Dialing in LH2.4 on the General Leif and dyno tuning basics?

Next Tuesday, Marc and I are going to try and dial in the General Leif on the dyno(Dynojet). I have been on the dyno before but never played with tuning on it. I have been "street" tuning after driving the car on the track(without a passenger seat, so drive, pull over, change and drive again), but it went wacko on me at the Pacific race earlier this month and didn't respond properly when I tried to lean things out. This past weekend I read the fault codes and found that it had stored an (internal?) ECM fault that may have come up during flashing, so I wonder if that had something to do with it maybe being in a limp mode? It would have turned the check engine light on if we had one.

Yes, I know, you will say ditch LH2.4. To that I say, once decently set up, it's been fine for us for the past 8 years we've been racing it!

SETUP:
-LH2.4 with an 012 MAF
-960 throttle body on a 90? adapter elbow with a B21F intake manifold with NO idle motor. [Switched to an 850 non-turbo throttle body with the 960 throttle shaft and new seals on both sides.]
-Orange 850T injectors
-B230 bottom end bored .030" over with 16V flycuts in b230F pistons
-405 head shaved .123"/3mm with mild(NO?) port work, stiffer springs, modified/opened up combustion chambers with a single groove on the larger quench pad.
-ENEM K13 camshaft
-"Stock" Elring headgasket.
-Combustion chambers were roughly measured as around 44-45CC which put our static compression ratio between 10.5:1 and 11:1 depending on what numbers I use for other information.
-Cranking compression with the K13 camshaft's intake clearances set to .015" is between 193-210psi(#4 is a bit low).

I did a track day in it a couple months ago and played around with the maps and fuel injector constants during the sessions. I got it to be pretty decent between 12.5-13:1 at WOT by the end of the day with Lambda feedback turned off. I can't hear detonation from the engine inside the car with the current engine mounts, so I put the ignition timing at what I felt was pretty safe based on what I've done in my street car and ran it without issue(both for the track day and the race).

After the track day I decided to bump up the IC1 one notch from 90 to 91, made some minor tweaks to the main fuel map and added some WOT switch fueling to try and help flatten the AFR curve for the next time we went out. I then turned on Lamdba feedback because usually these things lean out a little on me over time in my street car and then we went racing. It was too rich at WOT and would dip below 11:1 enough that the wideband wouldn't read above 4500rpm. I could at least let off the throttle and get it to start reading again, though.

For Sunday's race I dropped the IC1 back to 90 but when racing it was even richer! I couldn't get the wideband to read ANYTHING it was so rich, even at part throttle. I did NOT reset the computer after I experienced this, and later found it had a fault code suggesting an ECM or programming error(maybe for checksum?). I should have tried resetting the computer during the race to see if that would have an affect but didn't think about it. I have not since driven the car to see if it's any better. There's a possibility that something else amiss is going on, but I haven't found anything. I will try and drive it this weekend on the street to see if it's acting more like it did at the track day before going to the dyno(fingers crossed that it is!).

CURRENT PROBLEMS:
- 1. Too rich, which should be a generally easy fix.
- 2. Unstable idle once warmed up. When cold and warming up, I can gently bring the engine to an idle speed(1000-1100rpm) and it will stay right around there pretty steady. Once it's warmed up though, the AFR will cycle from 14-18:1 and the idle will bounce up and down accordingly! I have not been able to figure this out. The car has never done this before, but it has ever since we went to running the 960 MAF and throttle body. This has also been with a different B21F intake manifold and different engine than what we had run previously. I checked for a vacuum leak with brake cleaner but it was inconclusive. We should smoke test it, but otherwise I'm at a loss and open to suggestions! [There was a vacuum leak through the throttle shaft on the 960 throttle body and we replaced it with an 850 non-turbo throttle body using the 960's shaft for proper TPS clocking and put new seals on it. Problem fixed!]

GOALS:
- 1. Dial in the injector constants so the car is using most of the map. I mostly only play with Injector Constant 1 because when I've adjusted IC4 it moves where the ECU is pulling data from on the load scale.
- 2. Dial in the fuel map and WOT switch to get an even 12.5-13.0:1 or somewhere even closer in between there. Ideally I think I want to be around 12.8-13:1 for decent fuel mileage.
- 3. Play with camshaft timing to see where it likes it best for power between 3,000-6,500rpm and see how hard it falls off above that for when we need to rev closer to 7,000.
- 4. Recheck fueling then move on to play with ignition timing. HOW?! I can easily bump up the whole map a couple notches at a time, but should I just bump little sections of the map and do a pull, listening for knock, then move onto another section of the map and check there, rinsing and repeating until I've gone through almost the whole map by 2? and then start over again?
- 5. Recheck fueling once timing is dialed in.

Any thoughts on what's going on and anything more specific about what to do when on the dyno?

Thanks in advance!

TL, DR:
- 1. We have an endurance race car running LH2.4 without an idle motor, with 960MAF and TB. What typically causes the idle AFR to bounce? [a vacuum leak! Throttle body in this circumstance, see above.]
- 2. I'm going to do my first dyno tuning session, how should I go about dialing in the ignition timing? [Dyno operator suggested dialing in the fuel first, then drop the whole map 2 degrees and see how it responds. Read on!]

