RomRaider

I looked at the first attached default.bin. That is a successful ROM dump. Not sure why it appeared not to work. If you post up the logfile.txt I can have a look. I didn't look at the second attachment - I assume this is a duplicate?

It's a 5EAT Hitachi M32R TCU 512KB
SYS ID: A61022
CAL ID: 91A0217400

It probably won't have the ability to change shift pressures via SSM / FreeSSM (so far I've only seen Denso TCUs with that ability).

The list of pointers to the switching tables is at 0x170fc. That should give you a start on finding the shift tables and the specific ones you want to change. Usually there are loads of switching tables for the various drives modes and drive conditions. Might need to do some logging to really nail things down.

Ok awesome! Yeah it just kept on reading data with no end in sight so not sure what happened there?

Ive added the log, It has all the previous attempts. The last one was the one that was successful.

That makes sense, Ive tried changing the available values on FreeSSM and it fails.

Have you got any recommendations for software to define and edit those tables or resources on how to do so. Complete noob at this but would like to learn.

From the log, it looks like comms dropped out occasionally. Might be loose connection, cable drivers, software interrupts... hard to know. The good thing is it worked. Unreliable comms will make flashing risky, so it could still be worth trying to find out what is causing the flaky comms.

The Hitachi ECU/TCUs use a different table format from Denso ECU/TCUs. Most of the software out there (eg: Romraider) only works with the Denso table format. Not sure about Scoobyrom, I suspect it also only works with the Denso table format. The Hitachi table format is described in this thread a while back. One approach is to export the table data to Excel and use Excel to create the shift curves. That step is reasonably straightforward. The tricky bit is the reverse engineering to figure out which table is used in which circumstance (drive mode etc)...

Any changes made to the ROM will also require the ROM's checksum to be updated.

Yeah definitely wouldn't feel comfortable flashing unless the connection is 100%, I'm going to try again with the Openport and see if the results are better. Will only be able to later in the week. Currently the car is in a few pieces while I build a new set of headers for it.

Will give the whole thread a read through again.

Thanks for the info and patience!

Hello, ive managed to get my TCU ROM dump, but i dont know where to go from there, can i get like a guide for dummies?

TCU ID: AD E0 23 60 00
Using hitachi kline

Looks similar to the JDM 06/07 LGT 5EAT. Where to go depends on what your goal is... something not working right or just want to tinker for performance or something else?

Waselon - at this stage, TCU tuning relies on someone digging through the ROM, finding relevant parameters to change (by a combination of reverse engineering and logging), changing the parameters, updating the ROM checksum and then flashing it back. In the longer term, as we get familiar with how the ROMs work, we'll be able to standardise key tables & parameters that become easy to find, similar to how it currently operates for ECUs. But that requires some folks keen enough to spend the time on the journey with me and others...

Let me know if you want to volunteer!

I want to tune line pressure to improve shift speeds. Other way would be buying a reprograming kit, but that involves getting soaked in ATF so i thought i would try this first.



I did some decoding with IDA, got some functions and data... but again i have no idea of what to do with all that as i have limited knowledge and no experience in reverse engineering.

There is a way to adjust line pressure across the board (and/or each clutch/brake) without flashing. See around p18 of this thread.

This relies on the TCU ROM having this capability. You can find out by trying (no harm done if it doesn't have the capability).

SSM has only 4wd pressure correction for my tcu, and it is not permanent afaik?

FreeSSM doesn’t support the changes described on p18 of this thread.

I suspect SSM proper also doesn’t support it because I couldn’t find any mention in the SSM-III manual.

Try the extra menu item in v0.28 of FastECU from the repo and see if you are able to change the values.

I can also look in your ROM to see whether it has the capability, but trying v0.28 will be faster.

All corrections in RR logger read 0
And additional menu to set params doesnt pop up

Hmmm... interesting.

Waselon's ROM is Hitachi. Had a quick look at it. It does not have the pressure adjustments at the same SSM offsets as I have so far seen in Denso ROMs. However, it looks like there *are* some sort of pressure adjustments at a different SSM offset. But tracking it all down and how it works will take a long time...

STI TG5 2011

An update: I've now worked out most of the process that controls the brake / clutch pressure profiles during gear changes to get profiles like the attached pic (this is from a Nissan AT similar to the 5EAT). It's very complicated. Each brake / clutch can be in one of 34 different states, and there are different calculations for each state, different pathways through the states and different values used for each particular gear change. Many of these states are abnormal or error states of various types. This results in a multitude of different tables.

In each state the basic approach is to calculate a target pressure (for that state), and the number of steps to reach that target pressure. Then pressure is increased accordingly. Therefore, the simplest approach to getting faster changes would be to reduce the number of steps. This should essentially compress the time axis of the below charts.