some general tuning questions

[Yves]

I think I have established what looks a basic VE table with autotune, although this far from right in my opinion.

During the whole process where I used autotune, I had AE on, since I was having some rather strong backfires through the intake. I currently also still have some popping in the exhaust during decel.
I read on here that normally both autotune and VEAL in megalogviewer take the AE and decel into account. Can I therefor assume that at least the values in the cells I have hit most are correct ?
Does autotune and veal actually use the AFR tables to get the required VE value ?

I was also thinking on going to incorporate AFR. What would be required to go this route ? Recalculate the values in the VE tables taking into account the value in the AFR table vs the stoich AFR ?

And finally, to go and fine tune everything, do I put the TPSdot high enough to prevent the AE from kicking in ? This will probably make the engine backfire again.

Thanks

[kaeman]

Hey Yves, I have been using incorporate afr in my fueling since I first built the vehicle, I don't know the what the difference between incorporate afr and non is, but I believe that if you don't use incorporate afr, then tuner studio and megalog viewer try to make the ve table hit the stoichiometric ratio rather than your afr targets.
I have turned off my ae and driven my car... the engine responds pretty well, it makes it easier to tune some stuff without the ae kicking in. I believe that once you are close on your ve tables, your engine will not need much ae, because the table will compensate for your changes in throttle/map.

[Yves]

I'm going the incorporate AFR way as well. I just haven't grasped the interaction between the different tables and how the engine behaves. Basically, it idles around 69 kPa and it seems to cruise at constant speed in between 65 and 75-80 kPa. Acceleration is above 75 kPa and decel is at or below 60 kPa. I do not get the typical diagonal where in the table mostly operate. In my case it seems to have a horizontal plane in which it operates.

So I probably need to set AFR's accordingly, but not sure which AFR's belong in which zones. I assume it will have to look something like aroun 13.8-14 at the horizontal line (engine doesn't seem to like AFR's above 14), 12,5 in the zones above 75 kPa. and 14 below 65 kPa. But like I said I'm unsure.

[kaeman]

Yves, here is my current tune... look at my afr targets, my engine runs in the funny flat lines also.... it doesn't go diagonally unless I am using small acceleration. most of the time it goes up then over then when I reach speed in drops back down in a strait line.

2014-09-26_17.58.17.msq

[Yves]

Yves, here is my current tune... look at my afr targets, my engine runs in the funny flat lines also.... it doesn't go diagonally unless I am using small acceleration. most of the time it goes up then over then when I reach speed in drops back down in a strait line.

2014-09-26_17.58.17.msq

I'm going to input it in my system this evening and check it out.

What are the cam specs ? Do you really use 11.7/1 CR ? Seems awfully high if that's a 274° cam.

I'm thinking on how to tackle the conversion to incorporate AFR as well as where to set the ignition timing. It seems to have a big influence on lambda values I'm seeing.

Did you ever try the ITB load option ?

[Yves]

I opened it with megalogviewer.

Is it correct that your VE table only goes to 86 ?

Also I see your ignition advance at high kPa is below 20 ? Is that normal ?

Is there a way that you can open the msq in tunerstudio without it affecting my own tune and without opening a new project ?

Another question : would converting the values that are currently in my VE tables be as simple as making the division between the AFRtarget and stoich X the VE value ?

[strain]

Another question : would converting the values that are currently in my VE tables be as simple as making the division between the AFRtarget and stoich X the VE value ?

Yes, open it with the option "Use temporary Project"

[pit_celica]

There are some false information here.

Incorporate AFR target only changes the way the fuel PW is calculated. Look at the formulaes here :

http://www.msextra.com/doc/ms3/fuelcalc.html

The main advantage of using incorporate AFR target is that, once tuned, if you want to change the desired AFR target for a specifed cell, you only need to change value in the cell in the AFR target table. No need to play with the value in the cell in the VE table. The other advantage is that the VE table looks more like a "VE" table. The values at peak torque will be around 100-110.

If you do no play with your AFR target a lot or if your engine is already well tuned, I don't see the point of using incorporate AFR target. It will be a waste of time without any huge gains.


As for the AE event being ignored during VEAL or MegalogViewer VE analysis, this is true. But you need to understand that at low RPM/ low load, there is a lag between the actual combustion event and the reading of the AFR. So, even if TS ignored the AE event when they happen, the AFR still can be richer than it should because of this lag. This is why you should tune your Lambda Delay table before doing some VEAL. The other workaround is to do some really gentle acceleration movement on the throttle pedal when driving and VEAL. VEAL shoud be changing the VE table only during steady state condition. So, stay inside 1 cell for a long time until you are near your target and then move to another cell.

Sam

[Yves]

You mean the lambda delay table in autotune ? I haven't touched that since I didn't understand it. Would you care to explain ?

I have an older LM-1 as O2. Not sure if a good delay table exists for that one.

