Thanks, I already put my highest values from the boost table in their so I'll give it a 5% increase.
Attachment 104245
I reduced my flex spark since you said if anything reduce spark advance for now while I tune the ve table. I multiplied it by 0.4 so if I was going to convert it back to where it was later then I would need to multiply it by 2.5
Oh, I get it now. Thanks
Okay, so I uploaded the tune and started the car, at first the EQ err was off about 1% while the car was warming up, once warmed up, it was 5% too rich. Went for a quick drive around my neighborhood and I was overall lean. So I updated the ve table with the log data and then I went on a second longer ride.
Attachment 104277
I filled up my gas tank partly through the drive, that's why ethanol% jumps up to 81%. Gonna update the ve tables again and then go for another ride.
Okay, so I updated my ve tables and estimated my boost ve tables for the time being(probably way off). I just read somewhere that when in SD mode, the ECU will automatically use the low octane spark table instead of the high octane table. So I copied the high octane table into the low octane table.
Attachment 104280
I'm about to go for another ride.
Okay, went for my last ride for today, I've been getting slight knock here and there but I'm not sure if it's actually knock. I put my spark tables back to the way they were before during the previous logs and I don't see much of a difference.
Attachment 104283
Attachment 104284
Here is my current updated tune after log 5. Looks like I may have to scale down the ve and scale up the A/f.
E40 does not use the low octane table when in sd. Put it back so if you have some knock, it will pull timing. Not sure how your boost ve table got to be like that. Remove any MAF reference in channels in your scanner.
Looks like the knock is in the same area. At 16 degrees, spark may be too low.? Not much knock though.
Okay, I tried to hit boost and kept my eye on my a/f gauge to make sure I don't go too lean. I was aiming for the early cells in the boost able to start off with. Came home and to my surprise I never went over 105kpa. So I checked my charge pipes for leaks by pressurizing them and no leaks. Then I pressurized my intake manifold and the map sensor doesn't seem to be leaking either. I have a separate boost sensor that connects to my scan gauges but I can't hook up that and hptuners at the same time since they both use the obd port, I even tried with a 1 to 3 splitter and my scan gauges work but not vcm scanner. Anyway, I plugged my scan gauges in and pressurized the manifold to 11psi according to my boost sensor, plugged in vcm scanner and it shows 104kpa. Double checked map characteristics in vcm editor and looks correct.
Maybe something in vcm scanner is off?
EDIT: Nvm, I figured it out, I needed to add intake manifold absolute pressure instead of manifold absolute pressure and change to 2 bar manifold parameter in the graphs, my bad. I guess I should of paid more attention to your channels Lakegoat.
In the end, I think I want to run SD 100% of the time, but I want to be able to switch to maf easily in case my map sensor fails, or something else happens.
What if I set up maf to work in the tune but disconnect it via manual switch that will disconnect the maf sensor wires. That way I still have a iat sensor and if my map fails then I just need to flip a switch.
Also, I found a formula that converts maf to map for speed density, thinking of setting up the maf and doing some pulls and the convert those numbers to map to get in the ballpark and fine tune from there.
I have been running sd for years--don't have a maf at all. Never had a problem.
I had a maf tune that would run well enough to drive around, but not a switch. I am not sure how that switch would work with the dynamic airflow calc. If you need the maf for inspections, you will have to do something. Give it a try.
The maf and ve table are tuned the same way, shouldn't be that complicated. Of course, the more you learn, the easier it gets. Never hurts to try a different approach.
Thanks, not sure if I will end up wiring a switch or not but it may be a fun experiment.Quote:
Originally Posted by Lakegoat
I haven't made much progress on my ve tables yet. I have a manual so I wonder if that is affecting the results a little.(Coastdown, clutch slips, etc.) I added some duplicates of my EQ err tables with filtering like only during over 20% throttle, etc.
I keep getting occasional knock, usually always around 1700rpm.
Attachment 104326
I have a short log of my car idling while it still had a "professional" tune. Cylinder airmass shows about .09, right now at idle, my cylinder airmass shows about double that .19. So, if this is the case and I tried keeping the same spark tables my tuner set up(scaled down of course while ve tuning) Wouldn't I change the location of spark advance in the tables, since maf/airmass was scaled 50%?
Attachment 104327
Attachment 104328
Knock occurs around 1700rpm and .72,.80 airmass. If those spark tables are off because of previously scaled airmass, I should be getting the spark advance at 0.31,0.4 instead which is 0 instead of 1-2.5(ignore picture spark values).
