenoid relief valve
32 Solenoid switch valve

The car will also stall at idle with the AC on (hot day only). I tested and cleaned the BAC and it does work. I just bought a 10K resistor today and will measure the voltage tonight.

What do you think....bad ECU?

Does anyone know where to have them rebuilt or should I just buy a used one (6 mo warranty) from Mazda Recycling or Mazmart?

Do the people that install the performance chips, like MDS, rebuild the computers to meet factory specs or do they just solder in a new chip?

 

Disconnect the tps connector. Do you still have the 4v? Suspect the tps is bad or far out of range.

 

oops!

My 4VDC measurement was wrong! Bad DVM ground. It's actually 12.5 vdc on all three pins (DVM reference now battery neg) with the ignition in the on position. Disconnecting the TPS connector makes no difference. Looking at the TPS 6 pin connector (engine side) I measure the following voltages with the ign ON (TPS is disconnected):

Top Row (left) beginning with
green / red wire (pin 2F on ECU?)= 0.04
center pin = 5.0
far pin Black / green (pin 2G on ECU?)= 1.83

Bottom row beginnning under green / red wire,
0.02
0.02 center pin
5.0 far pin

I also tested the BAC with a 10K ohn resistor across the pins (engine side) and meaured 6.5 vac which fits with other measurements I found on this site. I've had this off recently and it functions OK with 12 vdc applied.

 

I think I can give you some insite on the voltage on the CHECK CONNECTOR. One wire will have 12v on it constantly. What you are seeing from the other two is the switching solenoid coil and the relief solenoid coil. If there is no ground being provided to the coil from the ecu, you will see 12v at the check connector. If there is a ground on the coil of the switching or the relief solenoid , you will see just a small voltage at the CHECK CONNECTOR. Take off the blue and the gray connectors(relief and switching) from the relief and switching solenoids. The voltage will disappear on the CHECK CONNECTOR with the exception of the one wire that is constantly 12v. Once the ground is provided by the ecu the voltage drops from 12v. I only have a 86/87 with the old tps so I cannot help on the voltages on the newer tps. Perhaps Irv, Keith's father can help.

 

Yes, this is the case. The 12vdc that is on the check connector all the time comes from the main relay. The other two will float up to 12vdc unless the ECU pulls it down to ground, thereby actuating the solenoids.

If the ECU is not providing the ground for the solenoids, the ECU is probably bad OR the connections at the ECU connector or TPS connector are corroded. That's where I'll look next. If everything is making good continuity, then the only thing left is the ECU.....correct?

 

I opened the ECU and here's what I've found....

Pin 2O (solenoid switch): the first component on the trace is burned. It's blue and looks like a resistor, but I don't think it is. Maybe it's a protective device of some kind. Hard to read the lettering.

Pin 2P (solenoid relief): the first component in the trace is burned. Same as above.

There are diodes next in series and they check out OK with the DVM.

What are these and does anyone know the value and where to buy them. Any idea of the cause?

I found a place that rebuilds computers for $250...1 day turn around. 800-254-3556. Hate to spend the money if all it takes is a $2 component.