Megasquirt Support Forum (MSEXTRA)

Matt Cramer wrote:JS4 to ground.

Will give a try and report back.

I've installed a 0,1 uf Cap between JS4 and GRND and now, with KOEO and clutch pushed in, I get some variation, take a look on the datalog. With clutch release, everything is stable.

Any idea?

What is the specific data field that is worrying you?

Adc7.

Doesnt remains to 0 when clutch is pushed in. When looking voltage at JS4, its constant.

How about posting a similar log, but use the space-bar to delineate when the clutch pedal is pushed down. Something like single space-bar when pedal is pushed, double bar when pedal is released.

I could if needed. Already done that test and remains 1023 when clutch is released.

The fluctuation when clutch in is only since cap installed. My concern us: why?

Will have to test on the road if it fix the "main" issue: flat shift yriggered without any input (clutch).

What quality is the switch? It may just have a bit more "on" resistance than you would like. The ADC counts look low enough to me that I think it will still be a TTL "low", may be no problem now.

Will give a try once I get some minutes and report back. Thanks

Still having issue, very strange. I did all you guys recommended me. On that datalog, launchcontrol and flatshift was disabled and got a RPM spike. What could cause this and wondering if that might be the same issue since the beggining but didnt catch it in logs.

Did you ever post an MSQ? I browsed through this thread, didn't find one. Regardless, post the current MSQ.

billr wrote:Did you ever post an MSQ? I browsed through this thread, didn't find one. Regardless, post the current MSQ.

Yes I did but removed them since. Here's a new one.

martroy wrote:Still having issue, very strange. I did all you guys recommended me. On that datalog, launchcontrol and flatshift was disabled and got a RPM spike. What could cause this and wondering if that might be the same issue since the beggining but didnt catch it in logs.

ADC is reading reasonably; with launch control locked out I can't see what effect it is having on status codes.

For the RPM spike, try noise filtering - start with tach period rejection and a 35% to 40% percentage setting.

"With "Basic Trigger" it is possible that a noise pulse came in generating a false high RPM which tripped the rev limiter briefly. With the relatively slow speed of datalogs (compared to the engine) you won't always capture that.

James"


With the launch issue resolved, we are back to the above, correct?

Matt Cramer wrote:For the RPM spike, try noise filtering - start with tach period rejection and a 35% to 40% percentage setting.

Will give a try. Is there some change "hardware" I could do instead of doing a change in software?

billr wrote:With the launch issue resolved, we are back to the above, correct?

I guess ! lol. Will try what Matt suggested.

With eight cylinders, a single coil, and wanting to rev to 6000+ you are probably already marginal on dwell/spark time available. Would you consider stepping up to missing-tooth and wasted-spark? That might eliminate the present problems as well as giving you a hotter spark and more accurate spark timing... lot's of "wins" there for you.

billr wrote:With eight cylinders, a single coil, and wanting to rev to 6000+ you are probably already marginal on dwell/spark time available. Would you consider stepping up to missing-tooth and wasted-spark? That might eliminate the present problems as well as giving you a hotter spark and more accurate spark timing... lot's of "wins" there for you.

I guess I'm not the only one that want to rev 6500 with a single coil and MSD ignition box without wasted-spark setup ? Do you know how to setup the noise filter curve? I did what Matts says regarding 30% but not sure about the curve itself.

What exact steps I need to follow to convert to wasted spark a SBF TFI setup ?

thanks.

If using an MSD box, you should be able to hit 6500 RPM no problem. A TFI module on its own would be marginal for that application.

Noise filtering is explained here.

http://www.msextra.com/doc/pdf/html/Meg ... .4-89.html

Matt Cramer wrote:If using an MSD box, you should be able to hit 6500 RPM no problem. A TFI module on its own would be marginal for that application.

Noise filtering is explained here.

http://www.msextra.com/doc/pdf/html/Meg ... .4-89.html

Already read that link but about the curve, do I leave it by default or something I should do?


Thanks