I was looking through my tune and to my surprise it seems I still have Clutch fuel cutoff active:doh: Hopefully tuning will be easier now with that deactivated.
Also, I don't know if this is related or not but since I started tuning SD, my idle will drop really low when I'm out of gear and coasting, like around 500rpm when I have the minimum rpm set at 800. Once the vehicle comes to a stop the rpms come back up. Coasting rpm is set at 800rpm and I've checked everything I could find related to airflow idle/coasting and compared it to stock. I've been playing with a few idle variables in between tunes and switching back to find a fix. I just changed a few variables in airflow final minimum, airflow under adaptive idle and Integral to test next.
Okay, I took the set spark tables done by my tuner, scaled the spark location by 50%, so spark at .08 shows up at .16 and so on. Then I put the differences between that and my high octane table(values I copied from Magnuson) in the flex fuel table so that in theory I would be running the same spark advance set by my tuner when filled up with e85.
Can someone take a look at this and tell me if this spark advance looks safe?
Attachment 104337 I also smoothed out the VE tables, Probably gonna have to start over for the most part.
Part of me suspects that the knock I was experiencing at 1800rpm and .76(g) could have been cause by not enough spark advance. This change increases spark in that cell by 10.
Open your Primary VE table. Open 3D view. Tap the [smooth selection] function 3 times. It will be a lot better. Then after every change that you make, smooth it once.
Add 2.00 to the 250 to 1600 rpm cells (all) in airflow final minimum and see what that does for coasting. Highlight those cells and do +2.00--enter. Easy to change back if it doesn't help.
Thanks for the tip on smoothing.
I tried to add 2 to the entire airflow final minimum and had no luck, I even tried 4 and didn't notice a difference.
Whenever I disabled clutch fuel cutoff in my tune, I also zeroed out all the prediction coefficients tables in the dynamic airflow section. Must of did something, cause when I started my car with this new tune, my idle was so lean the wideband couldn't even read it. Maybe my maf was still somewhat active.
I keep getting knock no matter what my spark advance is, usually happens around 1800rpms typically. I tried removing spark advance and it actually seems to cause even more knock with less advance. Maybe I'm overthinking it, idk?
This is my most recent log, I zeroed out the flex spark table in this tune to reduce advance.
Attachment 104367
Attachment 104364
This is a less recent log where I kept spark advance much higher, ironically, there is much less knock, knock still usually happens around the same rpms though.
Attachment 104365
Not sure how to go about this issue with knock and spark. I'm not even sure it its real knock or not or how I can figure it out. I don't think my exhaust is contacting anything, However I do have the hardcore Procharger Intake mod done to my car, https://www.ls1gto.com/threads/proch...le-mod.444935/.
However, I do have a giant rubber grommet inside that hole and a flex coupling before the Procharger inlet to compensate for engine movement, so I doubt this would be causing false knock.
Here is my most recent tune file. still a lot of progress to make.
Attachment 104366
Here is my most recent log, starting to make some progress on the VE tables. I reduced the spark advance even more by copy and pasting the stock IAT spark advance correction table. For idle, I copied and pasted the stock airflow final minimum table which helped reduce the surging a bit and get it a little more in control, still not perfect yet. Next I'm going to try to replace the zero values in that table with the other values in the nearby row and see what that does.
Still getting some knock.
Attachment 104370
Attachment 104371
Your knock sensor settings are stock. Set the high to 500.0000 across and the low to 100.0000 across.
When you get to +- 3 or 4 in your ve table, thats about as close as you need them. That will change everyday and you will be chasing your tail.
Thanks, the knock settings helped, no longer detecting knock which was probably false knock.Quote:
Originally Posted by Lakegoat
My high rpm VE cells are definitely off. I tried going high rpm low throttle and it was about 10% too lean. Think I'm going to remove my bypass valve later today so I can redline at wot without building boost to get some good data for my primary VE table. Should make tuning the boost VE table a breeze from there.
We had a cold front yesterday and car actually almost died on me when I approached a stop sign in my neighborhood and I pushed the clutch in to come to a stop. The idle surged and the rpms dropped down to 500, engine barely recovered. This was 68F weather.
Did a couple of WOT pulls with no boost and got some good data.
Should hopefully help with the boost VE table now.
I noticed over 6,000rpm the car felt like it fell on its face, I'm guessing because it got a little too rich at that point.
Attachment 104449
Yep, commanded is .788 and actual is .72
I bet you already have it corrected.