LSA LQ4 Tuning Help

[schiavonske1013]

So I've been reading for months, got this truck finished up after spending 4 years slowly building it. Figured I'd try my hand at tuning myself, I tried it on my previous setup and it went fairly well, but I feel like I am going around in circles so I'm finally giving in and posting. Went through enough bad mechanical issues from 2 sets of bad injectors, a bad head gasket, a broken valve spring, broken trunnion, that the tuning is just driving me nuts.

Mods are in my signature. I've been trying to tune this setup using STFTs for closed loop and wideband for boost.

My main issue is getting the fueling correct, I'm in speed density, 2 bar OEM MAP sensor. I have been trying to log and smooth out my VE table, not just copying and pasting % error and leaving spikes, but I'm confused. I know my file naming is terrible, but I got tired of long names at one point. So "again 28" is the tune this log is based off of. They seem to be at an acceptable percent error, but the boost areas are stumping me. Based on watching boost climb in the log, I am relatively close up to about 140-150kpa, but above that I am actually too rich, adjusting areas above 140 to bring fuel leaner actual creates a slight "cliff" in my VE table. Does this seem correct? "again 29" is my attempt at tuning the VE based on the log, you can see where it seems to dip ever so slightly. I've also got the typical lean spikes with quick throttle transitions, see the area about 5:37 in the log, somehow I hit 206kpa, not sure how I tapped almost 15psi of boost, that seems like another issue. Is this something I should be looking at transient fueling on, or do I need to up certain areas of the VE table?

I've also got some issues with return to idle and idle dipping down, but I don't have the idle stuff logged here, that I will tackle once I get the main fueling figured out.

I've waited years to get this running and finally did, and now the tuning is killing me.

EDIT: Though of this after, my base fuel pressure is set to 58psi. I was also messing with EOIT by setting it at what a stock CTS-V uses to try and see if that cleaned up some of my issues, no idea if this was a stupid move.

[04silverado6.0]

I believe this file will be a better start. May not be close at all on ve but it at least has the right shape. I believe your eoit could be later as well, try adding .2-.3 to the normal table and see if you get richer or clean up any idle smell.

[schiavonske1013]

I'll give that a shot and see what happens, I appreciate you taking a look at it.

I honestly thought I went overboard pushing the EOIT out as far as I did it the tune I posted, but I will try what you said.

[schiavonske1013]

May seem like a dumb added question, but is there a chance that my injectors being way more than I need for my application hurting me? I only plan on using 93, e85 is almost non existent where I’m at. At 58 psi base pressure ID data has my injectors flowing 117.49 lbs/hr. I didn’t got full pedal down on the log I posted, but previously when I had done it I maxed out my injectors at like 45% duty cycle. Is there any chance such a low duty cycle is leading to them opening later thus hurting transients? I was considering dropping base pressure to 3 bar to see what happens. I was also considering trying Lsa injectors to see what happens, it looks like i should be under the power requirements to run them.

[04silverado6.0]

Duty cycles under 60% i believe rob hp. Getting sized down to something that would get you in the 60-75% range will pick some power up. It does affect injection timing and may contribute to bad fuel vaporization at idle/injector not firing at all.

[kingtal0n]

Here is what I see


Maybe flush out that part of the VE table it appears to be more flat in those areas based on the trend.

Use the wideband gauge for reading wide open throttle numbers because of analog offset. Compare gauge to log to know the true values from the gauge and adjust the log so it matches the gauge.

