Still get oscillation coming to a stop
01-20-2017, 12:31 PM
#101
Moderator
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,446
Received 1,836 Likes
on
1,145 Posts
Took me a while to really digest your post. It was very well thought out and does illustrate the circular nature of idle tuning - one change necessitates another, which eventually affects the original change.
You'd think that if you just ran it at zero IAC warm idle, it would never dip, but then, the computer cuts fuel and timing, which makes it dip even worse. My blade is set to 0.67 volts, and I have a hole in the blade, and I idle in the 20's on IAC counts - if that helps you any. i've read that 0.76 volts is the limit for the ability to zero the TPS. I cannot confirm the truth of that, though.
A few things I learned trying to dial in the MSD intake - which completely flipped out all my previous idle tuning. Some of them absurdly simple:
* I had a weird situation where I had tons of follower and cracker air in, and it would hold 1600-2000 from any speed until dead stop, settle to an idle, and never come close to stalling. However, if I let the car just sit and bark for a few minutes, then pull forward slowly a few hundred feet and stop, it stalls every time. And I finally figured it out. Once I get out of adaptive idle, it idles high and does everything properly. If it never gets out of adaptive idle, that's when it dips and sometimes dies. The solution ended up being almost absurdly simple. Proportional Idle airflow. When the idle dipped down, it was only adding .1-.4 g/sec. I put 1 g/sec in at -150 rpm error, and 2 g/sec in the farthest cell, interpolated. Now, it rarely dips below 800, much less comes close to stalling. I might even increase it more. I did this in all three idle speed low proportional airflow tables, and it never gave me that problem regardless of AC compressor, clutch, etc. Bottom line though - no amount of throttle or cracker will completely correct for poorly functioning or unoptimized adaptive idle. I just had to figure out that it was the adaptive idle routines that were problematic.
* I compressed the lower portion of the VE table. The MSD idles at 65-70 KPa. I took the VE from 0-1200 RPM in my 70 row and copied it to my 0 row, multiplied by 0.80, and then interpolated. This served to greatly stabilize the dynamic airflow calcs at idle - resulting in smoother fueling calculations at idle.
* I now have a new theory, but I'm not sure quite how to test it. I'm thinking that if one is able to perfectly dial in the adaptive idle routines, the follower and cracker might become unnecessary. The hard part is testing this. Disable the TC and TF uncovers the problems, but this makes it very difficult to keep running long enough to dial in the adaptive idle. I can report that having increased the proportional idle correction like I did, I was able to greatly reduce the cracker air adder and follower air adder.
You'd think that if you just ran it at zero IAC warm idle, it would never dip, but then, the computer cuts fuel and timing, which makes it dip even worse. My blade is set to 0.67 volts, and I have a hole in the blade, and I idle in the 20's on IAC counts - if that helps you any. i've read that 0.76 volts is the limit for the ability to zero the TPS. I cannot confirm the truth of that, though.
A few things I learned trying to dial in the MSD intake - which completely flipped out all my previous idle tuning. Some of them absurdly simple:
* I had a weird situation where I had tons of follower and cracker air in, and it would hold 1600-2000 from any speed until dead stop, settle to an idle, and never come close to stalling. However, if I let the car just sit and bark for a few minutes, then pull forward slowly a few hundred feet and stop, it stalls every time. And I finally figured it out. Once I get out of adaptive idle, it idles high and does everything properly. If it never gets out of adaptive idle, that's when it dips and sometimes dies. The solution ended up being almost absurdly simple. Proportional Idle airflow. When the idle dipped down, it was only adding .1-.4 g/sec. I put 1 g/sec in at -150 rpm error, and 2 g/sec in the farthest cell, interpolated. Now, it rarely dips below 800, much less comes close to stalling. I might even increase it more. I did this in all three idle speed low proportional airflow tables, and it never gave me that problem regardless of AC compressor, clutch, etc. Bottom line though - no amount of throttle or cracker will completely correct for poorly functioning or unoptimized adaptive idle. I just had to figure out that it was the adaptive idle routines that were problematic.
* I compressed the lower portion of the VE table. The MSD idles at 65-70 KPa. I took the VE from 0-1200 RPM in my 70 row and copied it to my 0 row, multiplied by 0.80, and then interpolated. This served to greatly stabilize the dynamic airflow calcs at idle - resulting in smoother fueling calculations at idle.
