It will be interesting to see what happens with this.
The race is on to see if people decode the layout of the late model 2G ECUs or go the EVO VIII ECU swap route. As you'll note some progress has been made on getting EVOflash to support the MH7202 based MD35940X ECUs.
That is really awesome!! Have you tried an Evo8 O2 sensor for the front?
An O2 sensor is an O2 sensor. 0-1v signal and heater circuit. His O2 code is probably due to MAF scaling and Injector scaling. His car needs a tune. I just scaled his injectors down to 450cc. The car is still using a 2G maf, by no means is it anywhere near tuned, just runs. I was ready to get out of the shop and go party. Told the OP he will need to set up something with Scott to get tooned.
That is really awesome!! Have you tried an Evo8 O2 sensor for the front?
This is my next step. My guess is that the impedance may be different for the Evo 8 front 02 and the 2G front 02. I plan on calling up some 02 manufacturers tomorrow to get some info on this. If the Evo 8 02 has a different impedance, I'll go ahead any buy one of those and see if that fixes the issue. Once in a great while, my car is a little slow on entering into closed loop, which I attribute to this 02 issue.
On a brighter note, the car is running very well right now (with the exception of the CEL). I've done about 5-6 flashes since I picked the car up, and knock is now becoming a very rare occurrence as opposed to when I first started running it with the base ROM. I've reached the point now where I need to upgrade my injectors and clutch in order to tune for any more power. I plan on getting some RC Engineering 850-950s sometime this week, and then an ACT 2600 clutch sometime the weak after. If all goes well, I'll take it back down to RRE and see how she does on a few dyno pulls
An O2 sensor is an O2 sensor. 0-1v signal and heater circuit. His O2 code is probably due to MAF scaling and Injector scaling. His car needs a tune. I just scaled his injectors down to 450cc. The car is still using a 2G maf, by no means is it anywhere near tuned, just runs. I was ready to get out of the shop and go party. Told the OP he will need to set up something with Scott to get tooned.
I'll call you tomorrow to talk more about this. I actually did some MAF scaling myself, which seemed to help overall performance, but is not changing the absurdly low voltage im seeing from my front 02. We'll talk.
i have done the evo 8 ecu swap in a 4g64 and the o2 sensors are the same as i have used a 1g turbo o2 on up to a 64 nonturbo o2 sensor and am currently using an innovate o2 simulator off my wideband.. The o2 sensor may however be bad and need to be changed.. also if ## scaling is crazy off u may get a delay entering closed loop which will also raise ## idle and cause idle codes.... Try messing with that scaling -- set it higher than 450 and see if that helps ...
i have done the evo 8 ecu swap in a 4g64 and the o2 sensors are the same as i have used a 1g turbo o2 on up to a 64 nonturbo o2 sensor and am currently using an innovate o2 simulator off my wideband.. The o2 sensor may however be bad and need to be changed.. also if ## scaling is crazy off u may get a delay entering closed loop which will also raise ## idle and cause idle codes.... Try messing with that scaling -- set it higher than 450 and see if that helps ...
Hmmm.....maybe I went the wrong way with the scaling. I calculated that the difference between the Evo 8 MAF and and the 2G MAF to be about 28-30%. The original ROM was set to 357.5. I thought I needed to make it lower, so I went to 260. I'll try 450 and see if that makes any difference.
no dont raise the maf scalings --- i was referring to injector scaling only... for maf size i have 257.7 i attached my maf scalings for u to view... maybe u want to try something close to that and see.... its the mafscalings.jpg file... try my numbers ----- also what is ## wideband reading at just idle??? it ## running way rich or too lean then ## trims are nasty..... GET a program called scan tech it will tell u when ## car is in closed or open loop also -- im using 450 cc blue tops from a 1g and i use 424 injector scaling -- u may need to raise or lower depending on ## trims
no dont raise the maf scalings --- i was referring to injector scaling only... for maf size i have 257.7 i attached my maf scalings for u to view... maybe u want to try something close to that and see.... its the mafscalings.jpg file... try my numbers ----- also what is ## wideband reading at just idle??? it ## running way rich or too lean then ## trims are nasty..... GET a program called scan tech it will tell u when ## car is in closed or open loop also -- im using 450 cc blue tops from a 1g and i use 424 injector scaling -- u may need to raise or lower depending on ## trims
Thanks alot man, I really appreciate your help. I'll try this scaling on my ride home tonight and see if it makes any difference. Unfortunately, I don't have a wideband yet, so I don't know where I'm at during idle.
just remember what section ## working on because one little mistake in numbers causes fuel or timing problems.. Remember to adjust the maf size to 257.7 and for the scaling follow what i posted... Also u will have to play with injector scaling and injector latency to get good idle of course.. But if ## scalings are incorrect for the maf all that goes haywire altogether soo... Post up ## fuel trims from evo scan if u can --
just remember what section ## working on because one little mistake in numbers causes fuel or timing problems.. Remember to adjust the maf size to 257.7 and for the scaling follow what i posted... Also u will have to play with injector scaling and injector latency to get good idle of course.. But if ## scalings are incorrect for the maf all that goes haywire altogether soo... Post up ## fuel trims from evo scan if u can --
Will do. I already had my MAF size set to 260. I'm curious to see if this helps. I'll try it tonight.