Make sure you prepare analog offset term in the user maths like this 2007 thread explains otherwise wideband is useless might as well not even have one. I based my offset on your narrowband oxygen sensors but on the assumption you have 0-5v wideband and that that was at a specific RPM and the analog offset will change slightly depending on the RPM and electrical load, so it isn't a set and forget term, you will need to take multiple samples and prepare analog offsets maths for highrpm and lowrpm and sometimes even day/night situations.
https://forum.hptuners.com/showthrea...ll=1#post79124



2. The larger the injector the better. Shoot for low duty cycles 40% to 55% is nearly ideal. I'll use Haltech firing angle for reference because it is universal and easy to understand. Haltech references in before top dead center compression, so 360* BTDCC is firing an injector 360 degrees before compression TDC, in other words during overlap of intake and exhaust and the worst possible time to spray an injector possible. You prefer either timing in the range of 410*BTDCC which gives fuel a chance to vaporize, this is the factory method of letting fuel pool or smack an intake valve where it boils off into the runner and creates a column of fuel vapor if the valve is hot enough. The factory HPtuners value of 5.55 is probably around there. The second method is a timing between say 300*BTDCC and 215*BTDCC which is during the time when exhaust valve is closed but intake is still open, allowing you to get fuel in during the intake stroke, which may dramatically improve torque (cylinder filling) and engine response in some engines/combo. To do this however requires a large enough fuel injector to fire within quickly that range for those low to midrange torque situations, around 30% duty cycle maximum window. The larger injectors and reduced duty cycle also keeps the injector drivers cooler, the electronics more happy, less current draw keeping injectors open, larger injectors has all positive benefits for modern high impedance quality injectors.

[schiavonske1013]

Two interesting takes. I was attempting to do it the “right” way by getting steady state ve data in my table before trying to work on the transients. Was also considering dropping base pressure to see if that helped either. I was thinking lower pressure would cause the injectors to open sooner and help transients, but the more I think about it I’d still be injecting the same amount of fuel, just sooner.

[kingtal0n]

I run the least fuel pressure possible to make idle tuning easier and take stress off parts. 34-38psi is where I prefer to live, all my cars turbo all low pressure possible for highest potential fuel flow from the fuel pump at max boost. Lowest fuel temperature, very important. Its a secret of tuning gasoline forced induction apps.

The fuel pump will last longer if pressure is lower. There is less likely chance of leaking. And so forth

I talk frequently on tuning forums about these ideas
https://forums.holley.com/forum/holl...495#post385495
https://www.supraforums.com/threads/.../post-14037071
https://zilvia.net/f/showpost.php?p=6302262&postcount=9
https://www.corvetteforum.com/forums...post1607024948
https://www.thirdgen.org/forums/tpi/...ml#post6465375

etc...

[04silverado6.0]

Here is what I see


Maybe flush out that part of the VE table it appears to be more flat in those areas based on the trend.

Use the wideband gauge for reading wide open throttle numbers because of analog offset. Compare gauge to log to know the true values from the gauge and adjust the log so it matches the gauge.

Make sure you prepare analog offset term in the user maths like this 2007 thread explains otherwise wideband is useless might as well not even have one. I based my offset on your narrowband oxygen sensors but on the assumption you have 0-5v wideband and that that was at a specific RPM and the analog offset will change slightly depending on the RPM and electrical load, so it isn't a set and forget term, you will need to take multiple samples and prepare analog offsets maths for highrpm and lowrpm and sometimes even day/night situations.
https://forum.hptuners.com/showthrea...ll=1#post79124



2. The larger the injector the better. Shoot for low duty cycles 40% to 55% is nearly ideal. I'll use Haltech firing angle for reference because it is universal and easy to understand. Haltech references in before top dead center compression, so 360* BTDCC is firing an injector 360 degrees before compression TDC, in other words during overlap of intake and exhaust and the worst possible time to spray an injector possible. You prefer either timing in the range of 410*BTDCC which gives fuel a chance to vaporize, this is the factory method of letting fuel pool or smack an intake valve where it boils off into the runner and creates a column of fuel vapor if the valve is hot enough. The factory HPtuners value of 5.55 is probably around there. The second method is a timing between say 300*BTDCC and 215*BTDCC which is during the time when exhaust valve is closed but intake is still open, allowing you to get fuel in during the intake stroke, which may dramatically improve torque (cylinder filling) and engine response in some engines/combo. To do this however requires a large enough fuel injector to fire within quickly that range for those low to midrange torque situations, around 30% duty cycle maximum window. The larger injectors and reduced duty cycle also keeps the injector drivers cooler, the electronics more happy, less current draw keeping injectors open, larger injectors has all positive benefits for modern high impedance quality injectors.