* I now have a new theory, but I'm not sure quite how to test it. I'm thinking that if one is able to perfectly dial in the adaptive idle routines, the follower and cracker might become unnecessary. The hard part is testing this. Disable the TC and TF uncovers the problems, but this makes it very difficult to keep running long enough to dial in the adaptive idle. I can report that having increased the proportional idle correction like I did, I was able to greatly reduce the cracker air adder and follower air adder.
Last edited by Darth_V8r; 01-20-2017 at 02:07 PM. Reason: added more stuff
01-20-2017, 02:59 PM
#102
Hmm. I just need to know where the blade is optimally set lol. That becomes the circular nature. But I get where you are coming from. I'm surprised the MSD changed things so much. But the intake has a lot to do with the quality of the idle. And a FAST changes it over an stock manifold.
I have talked to James Short some about this and may just let him give it a go. He spends time on tuning the RAF/IAC Table, bottom of the VE, and adaptive idle when he tunes cars. So I may see what he can do and what (if anything) he does that's different than what I've been trying to do. It's very close, but with a cam this size, it should be perfect. And that's what I want.
I have talked to James Short some about this and may just let him give it a go. He spends time on tuning the RAF/IAC Table, bottom of the VE, and adaptive idle when he tunes cars. So I may see what he can do and what (if anything) he does that's different than what I've been trying to do. It's very close, but with a cam this size, it should be perfect. And that's what I want.
01-20-2017, 09:23 PM
#103
Staging Lane
Join Date: Nov 2015
Posts: 96
Likes: 0
Received 0 Likes
on
0 Posts
The blade should be set so to get you in the range of 30-50 IAC counts on a fully warm engine. With a FAST throttle body I would set a minimum of 40 IAC counts if the IAC port hasn't been hogged out. You then adjust the IAC effective area table so that your desired airflow matches your actual airflow.
01-20-2017, 10:04 PM
#105
Staging Lane
Join Date: Nov 2015
Posts: 96
Likes: 0
Received 0 Likes
on
0 Posts
01-21-2017, 08:26 PM
#107
Where it's open now is pretty close. Idles in the 30 counts hot. But still is a wavy line coming to a stop... IAC Effective area is close too. So I changed the base airflow. And added .5g increments during the log. Nothing seemed to make a difference. The baseline is close enough.
01-21-2017, 11:19 PM
#108
Moderator
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,446
Received 1,836 Likes
on
1,145 Posts
not sure how helpful this observation is, but I only see it doing the oscillation thing at the end of the log. in frame 19:53.561, the follower air is 0.8 g/s. The very next frame, 19:53.571, follower air is zero and never comes back up. Right then also, the fuel trims are seriously high, and the o2 mv readings go high and stay high until well after it stabilizes, at which point it starts trimming properly, which is right before you shut it down. Looks almost like the follower closed too fast, which put the whole system into a loop for a few oscillations. In trying to see what event happened first, between timing, fuel, and air, the follower instantly dropping to zero looks like the first event.
01-21-2017, 11:51 PM
#109
Yeah, I don't know. I might need to take the edge off the follower. It closes down almost instantly. Relying on cracker to save it.
But the oscillation is every time I come to a stop. It's not much mind you. But the loopy-loo dip is there. And that's after the TB has been shut for a while. So the follower would have decayed by then. It's just annoying cuz you can feel the engine bog and then bounce back. Like in my Honda and Toyota it does the same thing but rebounds instantly with electronic TB (I also image the newer LS cars do that too). But it shouldn't bog. That bog is what I don't like. I can probably stall it if I wanted to in a bog scenario. I just want the car to hold 1100 RPM until I hit zero. I may raise idle to 950 and see if that helps.
But the oscillation is every time I come to a stop. It's not much mind you. But the loopy-loo dip is there. And that's after the TB has been shut for a while. So the follower would have decayed by then. It's just annoying cuz you can feel the engine bog and then bounce back. Like in my Honda and Toyota it does the same thing but rebounds instantly with electronic TB (I also image the newer LS cars do that too). But it shouldn't bog. That bog is what I don't like. I can probably stall it if I wanted to in a bog scenario. I just want the car to hold 1100 RPM until I hit zero. I may raise idle to 950 and see if that helps.
01-22-2017, 11:59 AM
#111
Well I want it above idle. It dips below 850. So I want it to stay at least above that and then settle into the idle.
