1905 replies to this topic

[ozzimark]

JCviggen, on 03 December 2009 - 05:45 PM, said:

I disagree, that's just 100% load (max measurable airflow for the ECU) which is what you would expect in this case. The problem is the fuel map.

I guess I'll always have trouble wrapping my head around how the ECU does things.

I interpret what you just said as the max airflow the ECU can measure on the 2.75" MAF is apparently around 231g/sec, which is 40% more airflow than the most I've seen with a stock configuration, meaning that if you're doing an engine build that will be doing more than ~330bhp, the MAF cannot accurately read the airflow at WOT.

As such, the fueling on the ECU needs to then be done on the assumption that you will be getting X flow, but the ECU will read Y flow, so for Y flow and some engine RPM, you give an amount of fuel that is actually correct for an assumption of X flow.

Oi. Maybe a standalone isn't such a bad idea

[JCviggen]

The maximum number for the ECU doesn't change when you upgrade MAF diamater. You're just measuring less. The MAF stops accurately measuring airflow yes. Which is why you need to map fuel correctly for your target boost because if you boost less or more the AFR will change since the ECU doesn't know the difference until you drop back below the limit. Tuned cars (especially bigger turbo ones) will always max it out unless you go to 3" MAF. I can't hit 100% load with mine even at 25psi. 95% at 5K and then it drops back to 80s

[ozzimark]

Pffff, man. Now I get what you said some time ago about adding fuel down low affects fueling up top. It's a gigantic balancing game!

Does that also mean that you'll have to tune for a certain environmental condition, and if it isn't the temp, pressure and humidity that it was when the engine was tuned, the fueling and spark will be sub-optimal?

And further than that, if the ECU is reading 100% load and spark is based on load, this means there is a range of throttle position between where the airflow peaks out and WOT where both fueling and spark will be the same.

[JCviggen]

Technically that is correct. Spark can handle a pretty wide range of temps without needing changes but with considerably higher temperatures you can encounter timing pull that wasn't there before. Fueling same story, it's not too picky and you won't see any large swings unless you compare -10C winter to +30C summer.

[ozzimark]

Gtocha, thanks for explaining that So, moral of the story is to go standalone or use the 3" MAF and tune the ECU for it? Temps here do vary from below -10ºC to above 30ºC in the course of a year, and I like to do things the most technically "correct" way as possible. Messing with the fueling in that manner does not strike me as being correct in any sense

Perhaps moving to the 3" will help out Hussein's problems as well, once he gets his TT ECU back in there

[EricF]

Captain Bondo, on 03 December 2009 - 05:52 PM, said:

Agree with both posts above. It just keeps getting leaner and leaner, that just isn't healthy.

That makes me think that make the timing values might actually correct, oddly enough. It's possibly it's not pinging itself to death running 13:1 and leaner afr's at 17psi+ because it's got no timing.

If those timing values are correct, pretty much the whole map needs like, probably 10 degrees added. It would be a LOT faster a little richer and with that timing added in.

As it is if the logs are accurate I'd expect EGT's to be fairly high pre-turbo...

The EGT is my hangup... If he's getting negative timing, his EGTs will be off the charts, especially with that much boost and load.

That's why I think there is something wrong with those numbers. I also think the car would feel pretty anemic even with 25 pounds of boost if it didn't have any timing at all. Heck my car doesn't even feel that fast and I'm running a fixed timing map as posted a couple pages ago

[BlackT5]

I agree, the timing certainly seems off. Then again, i'm sure mine is too

Can you add TPS to the logs?

[Captain Bondo]

EricF, on 03 December 2009 - 07:34 PM, said:

The EGT is my hangup... If he's getting negative timing, his EGTs will be off the charts, especially with that much boost and load.

That's why I think there is something wrong with those numbers. I also think the car would feel pretty anemic even with 25 pounds of boost if it didn't have any timing at all. Heck my car doesn't even feel that fast and I'm running a fixed timing map as posted a couple pages ago

Well, it's difficult to say for certain what post turbo EGT readings translate to IMHO - the effect the turbine has via converting heat energy is hard to really account for, since it varies with pressure ratio, flow rate, etc. But I agree you'd think there'd be some indications of crazy EGTs one way or another.

