Mitsubishi Eclipse 3G Club banner
1 - 20 of 32 Posts

TurtleCatcher

· Registered
Joined
·
23 Posts
Discussion starter · #1 · (Edited)
Hello, I'm new to posting and wasn't totally sure where to post this problem, but I'm in the prep stages of a turbo build so thought this was as good a place as any.

I've swapped in FIC 775 injectors and EVO8 rail/FPR. Currently my LTFT MID and O2 trim at cruise are less than 5%. LTFT Low is maxed at 12.5, and the idle O2 feedback trim is around 10%. So, I've been adding latency to the whole curve to bring my LTFT Low back down, but no matter how much latency I add, it doesn't seem to impact my LTFT low.

Here's an example:
4.69v 5.328ms
7.03v 4.056ms
9.38v 2.784ms
11.72v 1.968ms
14.06v 1.440ms
16.41v 1.200ms
18.75v 0.936ms

LTFT Low:12.5
O2 trim idle: 10.69

So I've progressively added more latency (4 different times) and am now up to the following:
4.69v 6.120 (ECU flash won't take anything bigger)
7.03v 6.120 (ECU flash won't take anything bigger)
9.38v 6.048ms
11.72v 4.248ms
14.06v 3.144ms
16.41v 2.616ms
18.75v 2.040ms

LTFT Low:12.5
O2 trim idle: 9.48

Now I've never seen anybody report latency values anywhere near this high. I've made HUGE increases and seen basically zero impact on my trims. Basically, my injector scaling is the only thing that seems to have a real effect. Scaling is currently at 487 (way too low), but it's the only way I can get the ECU to add enough gas. It's like the latency adjustments aren't doing anything?

I've checked for leaks in the manifold, I can't hear any, and have sprayed starter fluid at various intake connections with no observances. The car actually idles alright, though it starts harder, throttle response is slower, and overall power feels diminished.

I have no other engine mods, and am running the stock ROM 98950010. Another weird thing, I tried to look in the definition file for some clues, but I can't find one that matches this ROM in my ECU flash folder, yet ECU flash still opens the ROM and lets me change it??? I've tried reading from my ECU and confirmed the settings that I flashed to it.

I've searched and searched, but have only found reference to latency multipliers in ECU flash being set to .024 instead of .015 (mine are 0.024), that was on Evom, but that guy's latencies were only about double what they should be, not super high like mine and having no effect on trims. I'm stumped:dunno:. Any ideas?
 
Discussion starter · #2 ·
Was reviewing a log and noticed some air flow values that didn't correlate to TPS or RPM values. If my MAF isn't reading right, that may explain why the ecu is dumping in so much gas at idle via the trims. I'm going to clean my MAF and see if that helps...
 
Discussion starter · #3 ·
Out of curiosity, what kind of airflow values are you guys getting at idle? Currently mine sits at 25. If airflow should generally be higher than this at idle, that would support the idea that the MAF may be reading incorrectly (too low) so the o2 sensor picks up the lean condition and adds fuel by adding trim.
 
Have you tried changing your IPW (main)? If not that could affect your latencies. Latencies are for idle (LTFT low) and injector scaling for WOT and cruise (LTFT mid). You could try running open loop and see if that cures your idle. There's a formula for figuring your new IPW settings.
 
Discussion starter · #6 ·
2000RS-Turbo: Thanks for the offer, I may take you up on that. Especially when I flash on an EVO ROM and when (if?) I actually get the turbo on. Part of the fun is learning this stuff though, so I don't think I'm ready to cry uncle just yet:) I get the feeling you may have a good idea of what's going on here, do you have any ideas I can try?

