I’m writing this with the hope that it may provide useful to some. I have encountered computer problems in the past with my 1985 302 EFI truck. The computer that came with the truck developed an issue where it wouldn’t control the injectors properly, and this forced me to go searching for some replacements. I was able to source some replacement computers and the truck was okay for some time. Recently, as fate would have it, my last working computer (Computer #1) developed an issue where it would no longer prime the fuel pump when starting the truck. The symptom was that the truck would not start, and more specifically one could not hear the fuel pump run for a second before attempting to turn over. I found that I was able to start the truck and run the vehicle if I jumped the fuel pump relay with the test connector (the same connector used to hook up a test light). Next, I replaced my computer with another damaged one (Computer #2) and found that Computer #2 was able to prime the pump.

Further troubleshooting confirmed that Computer #1 was not grounding the fuel pump relay. The images below were collected via use of an oscope.

The yellow signal is battery voltage to the computer, and the blue one is pin 22 (fuel pump relay control)

Yellow: Battery voltage to computer Blue: Fuel pump relay control (pin 22)

Computer #2 behavior looked like this

Yellow: Battery voltage to computer Blue: Fuel pump relay control (pin 22)


The blue line shows that Computer #2 was clearly holding the fuel pump relay ground low for about a second while Computer #1 was not. The next step was to open up the computer and back trace pin 22 to this transistor:

EEC-IV Computer internals showing transistor being tested

I tried to probe the device on Computer #1 and found the following:

Yellow: Battery voltage to computer Blue: Transistor Collector Purple: Transistor Base

Yellow is the battery voltage, blue is the transistor collector (fuel pump relay control pin 22), and purple is the computer control on the transistor base. We see a control voltage here on the base, but no change on the transistor collector. Compare this to the transistor on Computer #2

Yellow: Battery voltage to computer Blue: Transistor Collector Purple: Transistor Base


And one can see that the computer control is turning on the transistor, pulling down the collector, as it should. I tried to identify the transistor part and came up with this:
https://www.st.com/resource/en/datasheet/bd437.pdf
I checked the transistor using a multimeter and there was a difference between the transistor on Computer #1 vs Computer #2, so I replaced the transistor in Computer #1 with the one in Computer #2 and the truck now works again.

 

This is good information. Thanks for sharing.

 

I find it amazing what people like you know. Thanks for posting. I'm still lost looking at that.

Back in my day's at work, troubleshooting issues with Electrical design engineers, I could sometimes follow what he was looking for when he had time to guide me. But most of the time, I just watched in awe. I'm more mechanical.

 

That's some nice work there.

I usually did TTL troubleshooting with a meter. Scopes are nice, but for the purposes of checking transistor function, I always preferred my meter.

Now, I noticed your scope traces indicate your base voltage was jumping to about 5V on the bad unit and probably 0.75V on the good unit. This would be indicative of an open BE junction. Of course, your results bear this out. I expect any general purpose NPN transistor good for about an amp of collector current would have worked.

Congrats on a good fix

 

You are right. One could certainly troubleshoot this with a meter; the scope isn't necessary. Not having a schematic might be a little problematic, but by following the traces from the transistor base back to the integrated circuit, the emitter to the ground, and collector to pin 22 would allow you to figure out what type of transistor and pinout. You could then use a meter's diode test function to figure it out. Fortunately, I had another transistor to compare it with. In that case, one could test both transistors with a meter and see if they are different or the same. Different suggests a problem.

You are also right about the replacement. I'm sure that there are a lot of possible transistors that would work. Current requirements seem to be pretty high in this old stuff. For what it's doing, it seems like my computer get uncomfortably warm, but I suppose that might be normal.