[pit_celica]

The delay isn't necessarily based on the type of wideband you have. It's more likely the placement of the sensor in your exhaust piping. The closer to the engine it will be, less delay there will be.

There are no easy way to accurately fill this delay table. One way to do it is to make sure that AE is enabled and that you can easily create an AFR rich peak with a quick throttle stab while still staying at about the same RPM. Then, in the datalog, verify the time delay between the injector PW pulse (corresponding to the AE event) and the AFR rich peak. Do this at different RPM corresponding to the columns in your delay table. It can be hard to do it at different load value, so, as a first guess, you can set the same value for all the cells of a same column.

If I'm not mistaken, the units of the values in the table are millisecond (ms).

You should have higher delay at low RPM/low load and low delay at high RPM/high load.

When your delay table will be set correctly, the VEAL will do a better job during transient events.

Sam

[Yves]

Looks like switching off the ae is an easier way to go.

[Yves]

I did the following today, I used 'incorporate AFRtarget' and I then set the lambdasensor lag which was previously 60 to 95. I then went for a drive. It immediatly made changes and the car really felt a lot better than before when doing a couple of runs. Popping in the exhaust is also less.

[pit_celica]

Hi,

You are mixing two types of things.

First, you have the Lambda Averaging Lag Factor located under "Basic Setup"/"General, lags". This value (and all other lags value) apply a kind of dampening on the raw sensor signal to remove some jitter/noise. If your sensors signals are clean without any noise, it is better to have a value near 95-100%, essentially no filter at all.

Second, the lambda delay I was talking about is located under "Tune Analyse Live" tab, then under "Advanced Settings" tab, then press the "Lambda Delay" button. A 3x3 RPM vs Load vs LambdaDelay(ms) will appear on your screen. This is the table that needs to be correctly tuned before doing VEAL with throttle AE/DE.

In the log you sent me by PM, the only clear point where I can calculate the lambda delay is around 1314-1315 in the log. Based on this log, I would say that you have a delay of around 250ms at 2500 RPM and 80kPa. Another clear point is at around 1441 in the log, I can calculate 130ms of delay at 3500 RPM and 95kPa. For the low RPM/low load area of the table, it's hard to see from the log, because there are no clear relationship between an injection PW peak and the resulting AFR rich peak.

I think you now grasp the idea about Lambda delay table.

Sam

[Yves]

Thanks, appreciate that Sam.

Do you change the rpm cells and kPa cells in that table to input the correct values ?

Rpm on mine is set at 800-2500 and 7800.
kPa at 30-45.5 and 100

Seems that they are a lot higher at the low rpms.
The engine idles at 69 kPa so, I will not have much use for the 30 and 45 kPa it seems

I've tried to make some own calculations on peaks in AFR and it looks like I'm looking at 67 ms at 5500 rpm, around 600 at 2500 and almost 1000 ms at idle. Far cry from what is actually in the table.

[pit_celica]

You can edit the RPM/load values as you want. I agree with you, if you idle at 69kPa, no much point to have a 35kPa row. Tune it as you think it's right.

Sam

[Rick Finsta]

FWIW I idle in the high 50s or low 60s and I have logged my engine hitting less than 25kPa on throttle closing to take that FWIW. Not a big deal to not have a lot of resolution down there, but you might want to keep some bins for interpolation.

[Yves]

FWIW I idle in the high 50s or low 60s and I have logged my engine hitting less than 25kPa on throttle closing to take that FWIW. Not a big deal to not have a lot of resolution down there, but you might want to keep some bins for interpolation.

Mine does that too when in wot or acceleration and then taking your foot of it. Bottom rows are 30-40-50-60-65 and then more resolution.

If I keep a row of 35 in there it would only use that lambda delay during decel. Not sure if that is really necessary in the decel area.

[Yves]

Would it be usefull to use the scatterplots to calculate or see something ?

[kaeman]

Hey Yves, the reason you don't see any values in ve table 1 higher than 85 is because I went to table switching for this tune (my switch point is 85 kpa) to look more closely at the hard accel area of engine load. the upper part of the tune is in ve table 3. it is the values from 85 to 105 kpa. As far as the timing tables go I am not using timing or ignition control yet. I have a mechanical advance distributor set at 10degrees advanced and it adds another 23 degrees mechanical advance. for a total of about 33 degrees at 3000 rpm. I am still pulling fuel out of my lower rpm high kpa values to get rid of a rich misfire that occurs when I stab the throttle hard under 2000 rpm. I have my AE set to tpsdot of 200 giving a 2% shot, but I am getting ready to reduce that even more as the miss-fire only occurs when I get an AE event.
Good luck...
I am seriously considering trying the ITB mode. after looking at scatter plots.....

[Yves]

Not sure what you saw in the scatterplots.

I'm afraid I do not quite understand the idea behind the switchpoint graphs.

keep me updated on how it goes plz.