Hondata

Recently I ran into an interesting issue at the Dyno while tuning my latest build (using S300). The issue was related to the indicated RPM in the Dyno vs where the ECU RPM limit was set. The Dyno pull was run right to the overrun limit set by the ECU/

The ECU RPM limit was set to 9057
The Dyno plot below shows the ECU shutting down at ~8800RPM

Day 2 dynorunShare 2 by Baby NSX, on Flickr

Here is what my Rev-limit tab looks like:

RPM limit by Baby NSX, on Flickr

I took a video of the dyno pull and conducted spectral analysis on the audio. It looks to me like the ECU is shutting things down at about 8850 RPM which agrees pretty much with the dyno plot.

here is the video:
https://youtu.be/1_V3xU2Kf5k

If you'd like me to share with you how I conducted the audio analysis let me know - that might be helpful.

Any ideas on what might be happening

What gear was this in? Do you have a datalog showing this?

It was a 4th gear pull, I don't have the logs anymore...

We would need to see the datalogs to be able to get more details.

Spunkster wrote: Mon Jan 07, 2019 2:18 pm We would need to see the datalogs to be able to get more details.

Understood - I'll make sure I grab them next time I'm on the dyno. Just wanted to see if this might be something that has been seen before - seemed very strange to me.

It's most likely due to the gear ratios being set incorrectly. I see your 5th gear rev limiter is set to 8800. If the gear ratios are set wrong, the ecu could think the trans is in 5th gear instead of 4th gear, and use the 8800 rpm rev limit.
Or it could be the bug in the latest version of smanager that I found that Hondata still hasn't done anything about even after multiple proofs of it occurring: even with gear ratios set perfectly, logs show gear glitching up to the next higher gear completely randomly, and sometimes staying in the next higher gear than what it's actually in.
If your gear ratios are set properly, the HACK way to get around this glitch is to purposely set the ratios to slightly lower numbers than their actual values.