Tuning with MAF on a 20V 4AGE

All Megasquirt 1 support questions. See also the Documentation

Moderators: jsmcortina, muythaibxr

suberimakuri
Master MS/Extra'er
Posts: 632
Joined: Sun May 02, 2004 2:16 pm
Location: Auckland, NZ
Contact:

Post by suberimakuri »

1kpa steps may have it bouncying around a bit.. even at 2kpa steps need good dampening.

depends on your feed.

really though, after running map for ages.. when the load gets high, alpha n is king.
muythaibxr
Site Admin
Posts: 8230
Joined: Thu Oct 14, 2004 12:48 pm

Post by muythaibxr »

yeah, I wouldn't go in 1kpa steps... I'd something like this:

30,40,55,70,80,85,90,92,94,96,98,100

Tune that table completely, then tune an alpha-n table completely. Keep in mind that at low throttle positions it's probably impossible to get alpha-n to work right... You're going to have to tune by hand to get around 13.2:1 AFR when the engine is under load.

Once you get both tables tuned, just use my dual table hybrid alpha-n code.
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Guys,

Thanks for the heads up.

Ken
I am running those numbers from your VE table, working on getting the map based table tuned still seeing some big swings that I need to get worked out, thinking the viewer is making too big of corrections, on the large changes I go back and cut them in half and then logs some more and then adjust. Just takes time, lots of time.

Speaking of time, its time to go to the house, loading another table for the drive home, yesterday had to pull over and make some manual adjustments to get the lean swings out of the table. On this table I had time during lunch to take a good look at the new and compare it to the old and reduce some of the changes.

thanks again, wd
muythaibxr
Site Admin
Posts: 8230
Joined: Thu Oct 14, 2004 12:48 pm

Post by muythaibxr »

Some of the wild AFR swinging around could be accel enrichments...

I've never noticed a problem with the adjustments that megalogviewer is making.. so far they've done a good job... maybe you should get the newest version of megalogviewer, as it allows for some adjustment to the correction parameters.
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Ken,

I think I have the latest Viewer 2.40, and its set for auto updates. I just downloaded it about a week ago. So do you have to change it from the defaults settings? Did you have to change it?

But it could be the accel enrichments, I haven't tried to set them very much and I think they are set from your Decent_Tune.msq And I was looking at the logs tonignt. It appears that just about every time, it happens it is when I have made a throttle change. And I did notice driving home that I would not see the really big spikes if I kept the throttle movements slow with no sudden changes. I looked at the AE settings and I have it set for RPM, should I have it set for that? I mean, would it be better if it was set for MAP changes instead of RPM based?

I thought you were going on vacation? Or is yours like mine? Off work but to much to do at the house to go anywhere. HA!
later wd
muythaibxr
Site Admin
Posts: 8230
Joined: Thu Oct 14, 2004 12:48 pm

Post by muythaibxr »

I am on vacation, but decided to bring the new laptop that I got the day before I went on vacation. I can't go on vacation without my new toy!

I'm using different settings now for accel enrichment from the ones in the msq you have. In fact, almost all my settings are different.

I'm surprised your table was so rich if you were using my settings though... My full throttle part of the table is set for 13.2-13.0:1, and stays right there with a very small amount of change between runs.

I didn't change the defaults for the megalogviewer... but I thought your o2 sensor might be in a different place on your exhaust pipes.

RPM based accel just means you can change the accel parameters based on rpm... increase or decrease of rpm doesn't trigger the accel...

You probably just need to richen up the mixture in the areas where pushing the throttle causes spikes.

It is a good idea anyway when tuning the VE table to only use slow throttle movements.

I've gotten good results recently using MAPdot with RPM based and normal MAPdot. The best solution for the ms1 would be a combination of the two, but I'm not sure the chip has the resources necessary to add that code.
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Ken

I hear you on the new toy! =))

As far as the VE table goes, I haven't been up in the full throttle yet, maybe half throttle, breaking in a fresh rebuild, wanted to get a few miles on it first.
I am trying to hold the throttle as still as I can when logging, but of course when I hit hills, the map readings change and that is causing the lean and rich spikes. Its strange, when I am coasting down hill but still have a mid range MAP reading say about 70-80 then I am seeing the O2 go to about 16.5 Full off throttles sometimes give me a full lean O2 reading, 19+ ??? Is the full lean caused by the decel fuel cutoff? Should I turn that off until I get the VE table tuned?
When I go into the very high MAPs 96+ then I am seeing 11s or lower on the O2, yesterday I kick it one up to about 85mph about 48-49K revs and was seeing pegged full rich?
I am thinking that the differences between my exhaust, intake and yours is why your map is going so rich on my setup. Thoughts? I am using the stock air box and a full 2 1/4" exhaust with a stock type muffler.

