Issue with LM1815 on old v2.2 board
Page 1 of 1

Author:  turbohull [ Wed May 25, 2016 7:37 pm ]
Post subject:  Issue with LM1815 on old v2.2 board

Hey fellow teamswifters, I need help for a friend of mine.

I have not fiddled with MS related stuff in a long time, and I was asked to help install MS on a GT engine, using the stock wiring harness. When the engine warms up, it looses sync randomly. It becomes worse when accelerating. During warm up the issue is barely noticeable. I am monitoring the lost sync counter and it starts missing only when it's almost fully warmed up. They tried a stock GT distributer, a SOHC 4 teeth Dizzy, it does the same. I went to see the car with my scope and confirmed the VR wiring is ok, not reversed and properly isolated.(no noticeable noise)

The used MS came from another swift. It was (poorly) built with miswired P&H injector drivers from jbperf, and a LM1815 on some breadboard. I rewired the P&H driver and left the LM as it was (didn't feel like digging through the spec sheets that night). Now the question, is there something I need to do on the MCU side? or does the signal coming out of the LM good enough? what values have you guys used around the LM (especially the input circuit and the part that configures the output pulse)?

Author:  turbohull [ Mon Jun 13, 2016 12:03 pm ]
Post subject:  Re: Issue with LM1815 on old v2.2 board

It turned out to be a bad configuration of the LM1815.

first of all, it was set to use the one-shot output with no R/C connected to the R/C config pin. I put the chip in mode 0, disabled one-shot mode, now we had something that kind of worked, but it was missing teeth randomly at some rpms. With the heal of a logicl anlyser, I was able to trace it to the LM1815 output not providing a good enough ground for the MS2 input's pull-up. Previous owner wired the output directly to MS2's mcu, bypassing the opto-isolator in place, but keeping the opto'S pull up circuit. I had a 1.5V-4V signal to MS2, and it would sometimes NOT see the low state with 1.5V. We tried external pull down, lowering input voltage to 1.3 but it was still not good enough (doesn't make sense to install pull-down resistor on a pull-up resistor anyways....). I only wired the LM output to the opto input, and BAM, I had a super clean 0-5V at MS2's mcu.

The absolute maximum input voltage for that i/o on that mcu is 1.7V for a low state. I think it was in some kind of grey zone and it was missing some transitions.

Page 1 of 1 All times are UTC - 5 hours
Powered by phpBB® Forum Software © phpBB Group