AutoVE nearly complete - observations and questions

[Kevin Doe]

Ok guys, I think I am nearly complete with my auto ve tuning. My last log had bens mostly in the 98-99 range, and a few cells with a bit more correction than that.

1. My main VE table is something on the order of 40% less in the idle region than stock, and my hot starts are not that great. For a hot/warm start I have to push the throttle to get it to start. If I compare the 800 rpm and 400 rpm cells of my main VE to the cranking VE table, the values differ significantly. I know I need to modify the table, but how it is usually done? If my 800 rpm column is off by say 40% approximately, do I scale the entire VE map by 40%, or do I use the 400,800,and 1200 rpm columns and scale each of the cranking VE map accordingly, and make the percentage scale linear between those three rpm points. For example, if on my main VE table if the values differ by 30% from stock a 400, and the values differ by 40% from stock at 800, and the values differ by 20% (I'm just throwing numbers out for an example). Then would I scale the cranking VE table at those rpms by those percentages, and then try to scale it linearly between, so between 400 and 800 it would get scaled by 35%, etc.

2. My next question is w/o usage of an eddy current dyno, and sitting at high rpm points and high load for extended periods of time, how do you map those cells? Thus far to get all the cells I have, I used a lot of brake pedal modulation to hold me in those rpm cells while I kept the throttle in one particular MAP cell. My brakes just arent' up to the task of holding back near WOT at high rpm areas.

3. When taking off from a stop, I often get some KR (up to 4 or 5 degrees sometimes). This occurs right after I let out the clutch, and start rolling. Once I hit ~1800-2000 rpms with the vehicle speed around 10-15 mph, it gets some KR. I've looked at the logs, and sometimes it looks lean compared to the command, and sometimes its not. I targeted those cells paricularly hard when I was doing my last round of auto VE, those were the cells that corresponded to BENs over 1.0. That range was between BENs of 1.03 and 1.07. I hope that takes care of it. If you'd like, take a look at my before and after tunes, and the logs that lead me to make those changes.

4. I often stall when the car comes down to idle, mostly when stationary, sometimes when coasting to a stop. When the car gets very close to a stop, then it wants to idle very rough, and sometimes it will save itself, sometimes not. I have played with the overspeed/underspeed timing corrections a bit, but not sure I have enough. Can someone with a trained eye take a look and see if maybe I'm just not going agressive enough. Also, I made some changes to the throttle cracker, based upon the idle tuning tutorial in the stickied in the tutorials section. I have not drove the car since those changes. I hope that helps.

5. There is a table in the Fuel->Mixture folder, B3609 called "PE Delay Counter Adjustment". Stock my table is all filled wiht ones. I looked at many of the modified tunes on holdencrazy and it looked like all the other tables were ones. Joecar suggested that I make that table all zeros. Does anyone else have experience with this? I'm unsure of how this table works with the single adjustment parameter B3608 called "Delay before Engering PE Mode"

6. There are areas in my main VE talbe that are over 100, sometimes up to hte 105 range. I know these number arne't truely a volumetric efficiency, but does this seem realistic?

I would very much appreciate any and all help on these questions and concerns.

I have attached several items. Two are tune files. The first one is the before tune, then I have several logs. One log is about an hour long and consists of highway driving at varying loads trying to be VERY smooth on the throttle as much as possible. Then there is a log where I wanted to target idle BENs, so I did that seperately. Then there is a log with a fewll WOT or near WOT pulls. I have also attached the tune file I applied changes to based upon those three logs.

My basic modifications are MS4 cam, dual 3.5" intake, LS6 manifold, 1 7/8" long tubes, dual 3" exhuast, M6 trans, in a 2800 lb FD RX-7.


Once again it appears that there is a problem with the database and I cannot upload the files. I will upload them as soon as it allow me to. In case anyone wants to take a look before I am able to upload the files, I can email them as well. I'm near a computer 95% of the day, so I can email probalby within 5 minutes of you giving me your email address.

