Downshift Tuning

A forum for discussing applications and implementations of the MegaShift transmission controller code for the GPIO from B&G. This can control up to 8-speeds and 6 shift solenoids (plus a 16x9 table for controlling a PWM line pressure valve). It has manual and fully automatic modes (16x9 load x speed table), with under and over rev-limit protection, and full data logging of all inputs and outputs (among many other abilities). A TransStim to test your completed board is also available.
Buickv6
Posts: 12
Joined: Wed Jun 17, 2015 12:55 am

Re: Downshift Tuning

Post by Buickv6 »

Lance,

The issue concerning the downshift appears to be fixed so thanks for that. I now have another question as I need help figuring out why mshift appears to not be shifting at the correct speed according to my shift table. In my log you can see at -

@ 676.009, 31.5 avload (mshift), 99.1 km/hr (mshift) speed, 94.6 km/hr GPS speed the current gear is 3rd when according to shift table 1 it should be 4th.

then @ 694.720 , 27.2 avload (mshift), 107.2 km/hr (mshift) speed, 102.2 km/hr GPS speed mshift changes to 4th gear when it should have already been in 4th gear.

then @ 708.353, 25.3 avload (mshift), 104.6 km/hr (mshift) speed, 100.7 km/hr GPS speed mshift changes back to 3rd when it should stay in 4th according to the shift table 1.

The GPS speed I use is obtained from the QSTARZ BT-Q818XT 10Hz which is the unit recommended by efianalytics for use with tunerstudio and megasquirt. I used the GPS to check that the speed I was seeing from mshift was correct, while there is about a 5 km/hr difference compared to mshift speed I don't believe that this is the problem as the speed bins for 4th in table 1 are much lower at the avload shown in the log.

The SI Lengths are "activated" in can settings and the speed bins for table 1 show km/hr. My shift hysteresis speed 3.1 km/hr and shift hysteresis time 1.5 sec should have already been met, so I don't know why this is. Any ideas on what is going on?

Also do I need to start this in another thread?

Thanks
Attachments
current mshift tune_21.12.41.msq
(57.4 KiB) Downloaded 694 times
2015-06-21_22.12.16.zip
(283.25 KiB) Downloaded 521 times
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: Downshift Tuning

Post by Bernard Fife »

BuickV6,

Thanks for trying the downshift fix in the code, I'm glad it appears to work for you.

Things that can prevent shifts include hysteresis, rev limiting, decel mode, using the 'other' shift table accidentally, and a few other things.

In your case, the target gear stays at 3. This looks suspiciously like the code is using gear shift table 2 instead of gear shift table 1, especially because the shift does eventually happen at about 107kph. The log shows that the table switch is grounded (MShift.4WD = 0). It is in the datalog as "4WD", where:
- 0 (i.e. "logical low") means grounded (use table 2) and also use the speedo adjustment factor (when it is not == 1.000),
- 1 ("logical high") is not grounded (use table 1).

When I use your MSQ on my bench and set the input speed to 99.1 and the load to 31.2, the target gear and current gear both go to 4th as they should according to gear shift table 1. In fact, the code shifts from 3rd to 4th at just over 80kph (and 31.2 load) as you would expect. However, if I ground the 4wd/table switch, the controller shifts to 3rd at 99kph,31.1%, as you are seeing.

So I think you either need to make sure that the switch is wired correctly (or make sure the wire is not grounded if you always want to use table1) or else you can do your tuning on table 2.

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw
Buickv6
Posts: 12
Joined: Wed Jun 17, 2015 12:55 am

Re: Downshift Tuning

Post by Buickv6 »

Lance,

Thanks for looking at this for me, sorry i did not see it was switched to table 2 never looked there as pin 5 (circuit GPI1) has no wire in the GPIO amp seal in that spot and I am not using the switch function. Maybe I stuffed up when I built the circuit. Will try to find out why this is?

Cheers
Buickv6
Posts: 12
Joined: Wed Jun 17, 2015 12:55 am

Re: Downshift Tuning

Post by Buickv6 »

Lance,

Did not have the jumper for GPI1-PE1 installed causing it to default to table 2, thanks again for your help.
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: Downshift Tuning

Post by Bernard Fife »

BuickV6,

No problem at all, I'm glad it was relatively easy to sort out!

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw
Philip Lochner
Posts: 34
Joined: Thu Mar 08, 2012 9:00 am

Re: Downshift Tuning

Post by Philip Lochner »

Buickv6 wrote:Lance,

Loaded the new code to the gpio and then went to load the ini file and i get the following message -

A problem was encountered loading the project mshift : Reported error, invalid ini row in loaded ini (turbo baud), root problem:0
Using the ini from the latest release site, I got this exact same message. Using the ini file from this thread, things went well - in that regard. It seems to me the corrected ini file has not been loaded to the latest release site?
Best regards
Philip
'80 Jaguar XJS 5.3L V12 = MS-2
'74 Jensen Interceptor Mk3 440 V8 = MS-2 + GPIO (GM 4L60e)
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: Downshift Tuning

Post by Bernard Fife »

Philip,

Thanks for spotting this!

The corrected INI is here: http://www.msgpio.com/manuals/mshift/fi ... t_5102.ini (it should have today''s date near the top of the file).

It looks like when I fixed this file I was actually reverting to the version with the error in it, instead of reverting to an even earlier version prior to the error (there was a small change in the ordering in the INI created when cleaning up the INI that TS didn't like).

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw
Post Reply