VR Interface Issues

[LEMCO_Novi_2000]

I’ve been running a MS V2.2 since 02’ with a Wells DR-100 HEI module conditioning the signal from a MSD VR pickup. Recently upon startup I began to get an erratic trigger signal to the MS. With the computer on I could see the virtual tach jumping around all over the place. I use a dual VR distributor so that I can dedicate one pickup to my MSD 6A module and the other to drive the MS via the HEI module so the original single pickup distributor I used back when I had a carb is now a spare. Therefore I used it to troubleshoot this problem by connecting it to the HEI module driving the MS and lifting my injector fuses so my injectors wouldn’t fire as I conducted the test. As I spun the shaft on the rotor end by hand the virtual tach climbed but; since I could only get to about 400 to 600 RPM that way I put my ½” drill on the shaft and ran the speed up. As soon as I got over 800 to 900 the tach dropped out completely. When I pulled the drill trigger all the way in it went up to 2K. Further experimentation revealed that the module wouldn’t trigger anywhere between 850 to 1500 RPM so I went out and bought a BWD brand CBE4P HEI module from Advance because the DR-100 was discontinued. I installed and tested it and though I did get the full range of RPM (as fast as the drill would go at least) I still noticed an occasional random glitch where the tach would suddenly jump up a few hundred RPM. I verified the wiring from the pickup, the parameters in the TS program and since I found nothing wrong I switched out the 32 bit processor with my old 16 bit one and opened the old MS tuning program. Since the trigger signal problem persisted I went to the MS II manual where I remembered seeing the B&G easy VR interface, printed the image, bought the parts, assembled the board, and only then noticed that the circuit was designed for a 5v supply. Since I didn’t like the idea of taking something else off the 5v VCC supply of the MS, I wondered if I could get away with using a small 12vdc to 6vdc board out of a cell phone car charger to power the interface instead. That compelled me to look up the specs for the 4N 25 opto and in the process of doing so I stumbled upon an article someone had written about connecting a resistor with a value between 80 to 100K from pin 6 (the transistor base) of the 4N 25 to ground to add stability to the output of the opto. Upon reading that I checked the MS schematics and found that there isn’t anything connected to pin 6. Although I think there is a possibility that adding a resistor there could make the opto more stable I have to point out that when I scoped the output of the CBE4P I did witness a glitch in the waveform that directly coincided with the glitch of the virtual tach. Since the B&G EZ VR interface I built worked perfectly according to my scope I wired it in, grounded my 5v bench supply with the vehicle and powered it with that just to test it in the vehicle. When it failed to drive the MS I scoped the interface output while it was still connected to the MS and found a nice square wave but; it had an amplitude of only 2v. I thought about using 12v to supply the interface because a 2N 3004 is supposed to be good up to 50v but; I’m afraid I’ll smoke the 4N-25. Below is the link to the article I read on the 4N 25. Any suggestions? The engine doesn't run well enough to datalog and an MSQ is irrelevant because this is not a programming issue.

http://www.technogumbo.com/projects/Lea ... oIsolator/

[LEMCO_Novi_2000]

I think I'm answering my own question here but; since I posted about it I thought I should share what I found with the rest of you. Last night while experimenting with the B&G EZ VR interface I tried it with a 6v supply and even a 12v supply to no avail. I took the output of it through the old syle simulator but to do so I lifted one end of R9, pulled the 555, and soldered a lead to pin 24 to get the interface output into the 4N-25 via the Wing diode and 390 ohm R10 resistor. After it failed to work I began thinking about what the simulator had that the interface board didn't. The simulator uses a 555 to switch a 2N2222 with the emitter tied to ground and only a 100 ohm resistor between the collector and the source voltage terminal (which for some reason I don't understand) is supposed to 9v. Excluding the VCE drop and adding in the 390 ohm R10 resistor: 9v divided by 490 ohms equals about 18ma and the 4N-25 is turned on from 10 to 60ma according to all the spec sheets I looked at. The HEI module has a pull-up resistor that's 470 ohm with a 12v source. 470 plus 390 (R10) is 860 ohms and divided into 12v makes for about 14ma. According to the schematic for the interface the bias or pull-up resistor (whatever you want to call it) in the last transistor stage where the output is pulled off the collector is 4.7K. Add R10 to that and you get 5090 ohms. Divide that into 12v and you only get 2.4ma. No wonder it didn't work. I believe 4.7 K is a typographical error. I think it's supposed to be 470 ohms the same as goes across the HEI module but; then I had no way to verify that but to try it. Although the opto is typically turned on with 5v which is popular with TTL devices the voltage for the input really isn't important. It's the current that matters because it isn't a TTL device it's an LED circuit. The 2N 3904 will handle .2 A @ 40v so I didn't think the 25ma the 470 ohm resistor would allow through it would hurt it. The good news is I tried it with the 470 ohm resistor and a 12v supply (not a 5v supply) and now it works. The bad news is that it's more noisy than the CBE4P HEI module I recently purchased and installed. Rather than experiment with the Ed cap at this stage I'm going to wait for the AC Delco D1906 I ordered new from Ebay for $35 and see if it is any less noisy. After so many years without a glitch, I find it strange that this anomaly should just suddenly become a problem.