How long does it take to make a change on the LH ECUs, apply it, and be able to run the car on the dyno?
Can you adjust values and have them update immediately?

Basically immediately/yes. For the ignition, it's change and load right away. For the fuel, it's change, click save, then load. I do turn the car off before loading each change as well, just to be safe.

I have an Ostrich emulator on both the fuel and the EZK box. I would only be able to change one at a time but it doesn't take long to switch between the two(1 minute, tops?). Or, I could bring a second laptop and do both at the same time, but I don't think that's really necessary.

Mine did some really wacky stuff when I flipped to the 012, but I found I'd entered but not saved the values for the switch points. Once I changed those it drove completely fine, other than needing the fueling table completely reworked.

I've found on both the 140 and the wagon, the old addage of run the timing until it pings, then back off really didn't work well. The 140 on the dyno showed no real gain but the driveability dropped when timing was too high. The wagon loses spool when it's too high, even if it's not pinging.

One suggestion as well, for the spark side. If the shop has one, piggyback a scope onto the knock sensor signal into the EZK. That way you can see knock rather than trying to listen to it. We used Ryan's Snap-On, worked out pretty well.

I know this isn't the same but my microsquirt was doing some crazy Accel enrichment at idle because I could only get enough air to idle by adjusting the throttle plate open a little with the cable which made the switch unstable. I increased the Accel actuation percentage and fixed the problem, so maybe take a look at your tps adjustment.

For the idle, are you getting enough airflow at idle to get a decent MAF reading? Maybe check the MAF output signal and ensure its making sense at idle/low engine speed. I also had on MS lots of problems at idle with big injectors being at such a small duty cycle that there would be wide variation in the actual fuel delivered...it was hard to compensate for. Check the duty cycle/open time of the injectors and make sure its not too small to be reasonable.

Then again, isn't this a race car? Just set the idle higher and both the above problems should smooth out

gsellstr said:

Mine did some really wacky stuff when I flipped to the 012, but I found I'd entered but not saved the values for the switch points. Once I changed those it drove completely fine, other than needing the fueling table completely reworked.

I've found on both the 140 and the wagon, the old addage of run the timing until it pings, then back off really didn't work well. The 140 on the dyno showed no real gain but the driveability dropped when timing was too high. The wagon loses spool when it's too high, even if it's not pinging.

One suggestion as well, for the spark side. If the shop has one, piggyback a scope onto the knock sensor signal into the EZK. That way you can see knock rather than trying to listen to it. We used Ryan's Snap-On, worked out pretty well.

Click to expand...

I also didn't have the desired results when first trying to plug in the 012 look up data, but I hadn't entered and saved both tables apparently. It was ok after that.

That's good to hear on the 140 not showing a gain, that means it's in theory at the best.

I like the scope idea, I don't know if they have one but we could use the adapter from my Knocksense to easily monitor the signal.

VB242 said:

I know this isn't the same but my microsquirt was doing some crazy Accel enrichment at idle because I could only get enough air to idle by adjusting the throttle plate open a little with the cable which made the switch unstable. I increased the Accel actuation percentage and fixed the problem, so maybe take a look at your tps adjustment.

Click to expand...

The switch is adjustable though and I did have a similar thought, but it's ok. I do for some reason wonder if it has something to do with the throttle body itself, and I have toyed with the idea of going to a smaller one from the 16V cars because it's so touchy at low rpm/idle, but we'd have to come up with an adapter to go from the larger AMM down to the 16V TB. Might still be worth doing for testing, at least??

MisterBLC said:

For the idle, are you getting enough airflow at idle to get a decent MAF reading? Maybe check the MAF output signal and ensure its making sense at idle/low engine speed. I also had on MS lots of problems at idle with big injectors being at such a small duty cycle that there would be wide variation in the actual fuel delivered...it was hard to compensate for. Check the duty cycle/open time of the injectors and make sure its not too small to be reasonable.

Then again, isn't this a race car? Just set the idle higher and both the above problems should smooth out

Click to expand...

That is a good question, I don't know but because others, like Gary above, have run this same setup on LH2.4 Turbo cars(with larger injectors), I'd be surprised if that is the issue. Maybe with more fine tuning of the injector constants on the dyno it will improve. The engine also has a 82? C thermostat in it and I can see that when at full temp it sometimes bounces between the 2nd and 3rd load cells from the max temp load cell, so I wonder if that has something to do with it.

Haha, yes, it is a race car and why I haven't really been too bothered by it, but it still annoys me! It's been like this for a year(2.1 races, 2 track days, 1 autox).

klr142 said:

The switch is adjustable though and I did have a similar thought, but it's ok. I do for some reason wonder if it has something to do with the throttle body itself, and I have toyed with the idea of going to a smaller one from the 16V cars because it's so touchy at low rpm/idle, but we'd have to come up with an adapter to go from the larger AMM down to the 16V TB. Might still be worth doing for testing, at least??

