PSI+timing+AFR sweet-spot?

[EfiniX]

I know it's different for every car, but I imagine there's a point of diminishing returns chasing more boost at the expense of less timing. Lot's of bits, but the relevant ones here are...

FIC 950cc injectors
Wally 255
T3-T04
264/264 cams
SD setup w/ ECM link
FP racing exhaust w/3" Fujitsubo all the way back
ETS GVR4 FMIC w/ welded pipes
Innovate WB

I've never found 93 octane in Portland, so I run 92. I'm currently pushing 21psi (that's from the OmniBar, not the gauge). I've got knock around 4500 RPM's I'm working out, but I want to make sure I'm going in the right direction. It's just a few counts, but still, you can feel 2 degrees of time getting pulled and it takes all the fun out of it.

Here's where I'm at in terms of VE, timing, and AFR










Generally, I see my knock right at a 2.4 load-factor between 4500 and 5000 RPM's. I'm also only seeing about 11:1 AFR, so I need to tweak my VE table a bit. That being said, what's (anyone here's) shoot-from-the-hip advice on which direction to go? Lower timing or lower boost? I'm guessing both will sort the knock out, but at this point, I would also wager that one will offer better performance numbers than the other.

 

[EfiniX]

I'm also noticing that my BoostEst is about 3PSI over my actual boost. Is this value (boostest) even relevant with a MAP installed?




I would also add that I've got a coolant offset of 15, so my coolant temp shouldn't have an impact, here. The knock bumps you see are 1deg (timing -0.4) and 7 (timing -7) respectively.

 

[CutlassJim]

Everyone has there own schools of thought on this. I like to run "lower" boost to keep the torque hit lower when boost comes on and then make power up top with timing. Also to keep the compressor on a better efficiency island. 16G cars I shoot for 22 psi on a stock head and 20 with cams/port work.

 

[ghostinthevr4]

Try smoothing out your AFR table first and see if the knock goes away, your 4500 to 5000 cells are not smooth. Leave all settings the same and just change the AFR table and go from there, if it's still knocking after that play with the timing table and smooth it out.

Also to run more timing on lower octane fuel you'll have to run your AFR's richer, the cells you have at 11.00 match them too the 10.5 cells.

Hopefully that will keep the knock at bay if it doesn't work lower it to 10.00 and see if the knock goes away, you gotta get everything balanced and smooth.

Good luck tunning pump gas It isn't easy.

 

[Tommyd]

Personally, with the fuel and turbo you have I would start at 16psi and play with timing. You'll probably be able to get more out of it that way. Once you get it dialed in there, you can play with a little more boost if you still aren't happy, but you may end up needing some water/meth to make everything happy.

Your AFR chart looks pretty good but go from 11 to 10.5 over 2-3 cells. It seems like you are trying to add more fuel by tweaking the VE table and not adjusting your sliders. Try not to exeed 100 on any of the cells, though some say below 105 is still good. If you have to do that, there's something wrong elsewhere. Lower boost, get afrest and wideband to be the same, fix VE table and that should then fix your boostest. Then play with timing

 

[GSTwithPSI]

Giving advice without being able to see the whole log is pretty much worthless. Post a log of a 3rd gear pull from about 3k to redline.

Is your TPS calibrated correctly? If 98% is WOT, it's wrong.

 

[prove_it]

That, and looks like you need to fine tune the fuel calibration. You shouldn't have so much spread between estimate and logged. Don't hang up on it though.

There is such a thing as rich knock too.

Post a log, it's way easier to understand what is happening when you can see the whole run event.

 

[EfiniX]

here's the log that pull is from. It does look like my TPS could use an adjustment. I'll do that today.

click

Followup question: Is there a way to adjust the entire VE table at the same time? If I start adjusting global fuel, I'm going to need to update the VE table for these changes. Assuming it stays relatively linear, I've got a good idea about how much to drop a VE cell for each tick on the global fuel.

