Fix flaring / flare detection mitigation

[rnd-ash]

Intro to flaring on the 722.6 gearbox

Flaring is the term given when the gearbox will momentarily go into a Neutral like state during a gear change, allowing the input RPM to rise drastically. If the flare is drastic enough (as seen in this video), this will cause the OEM TCM to enter a loss-of-drive situation, as the gear ratio is now completely implausible!

Historically, on the 722.6 box, this issue is linked to bad overlap control sleeves in the hydraulic valve body (Fix kits available by Sonnax, however through testing with the UN52 TCM and tracing the OEM EGS control units, I think I have found that this issue can actually be fixed in software!

Root cause from bad software design on the OEM TCM

When the stock EGS controller commands a shift, it will only manipulate the SPC pressure line, and keep the MPC pressure line at a pre-determined value according to a map it has hard-coded in EEPROM.

The issue with this is that the OEM TCM will see that the box shifted too slowly (Due to worn overlap sleeves), and it will increase the SPC pressure as 'adaptation', without also increasing the MPC pressure line (Which directly feeds SPC).

This has the effect of making the flaring worse and worse over time, as the controller never compensates for the dropping pressure on the MPC pressure line, whos job is to keep the current clutches applied on the gearbox until the SPC pressure line forces the next gear's clutches into place.

The job of keeping MPC and SPC pressure in check is usually performed by the overlap control sleeves, but if they are worn, the can no longer function correctly, and allow for the MPC pressure to drop too quickly.

Through trial and error, I have found that increasing MPC pressure will indeed fix this issue, however it will make the shift a lot firmer without decreasing SPC shift pressure.

Therefore I propose 2 potential solutions:

Solution 1

When the UN52 TCM detects a flare, it will remember next time to increase MPC pressure momentarily (For approx 0.5 seconds), then relax MPC pressure as the shift begins, this momentary bump in MPC pressure should be enough to compensate for the worn out overlap control sleeves

Solution 2

Currently we have a SPC Ramp speed value, so we can also make one for MPC pressure line, so the TCM will try feeding MPC with a constant higher-than-usual pressure, allowing for the compensation of the SPC pressure line, again this will need testing!

[rnd-ash]

Commit d9ef803 begins the implementation of Solution 2 (Trying to ramp MPC along with ramping SPC pressure on the box)