It was somewhat better this morning on the drive in as far the spikes go, I was seeing them mostly when I went up steep hills, I would feel the car buck and then see the lean spike about 16.5 with a high 80s to low 90s Map. My cruising MAPs are coming down into the 80s now that is running in the low 14s O2 on cruise. It must just be the step changes on the bins that are causing these lean to rich surges.

Well thanks for all the help even while you are vacating, later wd

"I've gotten good results recently using MAPdot with RPM based and normal MAPdot. The best solution for the ms1 would be a combination of the two, but I'm not sure the chip has the resources necessary to add that code. "

PS Does this mean its a either or thing, meaning either MAPdot with RPM based OR normal MAPdot? If I understand it right, if RPM based then the normal is turn off ????
thanks wd
muythaibxr
Site Admin
Posts: 8230
Joined: Thu Oct 14, 2004 12:48 pm

Post by muythaibxr »

wrenchdad wrote:Ken

I hear you on the new toy! =))

As far as the VE table goes, I haven't been up in the full throttle yet, maybe half throttle, breaking in a fresh rebuild, wanted to get a few miles on it first.
I am trying to hold the throttle as still as I can when logging, but of course when I hit hills, the map readings change and that is causing the lean and rich spikes. Its strange, when I am coasting down hill but still have a mid range MAP reading say about 70-80 then I am seeing the O2 go to about 16.5 Full off throttles sometimes give me a full lean O2 reading, 19+ ??? Is the full lean caused by the decel fuel cutoff? Should I turn that off until I get the VE table tuned?
When I go into the very high MAPs 96+ then I am seeing 11s or lower on the O2, yesterday I kick it one up to about 85mph about 48-49K revs and was seeing pegged full rich?
I am thinking that the differences between my exhaust, intake and yours is why your map is going so rich on my setup. Thoughts? I am using the stock air box and a full 2 1/4" exhaust with a stock type muffler.
I'm running a TRD 4-1 header, magnaflow high-flow cat, and a 60mm catback (RSR exmag GT2). I'm also running open throttles with velocity stacks, so it wouldn't surprise me if my table makes your engine run rich.

Coasting down hills I see 30-60 if I'm close to off the throttle depending on the hill... if I'm maintaining speed, any higher and I accelerate. On level ground maintaining speed can fall anywhere between 70 kPa to 85 kPa. Maintaining speed up hill can get me to 90, and accelerating uphill in 5th can get me above 90.
It was somewhat better this morning on the drive in as far the spikes go, I was seeing them mostly when I went up steep hills, I would feel the car buck and then see the lean spike about 16.5 with a high 80s to low 90s Map. My cruising MAPs are coming down into the 80s now that is running in the low 14s O2 on cruise. It must just be the step changes on the bins that are causing these lean to rich surges.
I'm not sure what you're talking about with step changes.
Well thanks for all the help even while you are vacating, later wd

"I've gotten good results recently using MAPdot with RPM based and normal MAPdot. The best solution for the ms1 would be a combination of the two, but I'm not sure the chip has the resources necessary to add that code. "

PS Does this mean its a either or thing, meaning either MAPdot with RPM based OR normal MAPdot? If I understand it right, if RPM based then the normal is turn off ????
thanks wd
Yeah, it's either-or. To get rid of your lean spikes, you should probably increase the pulse width in the rpm range where you're having the problem.

Oh, here's my new spark table... it's gotten me close to double the fuel economy for cruise without changing the fuel to be leaner... it's also quite powerful on full-throttle, but you should be careful with it. If your dizzy isn't aligned exactly like what's left of mine, you could knock.

Image
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Ken,

Your MAP readings sound like close to what I am seeing. On the step changes, what I meant was that going from one area of the map to another mostly from a tune bin to an untune bin. Also when I make a big step change in the throttle position, I am seeing the lean to rich spikes. Like you said that is probably coming from my AE settings being off.

Thanks for the spark map, looks much different than mine with just a quick look, I will have to print out and do a comparsion.

I will just keep plugging away until I get it running like I want it to. Should I go ahead and try this spark table or wait until I get the VE table set?

thanks again, wd
muythaibxr
Site Admin
Posts: 8230
Joined: Thu Oct 14, 2004 12:48 pm

Post by muythaibxr »

wrenchdad wrote:Ken,

Your MAP readings sound like close to what I am seeing. On the step changes, what I meant was that going from one area of the map to another mostly from a tune bin to an untune bin. Also when I make a big step change in the throttle position, I am seeing the lean to rich spikes. Like you said that is probably coming from my AE settings being off.

