Tephra 5.10 based. Use your usual xml.

It is experimental in the sense that the Tephra DEAD variable is moved to 0xffffa800, otherwise all the same stuff has been tested on JDM 8MR with success. If it works as planned, this will mean that the maps are copied from ROM to RAM after reflashing. If it does not work as planned it might need the middle ECU plug (at least on the 3 plug ECUs) to be pulled before the maps will copy from ROM to RAM after a reflash. Or it might keep copying ROM to RAM if 0xffffa800 is getting overwitten, but it does look to be free.

1 byte load is on request ID 00, and used by the PC application for map tracing. I've changed 1 byte load to use the temperature/baro compensated load as this should trace the ignition map more accurately.

Read this thread for info on testing:
https://www.evolutionm.net/forums/ec...-software.html

Download:
banks.myzen.co.uk/96530006-v5.10-mods-dma.zip

 

Very, very nice, John.

OK, I think this will push me to finally convert over to 96530006 from 96940011. I won't be able to test this today, but hope to be able to next weekend. Unfortunately, it's just a super busy time in my life at the moment (got divorced, sold my house, etc, etc).

If someone tests this before then, let us know.

I know that SD isn't patched to this ROM yet, but I know that mrfred has the MAT patch for this ROM, so when I test this, I may then add that patch to get ready for the SD to hopefully come soon. I have my ECU pin connectors and have been running a GM IAT for a while, as well as a GM 3.3bar, but I don't think the boost logging is done for this ROM just yet.

Again, thanks for all of the progress from you and mrfred lately.


Eric

 

I just realized that since I have been out of the loop with the tephramods (still running his original v1 for 96940011) for so long, this live mapping is dealing with the alt maps, correct?

So, I have to build a harness for map switching as well (or contact JackofTrades for one). I have to get all of my ducks in a row here. Looks like there are a ton of changes for me to configure hardware and software-wise.

 

Yes live mapping is with alt maps, plus anything else we choose to put in the 2k block (nothing so far on this ROM, on my own ROM, I also put SD maps in RAM, so they are live mappable but not switchable).

You can still test the comms and the reading/writing of maps without actually running them, but to make use of it you'll need some way of map switching. They way Tephra wrote it, it looks like it cannot be easily switched by just poking a RAM variable, so you will need a switch.

 

l2r99gst, Direct boost is working just fine on the 96530006 ROM. I have been running the Tephra 5.10 version in place of my 0011 ROM for almost a year. (I disabled the Immobilizer, but there is an easy way to get the immobilizer to work.) Let me know if you need me to send my xml with direct boost and some other fun stuff in it.

I may be up for testing this in a few weeks when I get my alt-map switch installed.

 

Hopefully I can get some time this week to test.

Thanks

l2r99gst, I think almost everything that is available for the 96940011 is ported over to this ROM, like fixem I have also been on this for quite a while.

 

Just had some time to play with this a bit....

Logging works as long as there isn't a ROM selected in the "Choose ECU ROM file". If there is a ROM selected it wont record the log to file.

For some reason Load is not being logged.

While logging, writing is confirmed with 2048 bytes verified.

With the engine off I changed the Fuel map values from 14.7 to 15.2 & AFRMAP changed as expected. With the engine running i made the same changes & AFRMAP values did not change.

I made changes to timing & they worked as expected.

Do we still need to flash the ROM with ECUFlash when done? After making changes I downloaded the ROM from the ECU & it was not changed.

I'll have more time to play with this later.

 

I'm going on travel for work on Monday, I'll test this out when I get back next weekend.

 

Thank you very much for your work to release this

 

>Logging works as long as there isn't a ROM selected in the "Choose ECU ROM file". If there is a ROM selected it wont record the log to file.

It should do, but it might store it in the application folder or the desktop. The "choose ECU ROM file" is completely different to the logging part.

>For some reason Load is not being logged.

In Evoscan or in my app? I have 1 byte load on request ID 00, is it map tracing on the ignition map?

>With the engine off I changed the Fuel map values from 14.7 to 15.2 & AFRMAP changed as expected. With the engine running i made the same changes & AFRMAP values did not change.

If running closed loop this will be the case.

>Do we still need to flash the ROM with ECUFlash when done? After making changes I downloaded the ROM from the ECU & it was not changed.

If you make changes in Ecuflash, choose ECU ROM file then click write in my app, then if you flash the same ECU ROM file the changes are permanent.
If you make changes using the fuel and timing tabs in my app they should be kept until you reflash again or kill the battery or disconnect the ECU. You could always mark an area and use my copy/paste buttons.

 

>It should do, but it might store it in the application folder or the desktop. The "choose ECU ROM file" is completely different to the logging part.

When I didn't have a ROM selected it did record the log to file.

>In Evoscan or in my app? I have 1 byte load on request ID 00, is it map tracing on the ignition map?

Derrr, I'm certian I copied my entire MUT table over, I'll change back to the table in you're ROM & make other changes manually.

>If you make changes in Ecuflash, choose ECU ROM file then click write in my app, then if you flash the same ECU ROM file the changes are permanent.
If you make changes using the fuel and timing tabs in my app they should be kept until you reflash again or kill the battery or disconnect the ECU. You could always mark an area and use my copy/paste buttons.

I was using ecuflash then clicking write in your app. I just wasn't sure if I needed to flash with Ecuflash to make things permenent....

I'll try & play with this some more after I get home from work this evening.

 

When you make a change to an alt map in Ecuflash, save it, then write it to RAM using my app, then it is writing to RAM not to flash memory. My RAM maps have been remembered for up to the few months I've left them there in between flashing, ECU removal or battery disconnect, but they are not permanent until you flash your final changes.

The idea is you flash the modified ROM, make your changes in RAM, test them, repeat, when happy you can then flash the final version. This final version will be contained in the latest Ecuflash file you saved and then wrote to the ECU's RAM. Of course if you change the name when you save you need to change the name in my app. However, when you click write in my app, the latest version of the ECU flash file thay you just saved with that name is read in at that point.

 

One thing I would like you to test if you haven't had chance to, is if you flash a slightly altered live map ROM (change anything just to force a flash) over a previous live map ROM that it reverts to the ROM copy of the alt maps - read the fuel and timing map in my app. Before I moved one of Tephra's variables a flash over a previous live map flash would corrupt the RAM fuel and timing maps, I want to make sure that is fixed.

 

I just checked - if you have a ROM file selected you still get a log, but it appears in the same directory as the ROM file, rather than in the PC application folder. Please let me know if yours is different, it shouldn't be.

If you want faster logging, you can turn off the knock tracing and also increase the baud rate in the app and the ECU to match. I logged rates at up to 2500 samples per second. I might make some more MUT tables, including in RAM, so that you can log a smaller group rather than the first 64 items in the MUT table. This could be done without even reflashing the ECU, just by improving the PC app.

 

Anyway to add injector scaling and latency maps to this. Would be great for dialing in some new injectors