Thanks,
Kevin

[Kevin Doe]

I was able to upload most of the files to ls1tech. The only one that I couldn't upoad was the hour long log I used for the main VE modifications. The file size was too large. See the rest of the attachments here:

http://www.ls1tech.com/forums/showthread.php?p=8046930#post8046930

I can still email anyone who wants to see the hour long VE tunin log. Just let me know and I'll email it asap.

[SSpdDmon]

email me a copy of your tune to saj79 at aol dot com and I'll take a look if I get a minute this weekend.

[Kevin Doe]

Emailed.

Another thing I noticed when looking at log_0053. At the beginning, right when I begin to move, it appears as if it enters PE mode, since the commanded AFR goes from 14.68 down into the 12-13s. I looked at the tune, and I don't understand. The MAP enable is set to 15 kPa, so thats alwasy met, but during that part of the log, the TPS is not above the threshold to enter PE mode. Why is it entering PE mode there. I want to have a better understanding of what is going on.

[SSpdDmon]

Ok guys, I think I am nearly complete with my auto ve tuning. My last log had bens mostly in the 98-99 range, and a few cells with a bit more correction than that.

1. My main VE table is something on the order of 40% less in the idle region than stock, and my hot starts are not that great. For a hot/warm start I have to push the throttle to get it to start. If I compare the 800 rpm and 400 rpm cells of my main VE to the cranking VE table, the values differ significantly. I know I need to modify the table, but how it is usually done? If my 800 rpm column is off by say 40% approximately, do I scale the entire VE map by 40%, or do I use the 400,800,and 1200 rpm columns and scale each of the cranking VE map accordingly, and make the percentage scale linear between those three rpm points. For example, if on my main VE table if the values differ by 30% from stock a 400, and the values differ by 40% from stock at 800, and the values differ by 20% (I'm just throwing numbers out for an example). Then would I scale the cranking VE table at those rpms by those percentages, and then try to scale it linearly between, so between 400 and 800 it would get scaled by 35%, etc.

That's an airflow issue. I left the cranking VE alone and made the changes to {B4343} Startup Friction Airflow Correction.

2. My next question is w/o usage of an eddy current dyno, and sitting at high rpm points and high load for extended periods of time, how do you map those cells? Thus far to get all the cells I have, I used a lot of brake pedal modulation to hold me in those rpm cells while I kept the throttle in one particular MAP cell. My brakes just arent' up to the task of holding back near WOT at high rpm areas.

Sometimes it's hit or miss. Trying to hit one cell is hard enough...but, you really want data from the center of the cells. Set your filters wisely and this can be achieved. If I get time later, I'll see if I can pull my filter data.

3. When taking off from a stop, I often get some KR (up to 4 or 5 degrees sometimes). This occurs right after I let out the clutch, and start rolling. Once I hit ~1800-2000 rpms with the vehicle speed around 10-15 mph, it gets some KR. I've looked at the logs, and sometimes it looks lean compared to the command, and sometimes its not. I targeted those cells paricularly hard when I was doing my last round of auto VE, those were the cells that corresponded to BENs over 1.0. That range was between BENs of 1.03 and 1.07. I hope that takes care of it. If you'd like, take a look at my before and after tunes, and the logs that lead me to make those changes.

BEN's at idle are going to be inaccurate with the duration of your cam, which means reverting back to closed loop most likely won't be an option for you. As for fueling, keep an eye on the WB. 15:1~15.5:1 should be ok for idle.

4. I often stall when the car comes down to idle, mostly when stationary, sometimes when coasting to a stop. When the car gets very close to a stop, then it wants to idle very rough, and sometimes it will save itself, sometimes not. I have played with the overspeed/underspeed timing corrections a bit, but not sure I have enough. Can someone with a trained eye take a look and see if maybe I'm just not going agressive enough. Also, I made some changes to the throttle cracker, based upon the idle tuning tutorial in the stickied in the tutorials section. I have not drove the car since those changes. I hope that helps.

The idle controls were adjusted to help get you in the right direction for this (too many to list).

