New tuner, need help - part 1

[JTB321]

Originally posted in the Tuning forum but not much response, can anyone offer some feedback please?



First I apologize for the long post but I'm getting bogged down with my tuning efforts and need to get back to basics which, for part 1 is some education for me.

What I have - SBF 302, E303 cam, GT40P heads, 19LB inj installed in a FFR Cobra. MS3/MS3X running TFI full sequential with two LC-1 WB sensors and fuel pump pressure input also being logged.
Using a modified version of the DIYautotune 'Muxstang.msq' for a very similar set up the car started first time and idles well. But, after a few unproductive TS datalogs/VE analysis tunes I'm not making much progress so I need to get back to basics which means fully understanding what it is I'm trying to achieve!!

So, before getting in to the tuning efforts and results can some one please explain the 'chicken and egg' dilemma I find myself struggling with.

To tune an engine we need an AFR table, IGN table and a VE table to effectively set the injector PW's to meet the required AFR. (hopefully this is correct).
So, when you start tuning it is assumed the AFR table is somewhere near, but what if it is not? The VE table (which is also another best guess) will try and tune to an AFR that won't be correct? All along, the IGN table is left at a default and not adjusted at this point.

I know TS has the option of generating AFR and VE tables but the information required is not known i.e. HP and torque figures so how can an accurate AFR and VE table be produced?

Rather than just blindly make changes I am trying to understand the theory and I'm obviously not there yet!!

Hopefully some of the experienced tuners out the can get me on the right road, the weather is great and my car is in the garage!!! HELP

.MSQ attached which was the result of a 30 minute data log and VE analysis update. This tune was an improvement in drive ability from the base tune but the next data log and VE analysis went backwards with respect to smooth running, in fact it was bad......

Cheers, John

[slow_hemi6]

You didn't respond to my suggestions or my question in the last thread.
Are you using VEAL, tuning manually by examining logs and making changes or live tuning?

[JTB321]

Apologies for that, I have just got back to the forum to check.
I am using TS to gather data logs, using Megalog Viewer to analyze the log and accepting the updated VE table. As I said I had some success on the first log/update but a second went backwards and the tune made the car worse to drive.
My main request is in trying to understand the generation of the VE, AFR and IGN tables. If you use tables generated by guess work (don't know HP, torque etc) how accurate is your starting point and how can you get to an acceptable tune?
As you suggested, if I include AFR table does that not just compound the problem of an uncertain starting point?

Hope this clarifies my issue, appreciate your feedback

John

[gboezio]

AFR are for the EGO control, it's just telling the ecu to tweak the PW to reach the table AFR if it has enough authority to do so.
Don't worry about AFR for now first thing is to tune advance and VE, disable the EGO control and use the Ve table to achieve the desired AFR, then by activating the EGO control it will need very little correction to meet target AFR.

Ignition : load each bin and find a way to measure power, of course dyno is best, arse dyno is second, but you can compare pulls, increase timing until you gain no more and use as little as possible for best power, too much and boom expensive parts will fly.
It's a long process to learn, good luck !

[JTB321]

Thanks, I'll try that approach first.

John

[slow_hemi6]

The Initial TS generated tables are not real good even when your figures are pretty close, in my experience with them.
If you have TS then I would suggest using VEAL to adjust as you go. One thing with adjusting tables through VEAL or VE Analyze in MLV as you are, is they will mess up your idle. With VEAL I get the engine up to temp and using my wide band readout and the 2D VE map I set the block of cells around my idle. Both IN and OUT of gear for my Automatic gearbox, manual just in neutral is enough. Then Go into VEAL and lock out those cells so they cant be changed by the program. MLV VE Analyzer allows the locking of cells as well through right clicking on them. Other than that VEAL is pretty good if you give it enough varied conditions on the heat map and enough time in each to get enough data.
Another side of the coin is Accell Enrich. You can set a ridiculous threshold and turn it off for initial tuning. You are likely to experience some transitional issues on more aggressive throttle openings. This will convince VEAL to make the VE cells richer to try and compensate for no AE. This then cause richer running and more correction from the EGO trying to reign that in when active. So when you get into tuning your AE you will likely find you can lean out your VE numbers a bit in certain areas to result in less ego correction happening. This however is more of a manual tuning exercise than trying to let VEAL or VE analyze do it.

[JTB321]

Sounds good, now I'm getting to understand and have an approach I can try out this weekend.
I'll post results.
John

[slow_hemi6]

Added some extra to my post above^

[JTB321]

Thanks, great advice. Looks like I am going to be busy this weekend.
Cheers, john

[darthmayndawg]

I'm going through old posts trying to find a similar builds to mine.

I wanted to get a copy of your Tune file as I have the exact same setup.

if you can help then let me know. Share file please!