Help to understand EAE

Tuning concepts, methods, tips etc.

Moderators: jsmcortina, muythaibxr

Post Reply
martin2day
Experienced MS/Extra'er
Posts: 169
Joined: Tue Nov 01, 2016 11:56 am

Help to understand EAE

Post by martin2day »

Hello,

i want to tune my car.
It is a small block chevy 350 with a TBI and a AFR composite manifold.
Because of the very large manifold it is a little bit hard to fine tune the VE table.
For any transition i want to switch completely to EAE. Today a make a drive and set clt and rpm correction to 100%. The normal AE i set to 1000%/s.
I drive the car and switch like the manual say to 3000rpm and hold the TPS%.

Now it is not real clear for me how i have to change the AEA settings.
It is clear if accel and the a lean spot i have to increase ATW. But witch value. If i go from 40-60kpa and see a lean spot i think i have to increase 60kpa.... but what is with the value above 60kpa...? must i also increase thees?

If i decel and see a lean spot, i have to decrease SFW... but what value. If i go from 60-40kpa i think i have to decrease 40kpa...? But what about the value under 40kpa...? :-)

Can someone have a look at my MSQ and the LOG and can tell me how und why i have to change what value?

https://www.dropbox.com/s/hpsp6fasgvjoy ... 4.zip?dl=0

Thank u very much.
Regards Martin
martin2day
Experienced MS/Extra'er
Posts: 169
Joined: Tue Nov 01, 2016 11:56 am

Re: Help to understand EAE

Post by martin2day »

Hello,

nobody can give me a hint?
At low TPS change i think the setup is okay... but if i hit the TPS harder and go to MAP over 70kPa it isn't good.
Today i change the settings very hard to see what happen... it is a litte bit better, but not as good as i think.... to lean.

Please have a look at my LOG and give me a hint :-)
I have change the EAE as seen in the picture....
Is it possible that i have t change the values above 70kPa higher and higher... i have a TBI so i thing at high kPa it at very much fuel to the wall...!

Thank u very much.
Martin
Bildschirmfoto 2017-06-19 um 20.35.08.png
2017-06-18_12_modified.msl
mrx79
Experienced MS/Extra'er
Posts: 299
Joined: Wed Oct 29, 2008 8:26 am
Location: Frankfurt / Germany

Re: Help to understand EAE

Post by mrx79 »

Hi Martin,

haven't you saied in one of our emails that your o2 sensor reaction time is quite long due to the position in the exhaust?
The first lean spikes occure about 0,4-0,6s after the initial TPS movement.
considering your long reaction time, this could be a missing initial fuel squirt.

A reasonable amount of EAE is triggered only 0.15s after the throttle change.
On my engine it's similar, just be seen by the o2 sensor earlier.

Try some TPS based AE in addition to the EAE. This is also a hint in the manual to cover the first demand prior the MAP movement has been recongized.

PS: did my hint with the negative load change helped?

Best regards
Honda CRX B16A1 Turbo | MS3 running | pre1.5.1 beta7 Firmware | 24/1 Dual Wheel | COP ignition | 725cc ID injectors
turbo conversion
Super MS/Extra'er
Posts: 1281
Joined: Thu Aug 19, 2010 6:22 pm
Location: White House, TN USA

Re: Help to understand EAE

Post by turbo conversion »

If you don't have your VE table pretty much spot on EAE will never function correctly.

David
1976 Datsun 280Z L28ET Garrett GT35R T3-T04E stage3 50 trim 63 A/R housing custom grind cam 2000-6000 rpm 440cc injectors intercooled 18 lbs. boost
3" exhaust turbo back LC-1 o2 sensor Hallman manual boost controller EDIS 6 ignition batch fire 60mm throttle body 5 spd T5 borg warner 3.54 lsd
svocapri
Helpful MS/Extra'er
Posts: 51
Joined: Sat Nov 29, 2008 7:43 pm
Location: Rome, Georgia

Re: Help to understand EAE

Post by svocapri »