You must not have read the very long and drawn out thread where it was proven that low duty cycle loses power. Some of the power can be made back pushing eoit past the point of good drive ability. Lower fuel pressure is great with a pwm pump at idle or cruise but who wants 30 psi at wot and a 200lb injector in a 500hp build?

[schiavonske1013]

You must not have read the very long and drawn out thread where it was proven that low duty cycle loses power. Some of the power can be made back pushing eoit past the point of good drive ability. Lower fuel pressure is great with a pwm pump at idle or cruise but who wants 30 psi at wot and a 200lb injector in a 500hp build?

I read about that as well, not to ignore your original file you provided, but just for gits and shiggles I took my base fuel pressure down to 43.5 and redid my injector data, using this file this is the resulting log. I toned my EOIT back to what the spreadsheets say it should be, I did notice more perceived responsiveness, but quick throttle blips are still my nemesis, as well as the oddity I'm noticing with my VE, If I review the error in the log and were to correct the PE sections based on what I see, it actually would give me a bit of a "hump" between about 100 and 130 kpa, unless I am interpreting wrong, as I would seemingly need to either raise a miniscule amount or keep it the same, while once past 125-130kpa I am actually too rich and would need to bring my VE down a bit. Almost as if there is some extra air hanging around causing a need for more fuel in low boost situations.

I considered picking up a set of cheap LSA injectors I found locally just to see what happens, I understand it may help transients and would definitely up my duty cycle, running a boost referenced FPR and bumping it back up to 58 psi it SEEMS like they may give me enough fuel.

[SiriusC1024]

More responsiveness because a larger proportion of injector flow is linear. Decreased pressure means increased duty cycle. Even with data to address this the data is often slightly inaccurate for aftermarket injectors.

LSA injectors fire at an angle. Standard straight spray injectors will have issues with port wetting.

When you reduce fuel pressure offset and SPA will be affected. No data available to compensate, so that's another problem. Offset can be roughed in by multiplying by SQRT(P2/P1), 0.866 in this case, but SPA is another matter.

[kingtal0n]

You must not have read the very long and drawn out thread where it was proven that low duty cycle loses power. Some of the power can be made back pushing eoit past the point of good drive ability. Lower fuel pressure is great with a pwm pump at idle or cruise but who wants 30 psi at wot and a 200lb injector in a 500hp build?

You do realize the numbers I quoted are factory for millions of vehicles? I didn't say or suggest anything that isn't routinely done around the world for factory and performance vehicles.

For example my Toyota Supra 44psi of base pressure with vacuum reference at 16"Hg is 36psi of idle fuel pressure stock. At 25psi of boost with a large single turbo that is 44+25 = 70psi of fuel pressure. Pretty fucking high and pushing the limits of a cheap in-tank fuel pump so in order to get some head room we drop the base fuel pressure from 44 to around 38psi and install larger camshafts which reduce the idle vacuum to about 10". Now the idle fuel pressure is around the same as factory and the boost fuel pressure is far superior for reliability, fuel pump flow rate, fuel pump longevity, etc... everything gets better.


Nobody is losing power with low duty cycle that is just ridiculous. Something you changed caused the difference not the duty cycle. If low duty killed power then direct injection would have the worst power conversion and it doesn't it has the best. Power comes from fuel not from the injector size.

[schiavonske1013]

More responsiveness because a larger proportion of injector flow is linear. Decreased pressure means increased duty cycle. Even with data to address this the data is often slightly inaccurate for aftermarket injectors.

LSA injectors fire at an angle. Standard straight spray injectors will have issues with port wetting.

When you reduce fuel pressure offset and SPA will be affected. No data available to compensate, so that's another problem. Offset can be roughed in by multiplying by SQRT(P2/P1), 0.866 in this case, but SPA is another matter.