When I drove it the other day with pretty much these settings, I stalled it twice in traffic. It shouldn't be stalling. But if I have to make quick reactions to idiot drivers and slam on my brakes, I shouldn't be able to pull the engine down to nothing. That's not safe at all and makes me not want to drive the car at all. I can't avoid traffic unless I trailer it around and just use it at the track. But that's not what the car is for.
When I drove it the other day with pretty much these settings, I stalled it twice in traffic. It shouldn't be stalling. But if I have to make quick reactions to idiot drivers and slam on my brakes, I shouldn't be able to pull the engine down to nothing. That's not safe at all and makes me not want to drive the car at all. I can't avoid traffic unless I trailer it around and just use it at the track. But that's not what the car is for.
01-22-2017, 08:34 PM
#112
Staging Lane
Join Date: Nov 2015
Posts: 96
Likes: 0
Received 0 Likes
on
0 Posts
I've had a look at your log and I can see the oscillation at the end of the log. Look at your MAP readings... they're crazy. Your IAC counts are going berserk. Now look at the IAC effective area in your tune file. There's 10 IAC counts between 58mm and 60mm, 6 IAC counts between 62mm and 64mm, 4 IAC counts between 70mm and 72mm... you can't just put random numbers in there and expect it to work properly. There's no way that you can even tell now much of the overshoot is actual overshoot and how much is bad IAC effective area table.
You can't fix this by just adding some air to your cracker table. I don't want to sound nasty, but give it to someone who knows what they're doing.
You can't fix this by just adding some air to your cracker table. I don't want to sound nasty, but give it to someone who knows what they're doing.
01-22-2017, 10:45 PM
#113
It's not random. Those are the GM numbers just moved to the right. I haven't seen anything on how to tune the numbers within the IAC table. Most don't mess with it at all. But I think that's the trick to getting a 102 to behave since it's not even close to stock.
I've had three tuners work on it. And here I am.
I am trying to learn on something on the fringe like this. I'd love to be able to log the IAC counts and map them correctly. I haven't seen anything on that anywhere. Because again, most don't mess with it and live with poor drivability with these TBs or are in the business of making money and don't want to explain how they do it. Or, make money and don't mess with it and don't care. Want stock drivability? Go stock I hear. That's a crappy answer. People make these things work. Not a lot mind you. But enough that it can be done.
But I'm going to have James Short try. He does tune the IAC and map those things. Maybe he can fix it? If not, I'll probably just pull the TB/FAST102 off and sell it. I'm supercharging it anyway. I don't care about the manifold or **** poor drivability when boost covers up the short comings of an LS6 or LS2 intake.
A lot of people mod these cars and then sell them. The modding is the fun part. Living with them modded is not.
I've had three tuners work on it. And here I am.
I am trying to learn on something on the fringe like this. I'd love to be able to log the IAC counts and map them correctly. I haven't seen anything on that anywhere. Because again, most don't mess with it and live with poor drivability with these TBs or are in the business of making money and don't want to explain how they do it. Or, make money and don't mess with it and don't care. Want stock drivability? Go stock I hear. That's a crappy answer. People make these things work. Not a lot mind you. But enough that it can be done.
But I'm going to have James Short try. He does tune the IAC and map those things. Maybe he can fix it? If not, I'll probably just pull the TB/FAST102 off and sell it. I'm supercharging it anyway. I don't care about the manifold or **** poor drivability when boost covers up the short comings of an LS6 or LS2 intake.
A lot of people mod these cars and then sell them. The modding is the fun part. Living with them modded is not.
01-22-2017, 11:43 PM
#114
Staging Lane
Join Date: Nov 2015
Posts: 96
Likes: 0
Received 0 Likes
on
0 Posts
I am trying to learn on something on the fringe like this. I'd love to be able to log the IAC counts and map them correctly. I haven't seen anything on that anywhere. Because again, most don't mess with it and live with poor drivability with these TBs or are in the business of making money and don't want to explain how they do it. Or, make money and don't mess with it and don't care.
Engine stalling is a safety issue and I try to help people with issues like that for free... but your issue is not a simple fix.
I don't know how good James Short is, but in your situation you should probably give him a try. I know you've already had three tuners work on it, but a stalling issue is an issue which really does need to be fixed.