And I'm starting to think my initial tune might have been a bit bunk. My current global fuel is 55.5. With the FIC 950 injectors I have, ECM says I should be around 53. that difference would certainly account for the high VE cell values. Based on my math, adding 2.4% to get me to 53.4 will richen me up by about 1.0:1. There might be some tweaking involved, but I'm going to start there.

And yes, being able to adjust all cells so that they lean out a bit would be great. Based on the work I've already done, a drop of 1 in a VE cell will lean me out about 1.2:1. That being said, it's a starting point. i'm guessing there's some logarithms in here, which means that my estimates are going to be off depending on where in the table I am.

 

[prove_it]

I used to have LinkTools until some asshat ruined that.

Don't get hung up on the SD table values being a certain number, they are reference only and work with the global fuel. It's not worth changing the global to make your VE table "look" right. Looking at the log, I'd say drop to 53% global as wbFACTOR is showing to add all across the pull. Use WBfactor to determine how much you need to add and subtract. Read up on how to effectively use it.

Also why is there no IAT? You should be running this. It can effect the tune.

Your MINoctane map is set to run LEANER as knock occurs. You need to swap the two maps around, then dial the 11.5 to 11.0.

Which Plugs are you using? You can try NGK BR7ES. These are non projected tips and I've seen knock vanish with them. Cheap idea.

 

[EfiniX]

I used to have LinkTools until some asshat ruined that.

Lame

Don't get hung up on the SD table values being a certain number, they are reference only and work with the global fuel. It's not worth changing the global to make your VE table "look" right. Looking at the log, I'd say drop to 53% global as wbFACTOR is showing to add all across the pull. Use WBfactor to determine how much you need to add and subtract. Read up on how to effectively use it.

roger

Also why is there no IAT? You should be running this. It can effect the tune.

Not running one. I was a bit surprised when I noticed this, too, but Lucas at English Racing assured me it's fine. I know it's better, but they did zero-out the temp weighting and they are the experts, so I trust their judgement. That being said, here I am...

Your MINoctane map is set to run LEANER as knock occurs. You need to swap the two maps around, then dial the 11.5 to 11.0.

This is something I need to understand better. I've just been tuning the MaxOctane timing and Open Loop tables. It almost sounds like you are suggesting the min table is used when knock is detected, but I thought it was based on the actual octane of the fuel in the car. I'm not logging the octane, so I'll start. But I would like to understand this relationship better.

Which Plugs are you using? You can try NGK BR7ES. These are non projected tips and I've seen knock vanish with them. Cheap idea.

yup, using those plugs. they have been great!

 

[prove_it]

The min and max tables work together pending on the learned octane value. The value starts at 255 and drops as knock is detected. The ECU then interpolates the two maps together until the learned octane is low enough that it uses the MIN octane map completely. You do not want to see any knock and the value should remain at 255.

It's a safe guard, but it takes long periods of knock to reduce the learned knock value. Once it's reduced it tends to stay there until the ecu is reset. It's not very useful to be honest, I usually set the maps the same and ignore this feature. If this is mostly a race car only, I'd set them the same and continue with the tune, that way it can't influence any varibles. Once tuned and you drive it all the time and take road trips, you can basically reduce all the "boost" cells to have less timing and more fuel just in case.

You only need to worry about the max octane maps for tuning. I noticed the min octane fuel table was the opposite of what you should have so I let you know.

 

[EfiniX]

Thanks! That makes all kinds of sense, now. I've got a pull logged. Idling in my car now. Posting in 5...

 

[EfiniX]

And here we go. Frankly, my AFR's and such look good to me. The WBFactor is low when it isn't 0, and I'm seeing the AFR's I want to see. Is there a functional reason I would touch the global fuel right now aside from using it to bring my VE cell values lower (so I don't have ones that go over 100). it's been suggested I do this, but is having a VE cell over 100 really an issue?

Some quick notes:
I have a coolant offset of 15 degrees, so running between 190-200F should be fine.
My TPS is calibrated correctly. I guess I didn't have the pedal to the floor on the last run.
I'm seeing knock in the same places as when this started, so yay for consistency.