5. There is a table in the Fuel->Mixture folder, B3609 called "PE Delay Counter Adjustment". Stock my table is all filled wiht ones. I looked at many of the modified tunes on holdencrazy and it looked like all the other tables were ones. Joecar suggested that I make that table all zeros. Does anyone else have experience with this? I'm unsure of how this table works with the single adjustment parameter B3608 called "Delay before Engering PE Mode"

Not sure if it's a source of problems or not. But, I reduced it anyway. Can't hurt.

6. There are areas in my main VE talbe that are over 100, sometimes up to hte 105 range. I know these number arne't truely a volumetric efficiency, but does this seem realistic?

Change your options to show VE as grams*K/kPa. Then, you don't have to worry about percentages anymore. :)

I would very much appreciate any and all help on these questions and concerns.

I have attached several items. Two are tune files. The first one is the before tune, then I have several logs. One log is about an hour long and consists of highway driving at varying loads trying to be VERY smooth on the throttle as much as possible. Then there is a log where I wanted to target idle BENs, so I did that seperately. Then there is a log with a fewll WOT or near WOT pulls. I have also attached the tune file I applied changes to based upon those three logs.

My basic modifications are MS4 cam, dual 3.5" intake, LS6 manifold, 1 7/8" long tubes, dual 3" exhuast, M6 trans, in a 2800 lb FD RX-7.


Once again it appears that there is a problem with the database and I cannot upload the files. I will upload them as soon as it allow me to. In case anyone wants to take a look before I am able to upload the files, I can email them as well. I'm near a computer 95% of the day, so I can email probalby within 5 minutes of you giving me your email address.

Thanks,
Kevin

Try this out and let me know if it's any better. A quick question first?

Why was WOT timing so low? 26~27* should be a safe number for that cam unless you've got some limiting factors like more compression, octane restrictions, etc.

I lowered the timing a little more just off idle to help with your initial KR. Let's see how it does...

As for your commanded AFR - remember you're in OLSD. It might not have been PE changing your AFR...rather the Commanded Fuel in Open Loop Table. I readjusted that table to 14.63 across the board at operating temps. As long as PE is enabled, it'll kick on....so don't worry about lean AFR's at higher MAP pressures.

-Jeff

*PM sent too*

[Kevin Doe]

Try this out and let me know if it's any better. A quick question first?

Why was WOT timing so low? 26~27* should be a safe number for that cam unless you've got some limiting factors like more compression, octane restrictions, etc.

I'm running 93 octane only. I had it very low to be overly safe during the tune phase, I was going to slowly ramp it up to look like what you gave me. I'm new to the LS1s, so I was unsure of how much timing they liked, so I went overly conservative. My compression is stock.

I lowered the timing a little more just off idle to help with your initial KR. Let's see how it does...

As for your commanded AFR - remember you're in OLSD. It might not have been PE changing your AFR...rather the Commanded Fuel in Open Loop Table. I readjusted that table to 14.63 across the board at operating temps. As long as PE is enabled, it'll kick on....so don't worry about lean AFR's at higher MAP pressures.

It looked like I had a misconception of how the PE worked. I thougth everything in the mixture folder was PE. It now makes sense that I know the PE is actually only the curves and not the commanded fuel in open loop table. Thanks for the clarification. Looks like you made some pretty wholesale changes in my tune. I will give this a try when I get home and report back with my results.

-Jeff

*PM sent too*

Jeff, Thanks for all the help. I greatly appreciate it. I'll let you and everybody else know my findings! Thanks.

I did some reading about some people reducing burst knock, and even turning it off. It looks like you did the opposite, turned up the amount of timing it pulls, and slowed the decay rate. What is your reasoning for this, I'd like to understand your logic, and why it differs from opinions that I've read.

Also, could you give more explanation on the cranking VE table, and why you wouldn't change it. It seems like the physical airflow characterisics have changed with the camshaft, intake, and exhaust. The main VE table changed signifcanly in the low rpm range, so to me it seems to make sense that the cranking VE table shoudl be lowered as well. Are you sure you're not covering up the cranking VE table being wrong by increasing the friction? I guess if it works it works, as they say the proof is in the pudding. But if you could elaborate I would appreciate that.

