EvolutionM - Mitsubishi Lancer and Lancer Evolution Community

- ECU Flash (https://www.evolutionm.net/forums/ecu-flash-179/)

- - Evo8 ECU in a 1G DSM in the works! (https://www.evolutionm.net/forums/ecu-flash/436196-evo8-ecu-1g-dsm-works.html)

Quote:

Originally Posted by 0xdead (Post 7331481)

how do you remove the limits? Aside from putting in an extremely long timer value.

Quote:

Originally Posted by the_mork (Post 7326097)

I'm not sure how different the 2g knock sensor is from the 8/9. The 1g sensor is very different. The thing to note though is that on dsmtuners we have a member named knockgoon who has used the evo8 knock sensor on his 2g and it has not solved the phantom knock issues he had. What has worked has been to raise the knock multipliers to around 22 for the low and mid areas. The high range is just fine as it is.

I might as well chime in since I started the MAF Adder thread posted here and I'm running a knock sensor out of an Evo 8.

I picked my knock sensor up used on eBay for $35 shipped. I would have posted in the classifides on here, but I don't that the post count to do that yet.

I really haven't noticed any difference between knock levels of the 2g sensor and the evo sensor. Upping the knock multiplier has been the only thing that has helped, but I'm still seeing knock. I'm actually going to try bumping the low multiplier up even higher than the 21 it's at now. I know it's not real knock because I've run 2g equivalent fuel and timing maps and still have the same knock.

I'm running a tephra mod v5.10 9055 rom.

My biggest problems have been hard cold starts and getting injectors scaled right. I think that I need to go and do the MAF scaling stuff.

I can't wait tell the bugs are worked out. I'm going to do this on my 4g63 elantra car.

Quote:

Originally Posted by wreckleford (Post 7331665)

Look in this thread:

https://www.evolutionm.net/forums/ec...g-guys-14.html

cheers marcus im going down the SD route to see how it goes on the earlies.:)

Found some good info regarding the 1G MAF scaling (assuming this stuff is accurate).

Here is the site:

http://www.hwnd.org/#maf%20compensation

According to this data, this is what might be correct for the 1G and 2G maf scaling???

http://dsm-ezine.com/evom/1g_2g_maf_scaling.bmp

Quote:

Originally Posted by Jack_of_Trades (Post 7337218)

Found some good info regarding the 1G MAF scaling (assuming this stuff is accurate).

Here is the site:

http://www.hwnd.org/#maf%20compensation

Might wanna share all your info your finding with the DSMtuner kids. Just a thought.

Quote:

Originally Posted by Evo_Kid (Post 7337226)

Might wanna share all your info your finding with the DSMtuner kids. Just a thought.

F*** that site. I would if I didn't have to jump through hoops to post in their sections. For that reason alone, I could care less what info that site gets from any findings of mine. If someone wants to do it themselves so they have it, by all means, do it.

The tables above are from the 1G and 95-96 2G, that use a Motorola cpu.
The 98-99 DSM and Evo5-10 use a Hitachi Renesas cpu.
The different input hardware and ADCs in the ecus, most likely will make the Motorola tables incorrect.

Have you tried anything like this yet?
Attachment 138189

Its the 98-99 DSM table with correction for an Evo adder.
The Smoothing Table is used in the Injector PulseWidth calculations so it also needs to be changed. From what I've found the Adder is only used with the Scaling Table, not with the Smoothing.

Is there really a difference from CPU's when they all are hex based code??? I personally used those scaled values with no success.

Quote:

Originally Posted by Jack_of_Trades (Post 7337514)

Is there really a difference from CPU's when they all are hex based code??? I personally used those scaled values with no success.

The 95 eprom and 98-99 flashable ecus have different MAF tables, the only thing different between those years is the ecu.

I think they adjust the screw on the bottom of the MAF at the factory, so different year 2G MAFs may be adjusted slightly more/less to match the ecu.

I had to adjust the first couple values of the Scaling Table a lot, when I put a 98 flashable ecu in my 95. But my MAF has always been quirky like that, even when I had a SAFC.

My flashable ecu is so much smoother than my 95 eprom, seems like they improved everything. I wish you lots of luck with your 1G.

I'm just gonna throw an Evo8 MAF in her for now and stop dealing with all of the variances involved. Then I will just switch over to speed density anyways.

That's what I did, I didn't see any reason to mess with getting the MAF right when I wanted a 3bar map anyway.

Don't worry about this getting to dsmtuners, it is being watched and cross-posted. Most of this stuff hasn't caught a whole lot of attention yet but it will.

I just changed the size and the compensation table as posted on the thread on tuners, and my trims were great when I ran closed loop, and my AFR's were right on when I switched to full time open loop.

I have since switched to speed density.

The REAL problemis the variance in 2G MAF's floating around out there. Like someone mentioned, the 95-96 MAF's may be calibrated differently from the factory than the 97-99's were. Then you add the fact that half of the 2G maf's out there have had their calibration screws F***ed with plus the honeycombs may be modified......now you have NO consistency for eople to work with. Safest bet, Evo8 MAF thats unmodified but I still think the speed density conversion will make this a non-issue.