Jakermeister: Thanks for the idea. I haven't really looked into the IPW tables yet, as I thought those were only active during warm-up, and even then only active under like 500rpm or something. My car seems to idle alright once it starts, it's just that I have to crank longer and give it some gas for a solid start. I'm alright with that for now because I suspect my wacked out trims may be contributing to it. But since you mentioned it, I looked and I don't' even have the IPW tables defined in my definition file, so can't mess with them just yet. Right now I'm mostly concerned with the amount of latency I seem to need. I graphed my data, and the adjustments are slowly lowering the trims, but at this rate I'll be at 6ms @ 14v just to get my 02 trims to zero and I'll still have that 12.5% to address on my LTFT low. I think I need to change my ECU flash latency formula from *0.024 to *0.015 for sure, but I don't think that will be enough. I wonder if the eclipse ROM has a different latency base than the EVO roms and that could be playing a factor? But latency base isn't defined for my ROM either. I have an EVO ROM with the IPW tables and latency base defined, but I didn't plan on flashing that on until I get my wideband installed. Obviously if I do that I'd update the fuel maps, timing maps, and MAF scalings to match my eclipse ROM, but I'd want the wideband to make sure everything is running right. I'm thinking about getting a hex editor and seeing if I can define the IPW tables and latency base for my eclipse ROM? Probably fall flat on my face with that :lol: but who knows until you try:)
 
Ipw certainly has a effect with latencys. To get it to crank better you'll need to mess with the ipw or "initial ipw". There was a short how to on evom about rom disassembly. Just give it a Google. If you don't have a wideband in you probably should.
 
Discussion starter · #10 ·
Thanks for the heads-up about the ROM disassembly information on EVOM, I'm going to take a look at that.

I bought the injectors new from FIC, so they came with a cert. For injector size, I started about 10% under the rating of 775cc, and for latency I started with the values from FIC. My approach was to first go after LTFT mid with scaling, then LTFT low with latencies, then fine tune as needed from there. Initially I had positive LTFT mid and low, so I progressively reduced scaling until the LTFT mid was within 5%. I ended up with a scaling value of 487. Then I went after my positive LTFT lows by adding latency, but that is where I started to questions things, as the latency adjustments had a barely noticeable effect on LTFT low. I suspected a fuel supply issue, but the only thing I changed were injectors and the FPR. My EVO8 FPR was used, so I got a brand new OEM EVO8 FPR, suspecting that my used one may be bad, but nothing changed. So I thought oh well, these are just numbers to get my trims under control, and I continued adding more and more latency. So now I’m bumping against the maximum latency values allowable in ECU flash, and I'm still stuck with LTFT Low pegged at 12.5%, and o2 feedback trim running around 10%.
 
Discussion starter · #11 ·
I did have an idea of what to go for next though. Probably a noobish oversite on my part. My next step is fuel filter replacement. Initially I ruled this out, because trims were around 4% with the stock injectors (no apparent problems), and I didn’t think that swapping the injectors, rail and FPR could impact the fuel filter or fuel pump. And since I’m still NA, my engine isn’t consuming any more fuel than before. So if the filter was fine before, why isn't it now? But what I realized is that even though fuel consumption hasn’t changed, By installing 775cc injectors (about 3x bigger than stock) I’m asking the fuel system to supply fuel at about a 3x faster rate than before, but for about a 3x shorter time duration. So maybe my fuel filter is mostly clogged, but could keep up with a 240cc injector. Now I ask it to flow the same about, but at a 3x faster rate through a 775cc injector, and it just can’t keep up. Since it can’t keep up, not enough gas gets in, o2 sensor says hey I’m lean give me more gas, so my ECU adds a bunch of trim to compensate for the poor flow. At least that is my current theory. I’ll also eventually be putting in a walbro 255. Again, I didn't think I'd need that until I got the turbo on, but now I'm not so sure about that. So for now, I’m going to go after 1) fuel filter, and 2) Walbro fuel pump. And if I still have problems after that….. I'll take it as it comes :)
 
Were the injectors new or used? 487 is way to low for a 775 injector at 40 psi or whatever the stock fpr lets through. No wonder its out of wack. And their not really flowing more fuel then stock. If they were you would never get the car to start. Especially not at 3x more fuel. So i wouldnt say fuel pump or anything in relation to that. If you want to test, get a pressure gauge on the line. I personally think its in the injectors or in the tune.
 