Yeah I had used their “universal” data to try and adjust offset, but haven’t gotten a response on if they have a complete set of data so I’m assuming it’s slightly off. Wasn’t sure if the responsiveness was the pressure change or the eoit change. I’m trying to get steady state ve data sorted before dealing with the transients, but they odd richness at higher boost is confusing me.

[kingtal0n]

I read about that as well, not to ignore your original file you provided, but just for gits and shiggles I took my base fuel pressure down to 43.5 and redid my injector data, using this file this is the resulting log. I toned my EOIT back to what the spreadsheets say it should be, I did notice more perceived responsiveness, but quick throttle blips are still my nemesis, as well as the oddity I'm noticing with my VE, If I review the error in the log and were to correct the PE sections based on what I see, it actually would give me a bit of a "hump" between about 100 and 130 kpa, unless I am interpreting wrong, as I would seemingly need to either raise a miniscule amount or keep it the same, while once past 125-130kpa I am actually too rich and would need to bring my VE down a bit. Almost as if there is some extra air hanging around causing a need for more fuel in low boost situations.

I considered picking up a set of cheap LSA injectors I found locally just to see what happens, I understand it may help transients and would definitely up my duty cycle, running a boost referenced FPR and bumping it back up to 58 psi it SEEMS like they may give me enough fuel.

Your boundary is strange and you could be spraying fuel after Intake Valve Closed which would explain the transient delay for the rising boost section 100-130kpa. I recommend go back to factory boundary and use the "normal" table only for adjustment, the way it is now I can't tell when you are spraying actually and I Don't think anybody else can either. Lowering fuel pressure will rise duty cycle and mask these affects which is why it could be helping because now you are spraying more often. The offset in the data you worried about is negligible perhaps 80uS to 130uS don't fuck with the injector data. The VE table will flatten out after 100KPA In fact you could put nearly the same number at 200KPA as 100KPA and it would be close. The VE actually drops in boost but the VE table number gets larger because brake specific fuel consumption of converting fuel to work gets worse due to parasitic crank and exhaust gas pumping loss for some exhaust configs so more fuel is needed to pay for those losses (typical low boost supercharger in street cars will absorb 40 to 80hp) and the VE table can provide that extra fuel as the number rise only slightly.
This is needed for your tune file because you chose to set PE to a single value. Which is fine for beginning tuning and novice stuff. But eventually you will want to interpolate the PE table so it provides a slight enrichment at first and more as boost rises. Naturally this means you can dial your torque model by adjusting the VE Table directly to any torque you desire (for transmission pressure or torque management or whatever) and then use the boost enrichment PE table to accomodate that tuning strategy.

[SiriusC1024]

Yeah I had used their “universal” data to try and adjust offset, but haven’t gotten a response on if they have a complete set of data so I’m assuming it’s slightly off. Wasn’t sure if the responsiveness was the pressure change or the eoit change. I’m trying to get steady state ve data sorted before dealing with the transients, but they odd richness at higher boost is confusing me.

It's never going to be right without GM format data.

[schiavonske1013]

It's never going to be right without GM format data.

So am I better off going back to 4 bar and using transient fueling to calm the spikes?

[kingtal0n]

It's never going to be right without GM format data.

The offset is around 1ms for general injectors around the world high impedance. At 150kpa his injector is 8.7ms pulse.

If the injector data offset is wrong by 0.2ms. That is around 2.3% Difference. Negligible for many reasons, cylinders are different, injectors are different. If you want we calculate injector pulse->fuel mass->air mass calc to show the air fuel ratio change is negligible with respect to analog offset order of error term, meaning that injector data issues at wide open throttle boost cannot be seen in the scanner logs. Not even with a canbus because of the way microcontroller averages what we see in the scanner.

[SiriusC1024]

The offset is around 1ms for general injectors around the world high impedance. At 150kpa his injector is 8.7ms pulse.

