Supra Forums banner
1 - 20 of 26 Posts

·
United States of America
Joined
·
1,469 Posts
Discussion Starter · #1 ·
Does anyone, with any datalogging software, experience this?
What could be the causes? noise, timing, detonation, slipping clutch,...

Blue = RPM
Green = A/F
Orange = Boost
Pink = TPS
 

·
United States of America
Joined
·
1,469 Posts
Discussion Starter · #3 ·
I don't notice/hear anyting - miss, ping, slip. No drop in power, that I can tell, either.
 

·
1,150+ rwhp club
Joined
·
4,438 Posts
I'm thinking it's something as simple as changing your plugs or just the signal.

If RPM was really that jagged I'd expect A/F and boost to be a bit more jagged too since they are based off RPM and Load.
 

·
Registered
Joined
·
13,581 Posts
To me that looks like it's just a signal processing problem and not related to actual revs. A good signal filter would be in order.
 

·
United States of America
Joined
·
1,469 Posts
Discussion Starter · #6 ·
Cool, Thanks guys.

I do have a change of plugs to install.

Now how to wire up a good signal filter?
 

·
Kwan Jul Sul Freak
Joined
·
928 Posts
Check the quality of your ground(s). Chassis, ECU, any other electronics that piggyback. I have seen jagged RPM plots like this with my Greddy Emanage and EMU and improving the ground has helped in both cases.

Also, I saw this when plotting data from my PLX Wideband controller. It was fixed by grounding the controller at the same point as the Emanage that was doing the logging.

You may also try a 0.1uF 50V capacitor to filter out some noise. Any 0.1uF 50V ceramic capacitor will work. I would check the grounds first however.
 

·
United States of America
Joined
·
1,469 Posts
Discussion Starter · #8 ·
I will investigate that too. I have noticed the line is smoother when I have my s-afr removed from the wiring. I was starting to think detonation because I had it configured to lean, but the a/f still looked very good.
 

·
Shawn Davis
Joined
·
1,662 Posts
It's signal processing. The three clean signals are all analog and your rpm signal is digital. Most likely your digital to analog conversion isn't accurate enough or is dropping counts.
 

·
United States of America
Joined
·
1,469 Posts

·
Shawn Davis
Joined
·
1,662 Posts
Most likely. You could try a filter as someone else mentioned but I doubt it's a noise issue since your analog singals look good. Placing a capacitor in line with the signal will give you a filtering effect. If you're not familiar with this technique do a quick google search on RC circuit filtering.
 

·
Registered
Joined
·
13,581 Posts
I think pretty much all datalogging software I've ever worked with (none were automotive related though) have built-in settings that can filter that out ...such as P.I.D. contollers (proportional, integral, derivative) where you can increase the "integral" values to reduce those signal fluctuations; in some cases it could be called something like "increasing hysteresis"
 

·
Moderator, l337 M0d3r4t0r
Joined
·
12,263 Posts
I have the same problem with my ZT2... It's all jagged and it's hard to tell exactly what RPM i'm doing. I export it to a excell work sheet and interpolate. I've emailed them servral times about it to try to get them to fix it but they won't respond. I've asked them to put a filter in the software to fix he RPM so it's not all over the place.
 

·
Moderator, l337 M0d3r4t0r
Joined
·
12,263 Posts
here is what mine looks like... Not it only does it under WOT.
 

·
United States of America
Joined
·
1,469 Posts
Discussion Starter · #15 · (Edited)
Hey ZaZZn thanks for posting. Does yours look dramatically worse when you put them all on the same graph? Can you post or email me a log or two for me to compare?

Blue = RPM
Green = A/F
White = Boost
Pink = TPS
Lt.Blue = VF

 

·
Shawn Davis
Joined
·
1,662 Posts
Ah, now it's obvious what the problem is. If things get worse when you collect more data at the same time it's not a filtering issue. It's a data rate issue. Try collecting just rpm with no other data streams and see if it looks better. The pulses from the crank sensor come pretty quickly so you need a fast data acquistion system to get a good reading.
 

·
United States of America
Joined
·
1,469 Posts
Discussion Starter · #17 ·
While it may be a data collection rate issue, I was thinking wondering if it was more, the format the data was viewed.
This is a pic of the same log with discrete data on separate graphs.

While there is still the 400rpm fluctuation in the data, it is exaggerate more in one view than the other.

 

·
Registered
Joined
·
1,602 Posts
It's the logger. Consumer-grade high-resolution counters do not contain any means of ignoring random signals. A max rate of change filter may help but you're not going to find that on any cheap logger.
 

·
United States of America
Joined
·
1,469 Posts
Discussion Starter · #19 ·
So would Goldfinger's filter suggestion be a good workaround?

Goldfinger911 said:
...You may also try a 0.1uF 50V capacitor to filter out some noise. Any 0.1uF 50V ceramic capacitor will work....
 

·
Registered
Joined
·
106 Posts
ZaZZn said:
I have the same problem with my ZT2... It's all jagged and it's hard to tell exactly what RPM i'm doing. I export it to a excell work sheet and interpolate. I've emailed them servral times about it to try to get them to fix it but they won't respond. I've asked them to put a filter in the software to fix he RPM so it's not all over the place.

i just emailed zeitronix and asked to reply to this thread,
All emails i sent were always very quick replied

Same graph here btw
 
1 - 20 of 26 Posts
Top