Hey Guys,

As you know I implemented a load threshold for the downshift blippers that use NLTS. Basically some users said they hit 100% TPS when downshifting and could I implement a load threshold.

Well that was done - however it introduced a unique bug.

Basically when you shift with NTLS the load is say 200+, however whilst the shift is happening the load drops. This sometimes reduces the load below the threshold.

I need a couple of logs of people NLTS shifting and NLTS downshift blipping.

I need to see how the load behaves when doing the above so I can find a way to fix the bug

Cheers
D>

 

Just set it so once the NLTS is engaged, it stays engaged until the TPS falls below the TPS setting. This would make adjusting the "NLTS load activation point" a useful tool again. If I set my load to 200 and TPS to 95%, if I NLTS shift and even slightly mess up my shift from one gear to the next, the load drops below the 200 mark and then I free-rev my engine.

 

I don't follow Jamie.?

 

My issue is that when NLTS shifting with the load threshold set high (like 200), if you take too long to shift and the load drops below 200, you'll disengage the NLTS function and free-rev the engine and bounce off the rev limiter. This is why I set my NLTS load threshold to 100 right now to limit this from even happening again. Though this does make me vulnerable to activating NLTS by accident sometimes.

I would fix this by having it set so if you exceed the load threshold setting (200 or whatever) and NLTS is activated, it won't disengage until you go below the TPS setting (aka, let your foot off the gas). Basically making the load threshold only useful for activating NLTS, not deactivating it.

 

I was the person who asked tephra to include a load threshold for the NLTS as I was hitting it while down shifting. I tested it a lot and I have mine set to 160. I found that this load is never hit when down shifting, and even if you stay on the NLTS for too long you still will be above this load point.

I have been running the NLTS load threshold for many moths and have never encountered a single problem, it just takes some fine tunning of the value to prevent yourself from running into this "bug"

 

Or we can just 'fix' it and not have to teach ourselves to be so studious about things ;-)

 

Yup that's my intention. But I couldn't get my head around it.

That's a good idea Jamie - using the load as a NLTS enable parameter, but not a disable one.

Sometimes my brain can't think - been reading too much ASM methinks

 

Psh, 5 minutes of reading any coding makes my brain shut off,lol.

 

I still would like to see a log of a NLTS down-shift blip..

just to see what loads you are hitting when you first hit the clutch!

 

ok I have added some code in - so that:

1) when you first shift it records the load at that point in time
2) subsequently when it decides to enable the temporary revlimiter or not it uses that saved load as the decision against the load threshold.
3) when you finally release the clutch it forgets the saved load

so in normal acceleration:
loads will be above 200 when you shift, it will then save 200 and even if you drop to 150 (ie slow slow shifter) then it will still keep the false rev-limiter enabled.

so really all you need todo is set your load threshold below your minimum load for shifting, so probably around the 170 mark..

of course there will be situations where you may think NLTS should kick in but wont - ie shifts at a funny RPM without full boost..

 

Awesome work tephra, and exactly what I was thinking Jack!

Aiming for this to be in V6?

Either way I can grab a couple of logs for your if you give me a better idea of what you are looking for... do you just want me to log 2 byte load, rpm and tps and do a couple of downshift blips? ...if you want a really hardcore log I can log a session when I'm at autobahn in 3 weeks also, but there is obviously a short wait for that...

 

I will probably be releasing a V5.8 which contains a few fixes + code cleanup.

V6 (or V7) will include the PID boost controller if it ever works

 

I need a tester for this - someone that understands the bug and can test to see if it still exists.

Obvious choice would be Dudical26, but I am not sure if he is near an Evo atm

Please post your ROMID when/if you volunteer

 

Tephra - I would love to test this but your right I am not near my Evo

If you wanna wait 5 more weeks I would be happy to though

 

no one else

I don't really want to *just release* it and hope for the best...