If the injector data offset is wrong by 0.2ms. That is around 2.3% Difference. Negligible for many reasons, cylinders are different, injectors are different. If you want we calculate injector pulse->fuel mass->air mass calc to show the air fuel ratio change is negligible with respect to analog offset order of error term, meaning that injector data issues at wide open throttle boost cannot be seen in the scanner logs. Not even with a canbus because of the way microcontroller averages what we see in the scanner.

Obviously. It's for normal driving. The problem is going to be magnified by these very much oversized injectors. Why do you think the offset table exists?

[schiavonske1013]

Obviously. It's for normal driving. The problem is going to be magnified by these very much oversized injectors. Why do you think the offset table exists?

I have an old set of fic650s that’ll run 720cc at 58 pounds that I may toss in and see what happens, should be running a higher DC with those if I can’t run Lsa injectors.

[kingtal0n]

Obviously. It's for normal driving. The problem is going to be magnified by these very much oversized injectors.

You should be able to tell by now I respond to everybody the same way. I do not care about praise or anger. I am not attacking you when I make your corrections I am guiding you. I only hit back when I am hit first.
Your tone indicates you think I have some kind of grudge on you but I do not. I will never care what you or anybody else thinks. I put the information down and I scrutinize everything, everybody, nobody is treated special or different.

Please learn these things for your tuning ability

1. You quoted his post

but they odd richness at higher boost is confusing me.

And responded telling him that it will never be correct, confusing him because clearly you knew deep inside offset data has nothing to do with his problem. Do you see this op? Do not be mislead. The offset is not causing your boost richness problem.

People on this forum are often very UNCLEAR in their responses. Do not be offended by any CORRECTIVE measures.

It's for normal driving.

2. What mean normal driving? DEFINE normal driving please. I assume you meant driving cruise/idle.
During 'normal driving' cruise/idle, closed loop feedback compensates for crap like bad offset and then saves the value as a fuel trim.

Bad offset is not going to directly cause air fuel ratio problems because whether tuning open loop or closed loop, the user or ECU is compensating for the incorrect air fuel ratio.
The real issue is what the VE table looks like once this is done with bad offset. Let us review what offset does to see why

Why do you think the offset table exists?

Glad you asked , here is why

The offset for the OEM ECU is intended linearize the low pulse portions of the injector near 1ms to 2ms , generally far from 'normal driving' in street cars still using narrowbands (sub 1200rwhp vehicles, after which most would be using widebands and much larger injectors and aftermarket ECU) making it easier to tune idle and keeps the VE table to 'make sense' because for OEM ecu the VE table making sense while not necessary is a helpful accessory for tuning as a novice beginner tuner. But not necessary for gen3 applications since torque model does very little and controls very little. For example it is easy to SCALE a VE table 1/2 and injectors 1/2 and this has no far reaching consequences the vehicles are frequently tuned this way in gen3 for large injectors. This tells us that offset is not important at all in gen3 applications because there are many ways to 'BREAK' the OEM ECU's torque model and still everything will work.

https://forum.hptuners.com/showthrea...cycle-question

Within this linear operating range, the output vs pulsewidth, when plotted on a graph, will follow a straight line.

If our injector data is correct, this straight line will exactly match that of the line defined by our offset and IFR values. As long as the output of the injector lies on this line, the fuel delivered to the motor will be as commanded.

If the offset is incorrect the VE Tables near low non-linear pulses will be nonsense, but the normal driving portions will barely notice because pulses are much higher. Unless the injector is very large for some reason.
In which case the VE can be adjusted to 'crazy numbers' that don't make sense. In a manual trans application this has no drawback I can see. With an automatic you can adjust the force motor to compensate anyways so again it doesn't really matter. In short, even if I put random offset values into my offset table for gen3 applications, near 1ms for example, the ECU will still be able to tuned for the engine easily and has no consequences for advanced tuning experts who know how to scale the ECU even when guessing offset values.

In aftermarket ECU it matters even less. I've done hundreds of megasquirts, power FC, AEM, Haltech, Motec, etc.... without knowing injector offset values. Because those ECU do not rely on any torque model so when the VE table values at low pulses become too large or too small for the rest of the table it has no consequences.