Super rich plugs fouled

[Dave WM]

Could be, one thing I did NOT try is wiggling the connector when it was acting up.

[Dave WM]

well there you go, the orig ECU worked fine from cold start to full operating temps. I left the replacement bolted in, so the orig was tested by just hanging on the connector, a good test for intermittent connections I suppose. It never missed a beat. I will leave the replacement in and keep the original as a backup.

[Zed Head]

So we don't reaaallly know if the original ECU was bad.  It didn't get warm/hot like it would in extended usage. Can't blame you for being safe, but your spare ECU might only be good for a few miles..  Ignition modules are typically heat sensitive when they go bad.  

[Captain Obvious]

Glad you're at least out of the woods for now. but of course you know you can count on the fact that if it's an intermittent electrical connection somewhere, it WILL return again.

But you can at least stay focused on electrical things and stop worrying about compression, manifold vacuum, fuel pressure, etc. Those kinds of things don't usually snap back and forth between fixed and not fixed. Sometimes, but not usually.

[Dave WM]

I was being an idiot about the fuel pressure confusing vacuum and pressure readings, so all is good now. I hear you about the intermittent issue, I really hate those. I put a few miles on the brand new plugs, looking good (not getting fouled).

[Zed Head]

You saw that the guy on zcar had a bad ECU after all?  http://www.zcar.com/forum/10-70-83-tech-discussion-forum/396890-280z-troubles-again-8.html#post2668378

I just took my bad 76 ignition module apart.  The area under the transistor looks cooked, along with a component or two nearby.  Would be nice to know how these things really fail.  This one would repeatably fail after revving to over ~3000 RPM.  Might be a good test case.  Who knows, there might a $3 fix instead of going to ZX or HEI modules.

[Dave WM]

yea I saw that, glad he got it worked out. The only component that I saw when I had my ign module apart that looked like it could give a problem is a single electrolytic cap, they do go bad can open or short, that could result in damage to a transistor. Other than that I cant see what could go wrong unless there was some kind of unloaded output or some kind of feed back pulse. what ever the case is I suspect the parts could be sourced and replaced.

[Zed Head]

Since this topic is current and because I mention my bad 1978 ECU and how I "fixed" it occasionally, I dug around on the internet for a transistor test and ran it on the two old 1978 ECU transistors.  Which I kept, for future study.

I know very little about electronics.  I remember learning about P's and N's and various other things but never had need to use the information so it has rusted away in the back of my brain.  I'll just put these test results here in case somebody that knows gets bored.

Fluke 115, diode setting.  NEC D411A M74 transistor markings.  Two leads, collector is case.

This procedure, because it sounded authoritative - http://www.vetco.net/blog/?p=184

These results.  Values in parens are what "should" be found.  Seems weird that both transistors would fail at exactly the same time.  Considering manufacturing technology, not too surprising that they fail the same way though.  If these numbers actually mean a fail.  They might be fine.  The + or - is the probe used.  I ran the same tests on a Radio Shack 2N3055 type, NPN and it passed all tests.  Posting for the record, no idea of value, not my league or my game.  Maybe I'll take my ignition module transistor out and test it if there's meaning here.

B+  --- E-  .660, .665 (.45 - .9)
B+ --- C-  .561, .559 (.45 - .9)
E+ --- B- 1.2, 1.2 (OL)
C+ --- B- OL, OL (OL)
C+ --- E- OL, OL, (OL)

[Captain Obvious]

The output transistors used the ECU's are not your traditional NPN transistors. They are what are known as "Darlington transistors" or "Darlington pairs". The Darlington configuration combines two transistors together inside the same package, and the end result is a device with higher than normal gain in the same sized package. The circuit diagram for an NPN Darlington looks like this:

Some Darlingtons also include internal resistors on the base connections of the transistors, and I believe that to be the case for the 2SD411's used in our ECUs. I dug a little on the web and didn't come up with a datasheet for the 2SD411, but I did find the 2SD412 which is also an NPN Darlington which includes base resistors. Here's a snippet from the 2SD412 datasheet that shows how the internal resistors are connected:

Why is all this important? Because the inclusion of those internal resistors can throw off the simple Ohmmeter test of a transistor and give you strange readings. The bottom line is that I think your transistors were fine and the problem you were having was somewhere else. I think that the reverse biased E to B numbers you got were simply reading the voltage drop across those base resistors.

[Captain Obvious]

Forgot two things... First is that the 2N3055 is a single normal transistor and not a Darlington. That would be why it passed your Ohmmeter testing.

Second, here's the datasheet for the 2SD412:
http://datasheet.octopart.com/2SD412-NEC-datasheet-519873.pdf

I wouldn't be surprised if the 2SD411 was similar in most ways to the 412. It's not uncommon for parts in a sequence to be very similar. The most common similarity I've seen is stuff like the collector to emitter rating. For example, the 412 is a max of 150 Volts collector to emitter, and the 411 might only be 100.

[Zed Head]

Oh well, it was fun to do, and cheap.  Good thing I was dumb or I never would have tried it.  Any chance they can test fine cold but fail when current starts flowing?

 

Edit - the main reason I even considered this was because I heard an interesting piece on NPR about how the development of transistors that could handle high voltage, or maybe it was current, allowed the development of electronic ignition.  So I had transistor failure in mind.  Seemed to fit the symptoms, with constant current through the (damaged) transistor causing the injectors to stay open.

 

 

Edited September 11, 2016 by Zed Head

[Captain Obvious]

It is absolutely possible that a component like that will test OK cold, but fail under load. I would highly doubt that type of failure to have occurred on both of them at the same time though. Could also have been a bad solder joint that you rejoined with the mechanical agitation.

How long ago did you do the replacement and did you replace them with the same part number? Doesn't appear that they are made anymore, but I'm sure there are suitable alternatives today.