Click to expand...

Isn't there an 850 NA throttle body which is the same diameter as the 960 TB but has a big plastic thing attached to the throttle plate, to limit the "touchiness" of the throttle? May be an easy thing to snag from a junkyard and throw on.

MisterBLC said:

Isn't there an 850 NA throttle body which is the same diameter as the 960 TB but has a big plastic thing attached to the throttle plate, to limit the "touchiness" of the throttle? May be an easy thing to snag from a junkyard and throw on.

Click to expand...

Yes, I did think of this previously as well and that would definitely be an easier way to do it... Who knows if it has some other affect of introducing turbulence that may or may not be of concern with our 90? elbow it's mounted to. That would help with the touchiness, but not necessarily whatever is causing the idle fuel mixture variation. I think the current throttle spool is somewhat junky so we might need to make another one of those pieces of junk so we can switch back and forth between the two easier.

hi

Hi Kyle

I hope you have disabled the Lambda feedback flags ?
Trying to tune something with those on are pretty useless.

Secondly, sometimes the LH (or EZK for that matter) just get a hiccup when working towards an Ostrich, and the only known cure for that is to disconnect all power to the ECU/EZK for a minute or two. Handy with a purpose made switch for that usage.

Regarding Idle stability it pays to have sane flat areas on the maps around idle (rpm & load) , and the tendency towards instability increases with more cam, compression etc. This goes for both LH and EZK, remember that the EZK uses a separate table when idle contact is closed on throttle.

Jaybee

mrjaybeeze said:

Hi Kyle

I hope you have disabled the Lambda feedback flags ?
Trying to tune something with those on are pretty useless.

Secondly, sometimes the LH (or EZK for that matter) just get a hiccup when working towards an Ostrich, and the only known cure for that is to disconnect all power to the ECU/EZK for a minute or two. Handy with a purpose made switch for that usage.

Regarding Idle stability it pays to have sane flat areas on the maps around idle (rpm & load) , and the tendency towards instability increases with more cam, compression etc. This goes for both LH and EZK, remember that the EZK uses a separate table when idle contact is closed on throttle.

Jaybee

Click to expand...

Hello Jaybee!

Yes, I did have the Lambda feedback turned off during the track day but I turned it on afterward for the race. I do think something got messed up when I loaded the leaner Bin into the computer though and I did not reset it at that time, sadly. That would have been an easy fix, potentially! A power switch would indeed be nice for that, we've just been unplugging the fuel pump relay instead.

I have played with most of the maps around idle and low rpm/higher load, but I can definitely flatten things out more. The fact that I don't think the computers are pulling from the proper cells with my current Injector Constants may have something to do with this, but while tracing the idle I don't see it pulling from higher load cells than expected. Again, I just have IC1 adjusted and not IC4. I will look into that more as my first steps on the dyno.

Thanks!
Kyle

I have only done IC4 and it seems to be a time based figure. Higher number for smaller injectors. Dropping this down also helped with idle quality and starting when not running stock injectors. I didn't notice a change in map hits when changing it either.

you definitely want to tune the ignition for minimum advance and max torque. many times the knock limits the timing but keep bumping it up until you stop gaining power or you get det.

Your idle AFRs are moving around a lot because your wideband is not seeing enough volume. You may even be getting a negative pulse, depending on how your exhaust is tuned. Ours is reads over 17, we have been advised to ignore it. Trust the O2!

gsellstr said:

I have only done IC4 and it seems to be a time based figure. Higher number for smaller injectors. Dropping this down also helped with idle quality and starting when not running stock injectors. I didn't notice a change in map hits when changing it either.

Click to expand...

I will be playing with it! In theory I should be able to just run the 950 bin’s Info, but it’s for a turbo so not quite ideal and I don’t know if it’ll work in a 933 ecu.

blkaplan said:

you definitely want to tune the ignition for minimum advance and max torque. many times the knock limits the timing but keep bumping it up until you stop gaining power or you get det.

Click to expand...

Yep yep.

shoestring said:

Your idle AFRs are moving around a lot because your wideband is not seeing enough volume. You may even be getting a negative pulse, depending on how your exhaust is tuned. Ours is reads over 17, we have been advised to ignore it. Trust the O2!

Click to expand...

The idle AFRs are definitely messed up, as it idles fine when warming up and the wideband reads properly(always has, we've been running it for years). Once it starts cycling a bit, the idle speed drops with the lean mixture and recovers when it comes back near stoich. This problem happens even with Lambda feedback turned off. We shall see!

Hi Kyle


Why on earth would you turn the lambda feedback back on again and mess up all your tuning effort ???


I have driven all my volvos withouht lambda feedback the last ten years and never had any issues with that whatsoever. even passes the tech inspection every time The sond is attached but not operative

We?re on our way to KO Racing in Oregon City to strap it down! What?s it going to make?

Noise

step 1) throw LH in the trash and install an ecu which is more workable for a not exactly street mannered engine like LH was designed for.

1.21 gigawatts