I wonder though if the simplest answer might be if H could send a sample log to Jan explaining he's trying to tell if his datalogging is working correctly, and was hoping he could confirm/deny if the timing values seem right? I assume Jan doesn't want to just start handing out what his maps are, but maybe he could just confirm if the readings H is getting are in the ballpark or way out to lunch at least?

[lookforjoe]

JCviggen, on 03 December 2009 - 05:28 PM, said:

I don't like your AFR...is this still Janne's ECU? You want a steady ~12 not start pig rich and end way lean.

Point I'm trying to make, why waste your time with an ECU you can't remap. This one is never going to work well with that AFR plot.

Yes, this is Janne's tune. Really, i just wanted to data log to use as a basis for the TT map.

He said 6200 is the last point he can plot fuel. It's after that that it really starts to lean out.

I'll have to email him again.

[lookforjoe]

Captain Bondo, on 03 December 2009 - 08:41 PM, said:

Well, it's difficult to say for certain what post turbo EGT readings translate to IMHO - the effect the turbine has via converting heat energy is hard to really account for, since it varies with pressure ratio, flow rate, etc. But I agree you'd think there'd be some indications of crazy EGTs one way or another.

I wonder though if the simplest answer might be if H could send a sample log to Jan explaining he's trying to tell if his datalogging is working correctly, and was hoping he could confirm/deny if the timing values seem right? I assume Jan doesn't want to just start handing out what his maps are, but maybe he could just confirm if the readings H is getting are in the ballpark or way out to lunch at least?

I've sent him some sample logs, he hasn't really commented directly on any aspect other than to say that as long as AFR's stay under Lambda .96, with EGT's not over 1400ºF (which is the case)then it's OK.

I still haven't got the boost scaled properly - as it's showing max boost of about 1.34 Kg/cm2, which isn't near the 1.65Kg/cm2 the Apexi is registering. The timing data is pulled via OBDII, not sure how I could adjust the scaling for that. I can try adding TPS to the mix. The more channels, the more choppy the data, but maybe that doesn't really matter right now.

JCviggen, on 03 December 2009 - 06:01 PM, said:

The maximum number for the ECU doesn't change when you upgrade MAF diamater. You're just measuring less. The MAF stops accurately measuring airflow yes. Which is why you need to map fuel correctly for your target boost because if you boost less or more the AFR will change since the ECU doesn't know the difference until you drop back below the limit. Tuned cars (especially bigger turbo ones) will always max it out unless you go to 3" MAF. I can't hit 100% load with mine even at 25psi. 95% at 5K and then it drops back to 80s

So you're using a 3" housing with TT? I have a 3" ME7 housing I can play with.

[Captain Bondo]

lookforjoe, on 04 December 2009 - 03:04 AM, said:

I've sent him some sample logs, he hasn't really commented directly on any aspect other than to say that as long as AFR's stay under Lambda .96, with EGT's not over 1400ºF (which is the case)then it's OK.

Does he actually mean post turbo egts? Post turbo egts are often 300 degrees cooler than pre turbo. 1400 post turbo is meltdown city.

[JCviggen]

You're too lean starting 5K onwards...so well before 6200. You really don't want to see above 12.5 AFR.

I am using a 3" MAF housing with the TT, but be advised it takes a lot of work to adapt the spark map since the load figures are completely off (30-35% low) and a MAF rescale function is not (yet) available.

[lookforjoe]

JCviggen, on 04 December 2009 - 07:17 AM, said:

You're too lean starting 5K onwards...so well before 6200. You really don't want to see above 12.5 AFR.

I am using a 3" MAF housing with the TT, but be advised it takes a lot of work to adapt the spark map since the load figures are completely off (30-35% low) and a MAF rescale function is not (yet) available.