Thanks for the spark map, looks much different than mine with just a quick look, I will have to print out and do a comparsion.

I will just keep plugging away until I get it running like I want it to. Should I go ahead and try this spark table or wait until I get the VE table set?

thanks again, wd
You can try it if you like, just be careful with the full throttle above 4000 settings.
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Ken,

I haven't done any full throttles with the table I have now! yet! Don't want to hurt the fresh rebuild but I will soon at the rate that I am clocking up miles, 110 miles a day min.

With that in mind, how do you tune the alpha-n table? Just do a bunch of full throttle blasts and then just use the viewer? And what goes in the column that normally has the Kpa settings, is it % or voltage or raw TPS? Can I use my Speed Density table as a starting point and just change the kpa over to alpha-n settings whatever they are?

I will compare the spark table with what I am using now and see how close they match before I try it.

thanks again, later wd
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Ken,

I have the VE table looking pretty good on my morning runs BUT in the afternoon its not very good, with lean to rich spikes. Its doing this with any changes to the table from morning to evening, the mornings I am seeing low 14s most of the time on the crusing side of things.
I think I need to adjust the air temp correction but I can not find anything to tell me how to adjust. I found where and all the temps have 100% beside them, is the 100% talking about fuel where over 100 would add fuel and less than 100 substract? I need some help, its a bear on the drive home!!

thanks wd
superchargingmachine
Experienced MS/Extra'er
Posts: 348
Joined: Tue May 04, 2004 3:46 pm
Location: Illinois, Central

Post by superchargingmachine »

I use clt related iat correction but I only use it to compensate for really heat soaked restarts.

You need to look at your IAT's and make sure the AFR's change is related to IAT.

If you decide to use clt related iat correction. You can correct based on clt or on iat. You make that selection on the second tab. The front page will always show clt temp in the bins. You may be able to change this in a config file but I never looked that deep. Greater than 100% adds fuel less than 100% takes fuel away.

Have a look at this thread for a lot of detail on iat correction.
http://www.msefi.com/viewtopic.php?t=19791
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Ken

Just a thought but could the switching back and forth from rich to lean and back again be because I am so close to correct tune on my VE table in cruise AND I have the EGO authority set a 10% ?

thoughts

thanks wd
superchargingmachine
Experienced MS/Extra'er
Posts: 348
Joined: Tue May 04, 2004 3:46 pm
Location: Illinois, Central

Post by superchargingmachine »

Yes, absolutely EGO at 10% is a lot. The car will feel notchy as it steps back and forth. Try 5% auth, maybe 8 events and 1% step.

I don't mean to jump in on your thread but I'm bored today.
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

superchargingmachine

Chime in anytime, so are you saying that after I get the VE table set, then the authority on the EGO will not be set at 100% ?? I was thinking that after the table is set it, you would give full (100%) control over to the EGO to make adjustments for different conditions using the VE table as a based to start from. Am I off on my thinking on how this works? I can understand the small step sizes but not the %?

Here is what I am seeing, in the mornings when its cooler, IAT at about 90 degrees F, I can cruise no problem, this morning seeing mostly low 14s on the O2, I ran a log thru the viewer after I got to work and watch as it made changes and after about a third of the way thru the log the VE numbers just stop changing, that tells me that I should be very very close to right on the VE table with the given inputs.
Then when I drive home with the IAT at 115 then I am feeling these surges and when I look at the O2 I see spikes of rich (13s) and then lean surges up to 16.5 +. These happen pretty quick and without any changes in MAP or TP, level ground and going up hills and going down hills I sometime see MAX lean 19.4+. Needless to say, this does not give me a warm fuzzy feeling about my tuning BUT I never see this in the mornings, the only big lean or rich spikes in the morning runs are when I go thru the gears to get up on cruise and right now I just about have the AE push up out of the way to get the VE table set.

This is what is pushing me to look at the IAT correction to see if that will stop this difference on running from morning to evening on the exact same VE table. I will be studying the logs to night to see if I can see anything that would be cause this. This is why I have not had much success with tuning using the afternoon logs, I would just get worse instead of better on the VE, just didn't "feel" the surges until I got the Table lean out some more.

So any ideas anyone?

thanks wd
superchargingmachine
Experienced MS/Extra'er
Posts: 348
Joined: Tue May 04, 2004 3:46 pm
Location: Illinois, Central

Post by superchargingmachine »

This is how I look at it. EGO correction is there to adjust for error or variations in conditions that the VE table can't account for. If your VE table yielded the exact O2 values in your target AFR table then EGO would do nothing. So what authority does is say that the max EGO can correct from a given VE. The step size is how much EGO corrects for after each event interval. If you allowed EGO 100% and you lost a plug wire or somthing EGO could pull 100% fuel and damage the engine.