I have to say, I'm moving a LOT of air with this turbo. T3/T4 w/ .54 clip. All things being equal, it doesn't feel like it kicks as hard as I would think it would. Of course I've noticed that when driving AWD cars, the turbo "kick" feels quite a bit different than with FWD or RWD, so it might just be me. Which takes me back to the original question: timing or boost? Generally, I think I'm going to have to pull my timing where I'm seeing knock because I can't imagine dumping more fuel in at 10.5:1 is going to help much. But with my setup, I'm wondering if at 18PSI and some better timing, do I get more power? My car dyno'ed at 336awhp running 20.5psi, which seemed weak for injectors, cams, 255 fuel pump, ported everything, t3/t4, ETS FMIC, 264 cams, 3" exhaust, etc...

click

 

[prove_it]

Try pulling 1-2 degrees and lean it to a full 11.0 across the pull.

Like I mentioned, sometimes you can get rich knock. If the knock is better, then good, if not, add some fuel and keep timing alone.

How old is your WB sensor and when was it last calibrated?

 

[EfiniX]

Try pulling 1-2 degrees and lean it to a full 11.0 across the pull.

Like I mentioned, sometimes you can get rich knock. If the knock is better, then good, if not, add some fuel and keep timing alone.

will do

How old is your WB sensor and when was it last calibrated?

To my detailed record-keeping! 7/2/13 for the Innovate WB installation. Last calibration on 3/20/15. I'm using the LinearWB input and I've got the voltage adjusted so there is negligible drift between the gauge (true) and Link (DAC).

 

[GSTwithPSI]

Quoting EfiniX:

Also why is there no IAT? You should be running this. It can effect the tune.

Not running one. I was a bit surprised when I noticed this, too, but Lucas at English Racing assured me it's fine. I know it's better, but they did zero-out the temp weighting and they are the experts, so I trust their judgement. That being said, here I am...




Sure, it's fine on a racecar where you can adjust the tune according to ambient temperature before you make a pass down the 1/4 mile. But for a street car that's driven under various conditions and temperatures, you should absolutely have one. Unless A) you gave him the impression he was tuning a pure racecar, or B) you are completely out of ECU inputs, that's sh*t advice.

Basically, your ECU has no concept of changing intake temperature. That's bad, especially when you're trying to dial in your tune and get rid of minimal counts of knock under a specific set of conditions. That's pretty much impossible when your ECU only knows a static temperature in a dynamic set of environmental variables. Simply, as conditions change, your ECU only has 2/3 of the available data used to make SD optimally accurate. That adds up when you trying to work out the little bugs like you're dealing with.

I'll look at your log later when I have more time. In the interim, I'd strongly consider picking up an IAT. Aside from English saying so, you really don't have a reason not to run one IMO.

 

[EfiniX]

Fair enough. I'll check my ECU inputs, but I'm guessing I have room for one. Assuming yes, I'll pick one up.

I'm assuming it can go right on the intake elbow? I'd hate to have to pull the head just to drill a hole in it.

 

[GSTwithPSI]

Yeah dude, you can locate it in the TB elbow just before the throttle plate. That's a common location to mount it.

 

[prove_it]

IAT is also a good way to check the turbo efficiency and if the intercooler is good enough for the pressure.

 

[EfiniX]

Well I'll be buggered...

I pulled a degree of timing and leaned out to 11:1. No knock on a fourth gear pull! My laptop died, so that's all I got, but looking good. I'll post the log later and get a 3rd gear pull in when I can. For now, I'm still playing with part-throttle timing and afr's. English had my knock sensor (which is new in the last few years) threshold set to 4000rpm's and 70% TP. That seemed excessive. I dropped it to 3500/40% and I can now see a few counts right before I hit 4000. Working on that, too.

Thanks, all. Keep the advice coming. I love learning about this stuff!

And I suppose I have one other question that's a but nebulous... should I even be expecting more power out of this setup than 340-ish? That just doesn't seem to square with what I've read. I mean there just isn't that "kick" I keep expecting...