dwell rises and falls at odd times

[durson1]

I've been running my MS2 for a few years now. Running pretty good. Allowing me time to look into smaller issues like this. Running MS2 V3 with I think its v2.89 now, on my 2.0L 8v VW. running a distributor with direct coil control via BIP373. In the log I posted at 147 sec. my dwell climbs from the normal running 2.2ms to 2.9ms, then steps down to the normal 2.2ms. I'm not accelerating. Its not causing any driveability issues. Would just like to know why my dwell rises at what seems to me to be odd times. I changed one of my Tunerstudio gauges to "dwell" to watch it, and was surprised to see how often it rises sometimes shortly after throttle changes.

for the dubbers out there its a 1984 GTI with 3a 2.0L block, 8v GTI head.

[Bernard Fife]

durson1,

Dwell is affected by the rpm (the time available for dwell + spark), the battery voltage, and the acceleration compensation. Your datalog looks about right considering these, I think. The factors are described here: http://www.megamanual.com/mt28.htm#iw

Lance.

[durson1]

Thanks for the response, but I'm wondering why at that point in my log it jumps from 2.2 to 2.9 then 2.2 then 2.9 then steps back down to 2.2. Battery voltage is constant 13.5v, no accel enrichment at that point, not even close to my TPS dot and Map dot thresholds. within 2300 rpm,s. Basically nothing changes that should effect my dwell that I know of. Am I missing something?

[Bernard Fife]

durson1,

The datalog values are neither synchronized nor comprehensive. That is, the values are gathered at slightly different times, and they do not include anywhere near all the samples of information that the processor uses to make decisions and adjustments.

Around the time you mention (147 seconds of the datalog), the voltage is varying between 13.5 and 13.2 Volts. It appears to be at 13.5 at the start of the period we are talking about, and while it does drop, it takes about 0.7 seconds to show on the log. However I suspect a small voltage dip was missed in the datalog. This can have an effect on dwell, though it's hard to say how much because it doesn't look like the data was captured. It could easily be 0.2 msec or more, though.

At the same time, the map goes from 38 to 43 kPa, the tps climbs from 25 to 33, and the rpm starts to climb. These values alone may not be sufficient to trigger your accel enrichment, but there is always a bit of noise superimposed on the signal, and that might cause accel to trigger. Noise is tough to spot, because it's typically so short. I suspect a brief, marginal accel event was not captured but was used by the processor (it would be similar to the one at 164.7 seconds in your log). Your MSQ shows you adding 0.4 msec on an accel event, so that's part of the 0.7 msec increase.

Because of the noise and the limitations of the datalog, the best bet is to look at at least a few similar instances of an event, rather than any one. Also, if you really want to know what's going on, you can try setting the accel dwell adjustment and the battery voltage dwell adjustments to zero, and see if that clears up the dwell events. Then add them back in a little at a time to see what causes the behavior you are interested in.

Lance.

[durson1]

I'll take a look at some other logs, and maybe try the changes you mention. It runs fine. I was just curious as to why. Thanks for the explanation.