There seems to be a fundamental difference in approach - he says Lambda .92 (13.55) is optimum, with +/-1% range for 13.1 or 14. The problem that I see is that I'm running well over 14 above 6500rpm, which would not be helped by more advance, would it?

He said spark is set for 25º >6200.

[Johann]

Like the others say..

Spark makes no sense
Lambda makes no sense
Basically nothing makes sense.

Don't go 100% on what the Innovate says. I have had similar experience assuming a new LC-1 was OK yet it was faulty. The OT-1 simply sucks.

Any other OBD tools to see what the lambda and timing does?

[JCviggen]

lookforjoe, on 04 December 2009 - 03:54 PM, said:

There seems to be a fundamental difference in approach - he says Lambda .92 (13.55) is optimum, with +/-1% range for 13.1 or 14. The problem that I see is that I'm running well over 14 above 6500rpm, which would not be helped by more advance, would it?

He said spark is set for 25º >6200.

13.55 at 1.5 bar is silly lean. He is using a normal amount of timing. 12.0 gives me the best power. If it gets to 12.3-12.5 the timing gets pulled out. Honestly 13.5 is too lean any way you slice it. And because your AFR isn't stable at all through the revs testing with this ECU is quite useless.

[lookforjoe]

Johann, on 04 December 2009 - 04:30 PM, said:

Like the others say..

Spark makes no sense
Lambda makes no sense
Basically nothing makes sense.

Don't go 100% on what the Innovate says. I have had similar experience assuming a new LC-1 was OK yet it was faulty. The OT-1 simply sucks.

Any other OBD tools to see what the lambda and timing does?

OK. Well since the car runs very well overall, something must make sense

The AFR data is pulled from my AEM wideband, not from OBDII. I can try an alternate OBDII tester, but it doesn't have data logging capabilities.

I drove the car with a Volvo fuel pressure tester taped to my windshield today, and the pressure does not deviate under WOT conditions.

Having said that, at two points the fuel cut, I saw it blip on the gauge, and I was nowhere near the RPM limit. I thought it was spark blowout, but the pressure definitely blipped momentarily at the points where the TPS % (last column) dropped.

[lookforjoe]

Somebody had said that WI makes the mix richer. Not my understanding. So, I did data logs with and without. As expected, with makes for a leaner mix, since you're adding unmetered oxygen. The methyl-alcohol component in washer fluid is obviously not enough to make any difference.

[JCviggen]

Still doesn't make sense WI doesn't add any unmetered oxygen..you're not breaking the water down into H and O2 (thank god) and all the air came past the MAF. The one where it runs lean, its also showing much higher boost? In fact, I don't understand why the car would run at all at WOT with such figures. Last night I drove the 850 and although it rips with 12.0, it already starts to stumble around 12.5-12.6 AFR and goes berserk at 13:1 AFR. Trust me, those are not good AFR's.

Perhaps the most puzzling part of this is why are you continuing to mess with this ECU which is never going to give you a clean AFR when you have another you can program yourself to deliver a nice steady AFR away from the silly lean that you're doing now.

I'll bet the house that your car is not accelerating as well as it should with 13.5 AFFR's up top.

[Johann]

JCviggen, on 12 December 2009 - 12:04 PM, said:

Perhaps the most puzzling part of this is why are you continuing to mess with this ECU which is never going to give you a clean AFR when you have another you can program yourself to deliver a nice steady AFR away from the silly lean that you're doing now.

I'll bet the house that your car is not accelerating as well as it should with 13.5 AFFR's up top.

If he insists the car is running well it could mean two things...

Either we are still looking for Joe or the numbers are simply wrong.

Boost at 1.45, AFR well in the 13's and a timing of 6 degree doesn't do anything.
There simply is no grunt.

If you like it this way than I would suggest to lower boost to 0.9 with a lot more timing. Car will probably be a whole lot faster.

Time to pop in that TT and start from scratch.

[Captain Bondo]

Agreed. If those numbers are right, the car needs a complete re-tune. If they're wrong, then the whole exercise is meaningless.
TT and dyno time will show what really works and what doesn't.