[SSpdDmon]

Some have seen it as a problem. I see it as a friend...a friend I'd like to stick around a little longer. :lol:

I didn't increase the timing that was pulled for burst knock. I only asked it to stick around a little longer....like a tenth of a second. I've found on my h/c application that I had....having it stick around a little longer helped. After all, the engineers put it there for a reason - so why get rid of it?

I didn't touch the delta airmass settings though. If you find that your commanded timing is not matching the actual, the delta airmass settings may be too aggressive even though they're still stock. This means you'll probably have to increase the B5951 multiplier by 1~2 full points.


As for the rest of the timing, just keep an eye out for KR at WOT - especially with the changes I made. Again, I don't know why your timing was only at 21~22* instead of the 27~28* I usually see in cam'd LS1's.

[Kevin Doe]

^ Read the bolded parts I replied to in yoru quote in post #6 for explanation of why the timing was low. I was basically saying that I am not familiar with LS1s and how much timing they like, so I made it overly conservative to be on the safe side.

How much if any KR, if any, is acceptable on WOT?

Also, do you have any thoughts on this:
Also, could you give more explanation on the cranking VE table, and why you wouldn't change it. It seems like the physical airflow characterisics have changed with the camshaft, intake, and exhaust. The main VE table changed signifcanly in the low rpm range, so to me it seems to make sense that the cranking VE table shoudl be lowered as well. Are you sure you're not covering up the cranking VE table being wrong by increasing the friction? I guess if it works it works, as they say the proof is in the pudding. But if you could elaborate I would appreciate that.

[SSpdDmon]

^ Read the bolded parts I replied to in yoru quote in post #6 for explanation of why the timing was low. I was basically saying that I am not familiar with LS1s and how much timing they like, so I made it overly conservative to be on the safe side.

How much if any KR, if any, is acceptable on WOT?

Also, do you have any thoughts on this:
Also, could you give more explanation on the cranking VE table, and why you wouldn't change it. It seems like the physical airflow characterisics have changed with the camshaft, intake, and exhaust. The main VE table changed signifcanly in the low rpm range, so to me it seems to make sense that the cranking VE table shoudl be lowered as well. Are you sure you're not covering up the cranking VE table being wrong by increasing the friction? I guess if it works it works, as they say the proof is in the pudding. But if you could elaborate I would appreciate that.
Gotcha...little blind this morning. :lol:

Ideally, you want the least amount of timing that produces the most amount of power. That means, ideally, you need a dyno to check what timing your engine likes best. If KR is present, then you may have to consider false knock, bad gas, faulty sensor, etc. But, I'd be very confident in saying that 27* isn't too much timing.

As for the cranking VE, you're only there for a split second or two. I never seen it as a problem leaving it stock. If you want to reduce it, I'd make the changes similar to your normal VE. In other words, if the 800rpm row is ~40% below stock, then reduce the 800rpm row by 40%. Do the same for the 400rpm and 1200rpm rows and blend in the rest. But, I still say...if it's not a problem, don't mess with it. Your starting issues should be resolved after the changes I made.

[Kevin Doe]

Ok, I loaded up the tune SSpdDmon suggested. I ran a cold startup and did the RAFIG process, then added those numbers to the desired airflow, reflashed and took a drive. I have a few observations in addition to my log.

1. The car pulls more vacuum at idle at 900 rpms than it did at 850 rpms. Not surprise here, but now I'm at ~65 kPa.

2. During the cold startup the commanded AFR was in the 12s or 13s, and the WBO2 feedback was in the 14s. Once it warmed up, the two matched within a few percent. My VE values at idle seem to vary from day to day by a few percent.

3. When comming to a stop, if I let the car coast down in gear, there was much less bucking than before. There was also less bucking on very light throttle low rpm, creeping around in 1st gear.