Idle is definitely improved. Vacuum at idle is also much better now (20hg as opposed to 14-15hg before). Unfortunately, it appears to have had no effect on the front 02 issue. In fact, the voltage didn't change at all. This leads me to back to my idea that the sensor may be bad. Still throwing codes P0505 (idle issue) and P0125 (too long to enter closed loop). My front 02 is only showing 3 different voltages...0.020, 0.039 and .059. No matter what I do to the MAF scaling, I still get just those three voltages.
Below are some logs of a few partial pulls. I still have the stock injectors and stock clutch, and although I have no boost controller installed, the T-28 seems to pay no attention to the stock boost solenoid and will hit 20psi if i go WOT. So until the injectors and clutch come in, this is the most aggressive pull i want to do. I'm just now starting to get some of the higher rpm knock to go away.
if you are sure the wire is connected properly and is correct wire then i would say the o2 sensor is bad... are you sure you have the heater wires and signal wires hooked up 100% right??? maybe to verify you want to switch the rear o2 pin with the front o2 pin to verify that the ecu is reading the front at all... this is to test the port on the ecu to make sure its working then you know you have a bad o2..
if you dont mind send me your rom i would like to look at it also...
Last edited by Defiant; 11-07-2008 at 06:33 AM.
Reason: "U" is not "you".
if u are sure the wire is connected properly and is correct wire then i would say the o2 sensor is bad... are u sure u have the heater wires and signal wires hooked up 100% right??? maybe to verify u want to switch the rear o2 pin with the front o2 pin to verify that the ecu is reading the front at all... this is to test the port on the ecu to make sure its working then u know u have a bad o2..
Agreed. I plan on checking that out this weekend hopefully.
It will be interesting to see what happens with this.
The race is on to see if people decode the layout of the late model 2G ECUs or go the EVO VIII ECU swap route. As you'll note some progress has been made on getting EVOflash to support the MH7202 based MD35940X ECUs.
I think there are fewer evo8 cars than 1995 eprom ecus. The better thing about the evo8 ecu is that there's a much better logging solution for it than for the slow 1995 ecu, right? I think the price of evo8 ecus will go up unless the DSM h8 ecu reprogramming gets off the ground. I really hope having the evo8 ecu 'swappable' doesn't slow progress of hacking the 1998-99 computers.
____________________________
Matt
dd '90 GST, Holset H1C
I am really excited about this though I have to say that I somehow doubt that there will be enough demand for a wiring harness for these ecus to really make any money. How many evo8 ecus are there? How many are still in evo8s? Maybe on a limited scale of production someone might make something but only if they cost less than the convenience is worth since undoubtedly with a little bit of time someone else will play with the wiring for long enough to let the secret out. If this comes out at $90 like the wiring extensions that come from some vendors I don't see it as worthwhile (though obviously some people do or they wouldn't sell them).
I can't say I like the idea of a proprietary cable at all, I expected that when GSXDNA came back we would have details on modifications but I can live. Maybe something good will come out of this and I can replace a bad experience with a good one.
theres a lot around plus they get rid of them for aem gear and such.. ITs a good ecu and i think you can run upwards of 500 hp on it without a problem.. ITs essentially a standalone and if you compare your costs maybe 200 for ecu, 100 for tactrix cable, free ecuflash, 30 for evoscan, wire yourself or pay for harness a few hundreded (say 300) then you have a standalone for around 630 -- i wired mine myself about 90 or so wires but its well worth it.. although my car is not turbo stock so..
Last edited by Defiant; 11-07-2008 at 06:34 AM.
Reason: "U" is not "you".
I'm pretty sure that there were only about 13,000 evo8s produced in the US, I have no idea how many worldwide but its a bit harder to get parts from overseas, beyond that we need to exclude the number that still have their ecus in them. For $630 total I will just get dsmlink thank you very much. I think your numbers are a bit inflated though. Also it is not essentially a standalone.
Well I spent $110 for an EVO ECU, $100 for a harness extension (Mach V), and $90 for a tactrix cable, for a grand total of $300. I'd say that's a pretty good deal for a standalone tuning option.
Well I spent $110 for an EVO ECU, $100 for a harness extension (Mach V), and $90 for a tactrix cable, for a grand total of $300. I'd say that's a pretty good deal for a standalone tuning option.
Its not a standalone. You still are using a factory ecu with load limitations.
i dont understand that??? why is it not a standalone -- i believe it has nothing to do with limitations to determine if it is a standalone.. The stock ecu in itself is a standalone just not a good one.. the evo 8 ecu is much better and programmable.. still not the best but none the less a standalone as it does not require anything but its sensors to run.. is that wrong? sorry but i am curious why the stock ecus are not standalones
Like an evo8 ecu swap? less talk about the definition of a standalone and more talk about results!!!
Thanks to everyone in RRE who did their part in getting this thing running! Has it been long enough for the ecu to finish updating? How's the fuel trims now that you've changed the maf compensation?
____________________________
Matt
dd '90 GST, Holset H1C