Yes, that is a crappy answer. You can get good drivability with mods but it's not always a simple thing to do. The more you change, the more the tune needs to change. Just doing a cam swap usually takes less tuning than doing H/C/I. People do make these things work, but sometimes it takes more knowledge than most backyard tuning enthusiasts have.
01-23-2017, 08:01 AM
#115
I've run a NW 102 TB on a ported Fast with good results & yes the 102 TB tests the tuner 
I've re cam'd three times looking for perfect manners, starting @17*, 10.5* & now 4.5* of overlap. Even @ 17* idle & return to idle with help we were able to get it right, the bucking low gear light load was not acceptable.
Here is a link to a HPT thread that really helped me with the 102 TB, it includes DBW & clothes line TB discussion. Hope it helps.
I'm sure James will nail it down quickly, very smart guy. I've always enjoyed statesman posts & readily agree with his comments about tuners that earn their livelihood from this. I'm a backyard home built tuner, I've no issue paying for wisdom of others.
http://www.hptuners.com/forum/showth...ired-AF-vs-MAF
I've re cam'd three times looking for perfect manners, starting @17*, 10.5* & now 4.5* of overlap. Even @ 17* idle & return to idle with help we were able to get it right, the bucking low gear light load was not acceptable.Here is a link to a HPT thread that really helped me with the 102 TB, it includes DBW & clothes line TB discussion. Hope it helps.
I'm sure James will nail it down quickly, very smart guy. I've always enjoyed statesman posts & readily agree with his comments about tuners that earn their livelihood from this. I'm a backyard home built tuner, I've no issue paying for wisdom of others.
http://www.hptuners.com/forum/showth...ired-AF-vs-MAF
01-23-2017, 08:34 AM
#116
I know. I'm not looking for that knowledge for free. But some idea of what's happening. It's the same thing in every thread. But I will look at that thread and see. Waiting to hear back from James.
Re: thread - I hate that none of the old files are on the forums anymore. Would be nice to see the cfg files.
Re: thread - I hate that none of the old files are on the forums anymore. Would be nice to see the cfg files.
Last edited by JakeFusion; 01-23-2017 at 08:40 AM.
01-23-2017, 11:54 AM
#117
Moderator
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,446
Received 1,836 Likes
on
1,145 Posts
So, I ran into this post on a different forum:
Jake, this might be worth checking in your logs - if after extensive idling, and your idle trims are negative, does the car latch onto that number instead of zeroing it out? If this is the case, NO amount of idle air fiddling, base airflow, cracker, etc, will get rid of it. Seems like int he quoted post, the car latches onto the idle trims until the next time it goes into idle mode instead of zeroing as soon as it leaves idle mode.
Hey guys,
I am tuning the car. The biggest details of what is going on were left out here...
Thanks for the idle airflow lesson However the problem is, even with STIT's set to ZERO as max (and LTIT's the same), the car carries the STIT offset with it everywhere. What I mean is.... Let's say the car is adding 0.2 g/sec idle airflow and you get it moving.... instead of ZERO'ing as it should, the STIT LOCKS into it's last value and holds that the whole time the car is moving. When you come to a stop, all of a sudden it snaps back to zero and they begin updating again. If you take off with the car adding a tenth or two, the dip at clutch in is almost nothing. However, if the current conditions dictate that the car is removing idle airflow (say -1g) and you take off, that -1g is ALWAYS being subtracted, clutching in with that -1 offset causes a dip. You can drive for 100 miles and until you stop, the locked STIT will continue to act.
This car came from being tuned in another place. The rest of the tune was not the issue, just the dip. I thought something was wrong in the file in a place I could not get to. I flashed the car back to 100% stock, moved in the relevant calibration portions and the problem persists. The car is a 2000. I *THEN* did a full flash of the PCM as a 99, moved over most of the cal data and disabled the security as a test.... it even does this with the 99 file.
The STIT's lock when the car moves instead of zero'ing. It's very clear WHAT the issue is, but not clear why right now.
I am tuning the car. The biggest details of what is going on were left out here...