4. When comming to a stop, the car seemed to have less tendancy to stall. Before I would stall often comming to a stop. Now, it seems as if it has a tendancy to stall if I push the clutch in while hte rpms are greater than say 2000 or 2500. A few times it dropped down to 300 rpms for a second, then saved itself. One time it idled at ~400-500 rpms for several seconds before comming back up to 900 rpms. If I coast down to a stop while in gear, then push in the clutch when the rpms are near idle, it comes into idle great.

5. I still get part throtle knock, possibly a touch less than before, its hard to tell. I see 1-3 degrees of KR around 1800-2000 rpms in 1st gear taking off from a stop. I'm not sure what to think about this.

6. No WOT or high throttle KR at all.



I've attached two logs. The first one is the cold start log. This log shows the command showing in the 12s or 13s, and the actual in the 14s during warmup.

The second log is a quick 10 minute drive with lots part throttle and also some 2nd gear and part 3rd gear WOT pulls.

On the second log, here are some areas of interest
This frame shows where I pushed in the clutch rolling to a stop and it idled at 400 rpms for a while before the timing jumped up and pulled it back to 900 rpms. Not sure why it took so long for the timin to jump up and bring it back to 900 rpms. Starts at frame 950.
These frames are where I see part throttle knock: 1570, 1960, 5810
Thes frames are the WOT frames: 2140, 2650, 2860

[SSpdDmon]

Ok, I loaded up the tune SSpdDmon suggested. I ran a cold startup and did the RAFIG process, then added those numbers to the desired airflow, reflashed and took a drive. I have a few observations in addition to my log.

1. The car pulls more vacuum at idle at 900 rpms than it did at 850 rpms. Not surprise here, but now I'm at ~65 kPa.

2. During the cold startup the commanded AFR was in the 12s or 13s, and the WBO2 feedback was in the 14s. Once it warmed up, the two matched within a few percent. My VE values at idle seem to vary from day to day by a few percent.

This is normal. Worry more about what the car does at operating temps.

3. When comming to a stop, if I let the car coast down in gear, there was much less bucking than before. There was also less bucking on very light throttle low rpm, creeping around in 1st gear.

:D

4. When comming to a stop, the car seemed to have less tendancy to stall. Before I would stall often comming to a stop. Now, it seems as if it has a tendancy to stall if I push the clutch in while hte rpms are greater than say 2000 or 2500. A few times it dropped down to 300 rpms for a second, then saved itself. One time it idled at ~400-500 rpms for several seconds before comming back up to 900 rpms. If I coast down to a stop while in gear, then push in the clutch when the rpms are near idle, it comes into idle great.

Try adding a little more across the top row of the throttle cracker (400rpm). I think I had you at 3grams/sec. Try 5grams/sec just for that row. While rolling, the idle spark overspeed/underspeed corrections are disabled. That's why it's a little slower to respond.

The other thing you can do is log your throttle follower. If the rpms are dropping too fast, the decay rates I gave you may be too high.

5. I still get part throtle knock, possibly a touch less than before, its hard to tell. I see 1-3 degrees of KR around 1800-2000 rpms in 1st gear taking off from a stop. I'm not sure what to think about this.

Log where this happens several times. If it's consistently in the same spot of the timing table, try pulling a little more timing there.

6. No WOT or high throttle KR at all.

:D

I've attached two logs. The first one is the cold start log. This log shows the command showing in the 12s or 13s, and the actual in the 14s during warmup.

The second log is a quick 10 minute drive with lots part throttle and also some 2nd gear and part 3rd gear WOT pulls.

On the second log, here are some areas of interest
This frame shows where I pushed in the clutch rolling to a stop and it idled at 400 rpms for a while before the timing jumped up and pulled it back to 900 rpms. Not sure why it took so long for the timin to jump up and bring it back to 900 rpms. Starts at frame 950.
These frames are where I see part throttle knock: 1570, 1960, 5810
Thes frames are the WOT frames: 2140, 2650, 2860

Glad it helped... :)