Clicking behind driver seat -Resisitor fuel pump relay

[britlude]

yup, the fuel pump has a resistor to slow it up, then the relay short circuits the resistor when the car decides it wants the full pressure..... and, yes, its switched earth (which makes the electronics easier, but can do your head in when sorting thru many logic gates )

quite common for people to just short the resistor connector when their fuel pump is getting tired

so the alternator rebuild is a separate issue?

have you monitored the supply voltage across the relay?

[ozon02]

I monitored the supply voltage as well as the signal from the ECU. i.e. ground. The problem is internal in the ECU. But I don't know what makes it. I can only guess from the cases described on Prime. The fault lies in the voltage.

[Old guy]

I don't think its the vehicle voltage causing the problem otherwise all the other relays would develop similar clicking noises.

Under light load operation the relay should be de energized. Did you confirm that you are getting an intermittent ground connection for the relay at the ECU terminals? If so, I would expect that it is more likely that you are getting a short to ground which is causing the relay to energize rather than the ground switching transistor in the ECU actually energizing. That would require some odd firmware problem that most other people don't have or a short on the board that is energizing the transistor.

Any short to ground at any point on the wire between the relay and the ECU will cause the relay to operate. The definitive way to locate the source of the short would be to de pin the relay control wire from the plug on the ECU and just let it hang. That way it is impossible for the ECU to cause the relay to operate. If the relay continues to click then you have a wiring problem between the ECU and the relay. If the relay stops clicking then the problem would be in the plug or internal to the relay. You would have to pop the cover off the ECU and examine the solder connection between the plug and the board to start looking for a failure point unless you just want to hand over the money for a replacement ECU.

I will note that the generation of micro controllers in use by the NSX are most likely 5 volt CMOS logic. The ECU will have an on board voltage regulator to drop the 12 V car voltage to the 5 V logic voltage and that regulator is most likely a low drop out regulator. That means that the ECU should continue to function just fine down to DC input voltages of around 7 volts. If that was happening there would be lots of other things happening on the car that would make a voltage induced problem obvious - this would include an engine that was seriously misbehaving because the injectors and the coils, which operate at 12 volts, would not be functioning correctly.

[ozon02]

I'm not willing to spend money on a second ECU. I diagnose my case according to the Manual. The control cable from the ECU to the Fuel bypass resistors relay, as you know, is very short and does not pose any risk of abrasion. As I mentioned in the first post, I am sure that the fault comes from the ECU. This is confirmed by connecting the oscilloscope with the recorder directly to the ECU by disconnecting the pin. I am as surprised as Old Guy by the way of solving the problem described on Prime. Old Guy was making his voice heard there. Maybe there is a capacitor in the internal +5V stabilizer system that is failing due to the passage of time?