Thanks for the idle airflow lesson However the problem is, even with STIT's set to ZERO as max (and LTIT's the same), the car carries the STIT offset with it everywhere. What I mean is.... Let's say the car is adding 0.2 g/sec idle airflow and you get it moving.... instead of ZERO'ing as it should, the STIT LOCKS into it's last value and holds that the whole time the car is moving. When you come to a stop, all of a sudden it snaps back to zero and they begin updating again. If you take off with the car adding a tenth or two, the dip at clutch in is almost nothing. However, if the current conditions dictate that the car is removing idle airflow (say -1g) and you take off, that -1g is ALWAYS being subtracted, clutching in with that -1 offset causes a dip. You can drive for 100 miles and until you stop, the locked STIT will continue to act.
This car came from being tuned in another place. The rest of the tune was not the issue, just the dip. I thought something was wrong in the file in a place I could not get to. I flashed the car back to 100% stock, moved in the relevant calibration portions and the problem persists. The car is a 2000. I *THEN* did a full flash of the PCM as a 99, moved over most of the cal data and disabled the security as a test.... it even does this with the 99 file.
The STIT's lock when the car moves instead of zero'ing. It's very clear WHAT the issue is, but not clear why right now.
01-23-2017, 12:05 PM
#118
I don't log the LTITs. I'll start doing that.
But as that thread from HPT above shows, you can get it very close if you can tune the IAC table. So even if it locks in some weird data, if it's locking in basically zero, it won't cause an issue. But I don't know if that's happening or not. Need to log it.
But as that thread from HPT above shows, you can get it very close if you can tune the IAC table. So even if it locks in some weird data, if it's locking in basically zero, it won't cause an issue. But I don't know if that's happening or not. Need to log it.
02-10-2017, 07:22 PM
#119
So, I've done a couple of things. James Short has helped as well. We have the IAC dialed in now and airflow within .1g/sec everywhere.
Changed the timing up a bit in the idle tables, zero'd out the cracker, leaned out idle a bit, went OL at idle (CL above 1200rpm), and worked on the Follower a lot. James had me rev in gear with brakes engaged and I've been adding quite a bit of follower airflow (while also adding a lot of decay) to make it so it lands softly. It was undershooting idle and dying out. As James told me, catching idle in gear is the hardest thing for the Gen III PCM to do. So if you can get it to catch idle there, it'll never stall.
It drives much better now because of it. No more surging coming to a stop. And no more dying out with quick lifts of throttle (which is how I kept stalling before in traffic). I just need to tweak it a bit so it doesn't add so much airflow on slight throttle tip-in (causes a little bit of a surge) and doesn't fall below 700rpm as it sometimes does. The nice thing is, there are airflow by TPS% and multiplier by RPM, so I've worked mostly on the multiplier and decay rates to get it to perform like I need. It's probably 3-5X where it is stock, which seems about right considering when the 102 closes, it cuts off a lot of airflow. And the idle airflow is nearly twice a stock vehicle.
Changed the timing up a bit in the idle tables, zero'd out the cracker, leaned out idle a bit, went OL at idle (CL above 1200rpm), and worked on the Follower a lot. James had me rev in gear with brakes engaged and I've been adding quite a bit of follower airflow (while also adding a lot of decay) to make it so it lands softly. It was undershooting idle and dying out. As James told me, catching idle in gear is the hardest thing for the Gen III PCM to do. So if you can get it to catch idle there, it'll never stall.
It drives much better now because of it. No more surging coming to a stop. And no more dying out with quick lifts of throttle (which is how I kept stalling before in traffic). I just need to tweak it a bit so it doesn't add so much airflow on slight throttle tip-in (causes a little bit of a surge) and doesn't fall below 700rpm as it sometimes does. The nice thing is, there are airflow by TPS% and multiplier by RPM, so I've worked mostly on the multiplier and decay rates to get it to perform like I need. It's probably 3-5X where it is stock, which seems about right considering when the 102 closes, it cuts off a lot of airflow. And the idle airflow is nearly twice a stock vehicle.
02-11-2017, 07:33 PM
#120
Moderator
Glad to hear of your success as I will follow in your footsteps as soon as the Michigan Winter allows.
Since I have the same "FAST LSXR 102/102 TB" as you, I assume I can use the same IAC tables? Or does it depend upon the size of any hole in the TB plate?
Has anyone optimized the Transient tables for the FAST LSXR?
Since I have the same "FAST LSXR 102/102 TB" as you, I assume I can use the same IAC tables? Or does it depend upon the size of any hole in the TB plate?
Has anyone optimized the Transient tables for the FAST LSXR?
Last edited by mrvedit; 02-18-2017 at 05:16 PM. Reason: Typo