I am working on a similar endeavor. Mine is a 302 ford with a Holley TBI and no exhaust gas heat crossover under the intake plenum. I don't believe anyone has successfully tuned a wet flow manifold with Megasquirt and EAE. I am trying but I am so far unsuccessful. One thing I have noticed is that autotune is mostly useless on my combination, and with your composite manifold I suspect yours may as well. The issue is that fuel puddles in the intake at higher MAP and lower RPM's. When you press the accelerator, the fuel will fall out of suspension and puddle in the manifold causing a lean spike. From that initial lean spike until the fuel puddle builds up and the AFR stabilizes can be upwards of 10 seconds. During that 10 seconds, autotune kicks in and increases the VE at that point in the VE table beyoud where it should be. Because my car is relatively light, by the time things are starting to stabilize, the engine is moving into the next higher RPM column so autotune doesn't have the chance to go back and correct itself. To combat this, I have enlisted my daughter to tune wile I drive. She calls is playing mathematical Whac-A-Mole. I drive as steady as possible and she watches until the the AFR stabilizes and then adjusts the cell rich or lean as needed. It is an extremely slow and tedious process but fortunately she sees it as a game. This has made a HUGE improvement in the drivability of my car, and it has made the lean rich swings much less severe with the default EAE tables. For a brief while, I inverted the sucked from walls so that at high KPA it sucked less than at low KPA. This seemed to work much better but that was before I realized that my autotuned VE table was all out of whack. I therefore put the EAE back to stock and I am re-tuning my VE tables, but I may invert the Sucked From Walls again once I get my VE table in check.

Julian Weber
turbo conversion
Super MS/Extra'er
Posts: 1281
Joined: Thu Aug 19, 2010 6:22 pm
Location: White House, TN USA

Re: Help to understand EAE

Post by turbo conversion »

Auto tune is designed for steady state driving and very slow accelerator movement.

If you accelerate quickly auto tune will see this as a lean condition for those cells and ad fuel.

If you are trying to tune EAE before your VE table is almost spot on you are chasing your tail.

Try tuning AE first (after your VE table is tuned) and once it is close then work on EAE to fine tune AE.

David
1976 Datsun 280Z L28ET Garrett GT35R T3-T04E stage3 50 trim 63 A/R housing custom grind cam 2000-6000 rpm 440cc injectors intercooled 18 lbs. boost
3" exhaust turbo back LC-1 o2 sensor Hallman manual boost controller EDIS 6 ignition batch fire 60mm throttle body 5 spd T5 borg warner 3.54 lsd
martin2day
Experienced MS/Extra'er
Posts: 169
Joined: Tue Nov 01, 2016 11:56 am

Re: Help to understand EAE

Post by martin2day »

Hello,

i understand the theory very well.
But i thing at some parts i don't understand whats going on in my engine.

Please have a look at 001.msl!
You see a quick run with about 27%TPS.... first it is to lean and than it goes rich... VE is 68... so i think have to increase AE and degrease the VE.
I have so tune the hole VE table... hold the TPS from 1000rpm to max rpm the same amount....

BUT than have a look at the 002.msl
You than see the result... with AEA and normal AE at quick trottle move it goes rich and than lean... if i hit it harder it goes to lean...
So i think the VE is not right....

Is it possible that at low rpm and high TPS my camshaft with overlap see the wrong AFR?
Or is the Ignition wrong?

Regards Martin
001.msl
002.msl
turbo conversion
Super MS/Extra'er
Posts: 1281
Joined: Thu Aug 19, 2010 6:22 pm
Location: White House, TN USA

Re: Help to understand EAE

Post by turbo conversion »

Start a new thread and post an msq and a data log.

David
1976 Datsun 280Z L28ET Garrett GT35R T3-T04E stage3 50 trim 63 A/R housing custom grind cam 2000-6000 rpm 440cc injectors intercooled 18 lbs. boost
3" exhaust turbo back LC-1 o2 sensor Hallman manual boost controller EDIS 6 ignition batch fire 60mm throttle body 5 spd T5 borg warner 3.54 lsd
svocapri
Helpful MS/Extra'er
Posts: 51
Joined: Sat Nov 29, 2008 7:43 pm
Location: Rome, Georgia

Re: Help to understand EAE

Post by svocapri »

Martin, your logs look very similar to mine. Here is how I see what is going on inside my manifold...
Imagine a small bowl sitting at the bottom of your manifold plenum. When your TPS is at 10%, the air is moving rapidly past the throttle blades and the fuel is vaporizing nicely. The fuel and air mixture is moving together like a cloud down the intake runner and into the cylinder. When you open the throttle the pressure in the manifold rapidly increases and that cloud of fuel and air suddenly starts raining gas all over the bottom of your manifold. in addition, the air that was moving rapidly past the throttle blades and catching the fuel droplets with it is now moving much slower and the fuel droplets are simply falling past the open throttle blades and into that imaginary bowl in the bottom of your plenum. The fuel puddling on the floor of the manifold and in the plenum area is where that initial lean spike comes from. Now if you continue to hold the throttle open, that puddle in the bowl in your plenum will get larger until enough fuel is evaporating out of it to make up for the lean condition and you will see your AFR stabilize. Finally, when you let off the throttle with that large amount of fuel resting in the bottom of the plenum, it will begin to vaporize rapidly and your AFR will drop until the puddle is fully evaporated at which point the AFR will climb and finally stabilize.

