Electronic – Flyback diode placement for solenoid valve simulation circuit


I am working on a project where I want to fake the presence of a solenoid valve in a control unit on a car so that the control unit does not give an error message.

The output of the control unit is a 12V PWM signal with about 2 kHz.
I measured the solenoid valve with an LCR meter and found a resistance of 5 ohms and an inductance of 10 mH.

I now want to simulate this solenoid valve with a simple coil. Since there are no coils with the exact inductance and DC resistance, I will use a 10 mH coil with 0.5 ohm resistance and a 4.5 ohm resistor in series.

Since I don't have any information on the ECU circuitry, I'm assuming I need a flyback diode for my inductor. Now I am not quite sure where to place it. Parallel to the coil with 0.5 ohm or parallel to the series connection of coil and resistor with 5 ohm?

The aim is that the control unit does not notice that the solenoid valve has been replaced by my coil.

I would be grateful if someone could give me an answer and a short explanation. The control unit costs 2,000€ from the manufacturer… I would be happy if it didn't go up in smoke as soon as I switched it on.

Flyback parallel to R + L
Flyback parallel to L

Best Answer

Your top diagram is a more accurate simulation of a coil with its distributed resistance plus external Schottky diode, but the bottom one might work too and is maybe a bit safer in case of reversal (though there is no guarantee the driver would survive passing several A rather than few hundred mA).

I'm not sure why you would add a diode if you are sure there is no diode in the solenoid you are replacing, but it should do no harm.

Whatever you do, I would check and double/triple check the polarity vs. diode orientation. Then go back and check it again. I once got an axial-lead diode direct from a Taiwan factory that was working perfectly well, but marked backwards, in one of several boxes of 1,000 (discovered during failure analysis of a rare production fallout). That had to have happened after automated testing but before marking.