Page 3 of 3

Re: Hysterisis issue on TPS load gear table

Posted: Tue Jul 23, 2013 10:04 am
by gboezio
EDIT: Disregard these posts go to next post


Removing the RPM checks did not make a difference, It was driving fine, except that it hung in second even tough I was swinging the throttle, same than in 4 th. If I break the speed hystersis, it will shift.

I have tried to use a trans only project to datalog so bit will be thinner, from then the TCU crashed twice, got me stuck in 3 rd gear, then it worked a bit upon reset, then got stuck in 1 st.

I forgot to replace the INI, I'm going for another spin, will post back in an hour, I will use the CAN datalog

In the meantime I'll post the crash datalog.

Re: Hysterisis issue on TPS load gear table

Posted: Tue Jul 23, 2013 11:22 am
by gboezio
Ok I think I have it pretty much figured out.

I have seen a huge improvement because there was no more gear hunting and the truck drove almost perfectly, but the TPS hysteresis never worked, it just acted as an infinite load hysteresis, but while punching the gas on the lower gear the rapid Speed increase on lower gears made me think it was downshifting from the busted TPS hyst, but I tried to set a larger value (20km/h) and it all become clear, the TPS does nothing until just above 50 %, I had WOT set to 55%, I moved it further up to 75%, but the double downshift still happened just above 50%, my TPS hyst was set to 4%, I would check if the TPS hyst could be scaled up like 10X .

I think the TCU crashes that I have experienced came from the power inverter that powers my laptop, it may close a ground loop, don't worry about this, I don't think it comes from your code.

Well I hope that helps and I'm sorry to have induced you in error thinking it worked mint.

Giovanni

Re: Hysterisis issue on TPS load gear table

Posted: Tue Jul 23, 2013 12:12 pm
by gboezio
BINGO !!!!

The TPS hysteresis works, but it's value is scaled up by 10x, I went for a spin and set speed hyst to a large value, and the TPS hyst to 1 (lowest value), it reacted like a 10% hysteresis value and downshifts from delta TPS.

To save you some headache from reading everything I write, what's left to do is :
Find why TPS hyst is scaled up
Fix the translate bin in the INI file for TTC no lock temp limit pictured in an earlier post.

Thanks Lance
Giovanni

Re: Hysterisis issue on TPS load gear table

Posted: Tue Jul 23, 2013 1:59 pm
by Bernard Fife
Giovanni,

I wondered about the TPS threshold scaling, and thought I got it right, but obviously I didn't.

I will post code in a few minutes that scales the TPS threshold properly, and I will fix the INI at the same time. Thanks for your help!

Lance.

Re: Hysterisis issue on TPS load gear table

Posted: Tue Jul 23, 2013 2:08 pm
by Bernard Fife
Giovanni,

Here is the test code:
Monitor_4126g.abs.s19
Test code, uses new INI, see text.
(71.08 KiB) Downloaded 707 times
I *think* this has the right scaling (I saw my mistake once you pointed out the issue; the TPS is used in two differently scaled values in the code, and I picked the wrong one). While you can use the old 4.126 INI with this code, I have fixed the noTCC_temp scaling factor in the new INI here:

http://www.msgpio.com/manuals/mshift/fi ... t_4126.ini

Many thanks for both your help and your patience!

Lance.

Re: Hysterisis issue on TPS load gear table

Posted: Tue Jul 23, 2013 3:32 pm
by gboezio
My pleasure, I love puzzles and to learn stuff.
Beside we both benefit from improvement

Well I went for another ride, it's just wonderful now, I have tried many settings, 10 km/h and 10% hyst is what works best for me, now I have full control of the tranny and engine on the same pedal, MAP is taking care of adjusting the pressure in relation to torque.

Thanks a bunch Lance
Giovanni

Re: Hysterisis issue on TPS load gear table

Posted: Tue Jul 23, 2013 3:48 pm
by Bernard Fife
Giovanni,

Great, I'm glad it works! This test code will be released as 4.127 code in few days if nothing else comes up in the meantime.

Lance.