EAE is supposed to know how much fuel is being added to that puddle and how much is evaporating from it at any given load and RPM. If you tell EAE the wrong information and it thinks the puddle is bigger than it is, then EAE will pull fuel even after the puddle has evaporated and you will see the engine go lean. Conversely, if you tell EAE that more fuel is being removed from the puddle than it actually happening, then it will not inject enough fuel because it thinks it is coming from the puddle, but also when it thinks it has exhausted the puddle it will start adding fuel, but since the puddle is still there in the manifold, your AFR will go rich until the puddle actually evaporates at which point the AFR will go lean again and then stabilize.

Keep in mind, all of this is based on MAP not TPS so if you open your TPS to 27% but let the RPM's climb, the MAP will initially go up almost to atmospheric at low RPM's but then decrease as the RPM's increase. At higher RPM's the Sucked From Walls and Adhered to Walls is different than at low RPM's and thus the need to adjust EAE for ONE RPM ONLY and then change the Sucked From Wall RPM correction and Adhered to Walls RPM correction for the other RPM's

Based on this, I have found that I can't tune EAE by using any sort of fast throttle movement. Also, the reason that the VE table must be correctly tuned is that EAE is assuming the puddle is a percentage of the VE. If the VE is wrong, so is EAE.

What I have been doing which is making things better is to start at about 2,000RPM in overdrive and then increase the TPS so that the MAP goes up about three cells in the VE table. I hold it there until the AFR stabilizes and then release the throttle until the MAP goes back to where I started. I then hold it there until the AFR stablizes again. As a general rule, if the AFR goes lean when you open the throttle, then increase the Adhered to Walls at the three MAP points where you tested. If it goes rich when you close the throttle, then increase the Sucked From Walls at those same MAP points.

Good luck and let me know if you find some other helpful methods as I am still struggling with this.

Julian Weber
martin2day
Experienced MS/Extra'er
Posts: 169
Joined: Tue Nov 01, 2016 11:56 am

Re: Help to understand EAE

Post by martin2day »

Hello,

thank u for your long and informative answer. :-)
As i say before i think i understand how it works but i don't understand my engine. :-)
I have the same problem as you say. Light car (2000lbs) and a lot of power.... the rpm goes high very quick.

So i think the best bet is to hold the throttle and the brake and simulate some parts of the VE Table.
This i have make and you see this in LOG_01.
log_01.msl
I have a look at wallfuel,tps and Map. If these 3 values are near constant i think the AFR shows the real value and i can edit the right VE value...

But than i am looking at some other LOG's and see that these area are to lean.
So the LOG_02... here i have some normal AE to, but i see it is to lean after the rich spot.... and this i don't understand...!?
log_02.msl
I see here that the map value goes down very quick and so the AEA value are 95%... make it lean because some fuel get sucked from the puddle.

But how can i react. At TPS press i see i need some more fuel... at AEA.... but if i accelerate and the MAP goes down i see AEA have to be smaller....
Can some one help? :-)

I think for the very large manifold and et very low rpm and hight MAP i must use much more fuel... or other words the puddle is very large....

Martin
svocapri
Helpful MS/Extra'er
Posts: 51
Joined: Sat Nov 29, 2008 7:43 pm
Location: Rome, Georgia

Re: Help to understand EAE

Post by svocapri »

So i think the best bet is to hold the throttle and the brake and simulate some parts of the VE Table.
That is what I have been doing to get the higher MAP bins tuned. If I had a load dyno this would be much easier.

So the LOG_02... here i have some normal AE to, but i see it is to lean after the rich spot.... and this i don't understand...!?
In that LOG it looks like you went from 43 to 73 KPA. That in my experience is too much to crate a good EAE curve. I would try going from 45 to 60 KPA and getting that area dialed in. Then try 55 to 70 KPA and on up to 100. Once you have the Adhered to Walls and Sucked From Walls curve correct from 40 to 100 KPA at 2,000 RPM and your ATW and SFW RPM Correction set to 100%, then you will need to do it again at 3000 but this time change the RPM correction not the ATW/SFW coeficient. You will also need to adjust all the RPM corrections from 1,000 to 6,000. Once the curve is correct for slow throttle movements and the RPM corrections are correct, then you will be able to adjust faster throttle movements with the time based accel enrichment. I have also found that on a wet flow manifold, you will need more like 6ms accel time instead of 3 to take care of the lean spike after the initial rich surge.

I have attached my dash layout that I use for tuning Accel Enrichment and hopefully it will help. The AFR gauge is sitting on top of the afrtgt1 gauge so I can easily see when the AFR doesn't match the AFR Target.
Screen Shot 2017-06-28 at 1.08.52 PM.png
Post Reply