So you might need to look at your target table to make sure it agrees with what you see on the O2 gauge. If they agreed the correction should be small and you shouldn't feel it. But, either way giving a smaller authority will keep you closer to the VE value and keep the swings from rich to lean a bit narrower.

I don't know why it feels better in the morning maybe the cooler air makes it just a tick richer and helps.

I wouldn't go to IAT correction yet it will just complicate things. If anything turn EGO correction off by setting the EGO RPM to like 8000 and try driving it home on a hot day. I'll bet good money the surging is gone. Just keep an eye on you O2's.
muythaibxr
Site Admin
Posts: 8230
Joined: Thu Oct 14, 2004 12:48 pm

Post by muythaibxr »

I would suspect that the air density curve in the air density .inc file might just be a bit off...

in the afternoon, with the IAT up that high, does the car idle at the same AFR as in the morning?

What octane gas are you running? Maybe if you're running 87 or 89 octane, the advance is right on the edge of too high for that temperature?

I have not noticed any of the nasty surging on my setup, and my IAT hovers between 120 and 140 on a hot day sitting in traffic.

However, I have noticed that on idle with temps that high, if I tuned idle at 90 deg F, the idle gets really lean at 120 deg F... so I used air density correction to tune it richer as the air temperature goes higher, and I made it "start turning off correction" at 3000 rpms, and made it turn off correction completely at 4000 rpms.

For the air density correction, you are supposed to set temperatures in the left column, and percentages in the right column, then open up the menu in that dialog box, and look for the settings, and set them reasonably according to the behavior you are observing.

Another option is to hand modify the airdenfactor.inc file (I think that's the name of it) and make the higher temperatures show denser air... that curve could just be off for the sensor you're using, or the temperature it's seeing.

I'm using an EGO authority of 15% and I'm not seeing the surging that you're seeing, so I don't think that's the problem. If your exhaust is a different shape from mine, or different diameter, or different length... you might need to adjust the step time to be a little slower or a little faster as well...
wrenchdad
Master MS/Extra'er
Posts: 614
Joined: Mon Feb 14, 2005 5:30 am
Location: Natchez, Mississippi

Post by wrenchdad »

Guys

First thanks for the replies. Here is what I am seeing in the logs, I had to look at it in Excel to be able to see what was going on. First the TP, MAP and IAT are all steady state when I see this. I am seeing in the Afternoon logs sometimes a 1v step change in the O2 reading, this happens from one recorded line to the next, like in .1 sec, it happens mostly when the GEGO is turning around changing direction on its correction. In the mornings, I am seeing close to the same thing, just the step size is smaller about .6vdc. I am seeing the GEGO rolling, for lack of a better word, from 95 up to 105 in the evenings over a 5 secs time frame with a Gair of 94. The morning log rolls from 95 to 105 in 4 secs with a Gair of 98.

SO it looks like its a combination of both, the O2 settings are not quite right and the lack of temp correction in the afternoon makes the swings last longer. I am running 92 octane and your advance table. Which BTW is great!

I will have to take a look at the MAP readings morning and afternoon and see if they are different, I know the idle speed is different so they are probably different too. The idle is higher in the afternoons.

Here are my GEO settings right now
EGO Switch point 2.509v
Enents 120
step size % 1
Authority 10%
Active > 150 degrees
Active > RPM 1300
Step Counter Msec

Maybe I need to switch to ignition events from the Msec to speed things up some? It seems on the logs that its trying to correct and making changes but its too little to late so of thing and then it been doing it long enough and kind of winds up the proportional gain and bang makes a large jump which is too much putting the control into swinging. Does that make any sense?

Let me know if ya'll need anything else and what you think.

thanks again, wd
superchargingmachine
Experienced MS/Extra'er
Posts: 348
Joined: Tue May 04, 2004 3:46 pm
Location: Illinois, Central

Post by superchargingmachine »

I would cut your events way down maybe 16-32 to start. It may be that you swing low and hang there the way you have it, it then wait too long and over correct. As I suggested earlier you may also need to adjust you targetAFR table a bit.

This quetion is for both you guys,
With ITB's do you have to cruise closer to stocih(14.7) because I know a few people mostly with turbo's cruising with 4A's that are leaner without any trouble. I cruise right around 15.8-16.2 with no issues. My thought is that you could possibly get better economy. I cuirse in the 3200-4000 range and get 35MPG highway.
Post Reply