So I was reviewing my data logs from this past weekends track day and have noticed this weird rpm spike.



I checked the spacing between the crank angle sensor and the crank angle sensor plate. It is set to 1mm (measured with feeler gauge) which is the minimum recommended distance from the manual (1-2mm). I also check the resistance of the crank angle sensor, and it was within spec.

I wanted to know what else could cause these spikes and if anyone else has noticed the same thing?

Thanks,
Alvaro

 

Did you see the tach needle actually go up at the same time? It could be electrical noise. Have you cleaned all your grounds?

 

To be frank, I don't look to much at my tach so I am not sure if it jumps (to busy looking at the track). But during street driving I have never seen the tach move to indicate any spikes. However I don't think it would respond fast enough anyways, as it only lasts a few hundred milliseconds.

Here's the log screen shot of the spike condition.



What ground would be most important? PFC ground? I have not cleaned the grounds since I installed the PFC a few years ago. I will do that before my next log session.

Anything else?

Have you ever seen this before on any logs Raymond?

Thanks for your help!

 

Chart speed, rpms, TPS, and boost. Maybe clutch slippage when shifting.

 

certainly possible

I have seen this kind of thing but it's usually when letting off the gas... lightweight flywheel especially

 

RFI ,, possibly a shielding issue
are you using resistive plugs and or leads?

 

I don't think it is the clutch slipping considering it even does it on decel. I have updated the chart to circle the multiple instances where I noticed the spikes. You can see I get it on decel, and initial tip in.

I am using stock trailing plugs all the way around (resistive), and 8.5mm MSD wire (40-50ohm per foot).

I will try to add some additional shielding on the crank angle sensor harness, and see if that helps.

So I take it from the responses, this is not a fairly common problem?

 

it is a common issue with some ecu platforms,, not many bother to log and see for themselves
when i suspect rfi i ask for an RPM log,, and look for that sort of signature plot

however sometimes the issue is caused by slack in the gear dive,, or runout of the crank nose
( either axially = endfloat or radially = wobble crank )
or even having the CAS gear drive installed backwards

these can amount to some shift in the pickup on the decel or clutch application
and the change in the pickup affects the algorithm that counts the acceleration between timing events
and so represents on a plot as these small glitches that only appear on decel or clutch

- there was a reason that mazda moved to a true crank based pickup for the FD onwards