Discussion starter · #13 ·
I bought them new, which is why I don't suspect the injectors are the problem. I agree 487 is way too low for 775's, which I said in my first post. But that's what it took to get ltft mid down, and i suspected I'd have to increase that number after I added latency to address the high ltft low. I think you mis-understood what is was saying about getting more fuel. I agree, the total amount of fuel going in per pulse is the same (roughly) as before. I was just saying that because the injector is 3x bigger, that same amount of fuel would dump in 3x quicker than before. So I tell the ecu that I have 3x bigger injectors now, and it says ok, I I'll only hold them open for 1/3 the old time so that I still get the same amount of fuel (fuel is coming out at 3x the rate). But assuming my electronics are working right, the ecu is trying to tell me that for some reason i'm not getting 3x the flow through these 3x bigger injectors, so its adding trim to compensate. My hope is that a clogged fuel filter is the reason that I'm not getting that 3x flow. I'm getting it changed tomorrow after work so I should have something to report soon ( hopefully good news)
 
I bought them new, which is why I don't suspect the injectors are the problem. I agree 487 is way too low for 775's, which I said in my first post. But that's what it took to get ltft mid down, and i suspected I'd have to increase that number after I added latency to address the high ltft low. I think you mis-understood what is was saying about getting more fuel. I agree, the total amount of fuel going in per pulse is the same (roughly) as before. I was just saying that because the injector is 3x bigger, that same amount of fuel would dump in 3x quicker than before. So I tell the ecu that I have 3x bigger injectors now, and it says ok, I I'll only hold them open for 1/3 the old time so that I still get the same amount of fuel (fuel is coming out at 3x the rate). But assuming my electronics are working right, the ecu is trying to tell me that for some reason i'm not getting 3x the flow through these 3x bigger injectors, so its adding trim to compensate. My hope is that a clogged fuel filter is the reason that I'm not getting that 3x flow. I'm getting it changed tomorrow after work so I should have something to report soon ( hopefully good news)

Alright. I understand where your coming from. I still think the issue is with the injectors. If all was working fine before the swap, it should work now. My reasoning is as follows,

Your ecu thinks you have 487cc injectors. Your ecu is telling them to open up to an amount that would flow through a 487 lcc injector at idle. But inreality its almost 1.8 times as large. Im not familiar with the actual pulse sent from the ecu to the injector is, but the voltage is only to open the injector slightly, by sending the same pulse to an injector 1.8x its size, of course its going to be rich. Hence the trim and o2 feedback.
 
Im not too sure about the evo, but if i recall the evo runs off a variable volt fuel pump. Or something of the sort which allows the fuel pump to change the amount of fuel sent up the rail based on loaf or rpm. Honestly not sure. I used the evo fpr with a wally 255 and didnt have any issues. But with the stock pump it might not be enough fuel to get the fpr to work properly. Just an idea.
 
I believe the evo fuel pump has 2 fuel modes from what I've read. There's the idle/cruise and then there's the WOT mode. Does the car idle fine in open loop? Since you're running more fuel your MAF scaling may need adjusting. You should try switching your rom info into the evo rom and see how it does with the LTFT low (idle). You'll at least be able to mess with the IPW settings as well.
 
I didnt see a mention for boost. I dont get why everyone keeps saying hes running more fuel. For a 14.7 afr, he needs to run the exact same amount of fuel, just with a smaller pulse width.

I ran a wally 255, evo rail and 450cc blue tops dsm injectors and set them up without touching anything other then injector size and latencies. I believe anyways, its been a while.

Are you running the stock intake? Otherwise i dont see a need for maf adjustments. Unless its dirty/dieing
 
Well he could be running more fuel. He doesn't have the wideband installed. Bigger injectors require a smaller ipw. Until he gets a rom with the ipw defined he's probably going to have the problem. You should get the wideband installed asap. It's great to log the car before boost. That way if you get stumped while tuning for boost you can go back and reference what the car did before boost. It helps with idle issues.
 
Discussion starter · #20 ·
Well, the shop didn't have the filter, so I haven't got to test that out yet. I'll swap that out when I install the walbro. From my searching, it looks like GSS-342 is the part number people are using on the 3g's, correct?
 
1 - 20 of 32 Posts