MAF Scaling questions (KSTech 73mm)

So I just bought a 73mm KStech cold air intake to give my turbo some headroom as I'm hitting around 4.9v on the stock intake.

 

What is the best approach to getting the scaling sorted out? (I've already read the various guides but want to see if my stupid idea below makes sense)

 

I was considering taking the difference in area of the stock maf and the kstech and multiplying that with the maf g/s values in the rom to get something to at least work with as a rough scale.

 

Then scale the low end of closed loop (idle, and "around town" maf g/s) using the romraider tool so the thing idles and can at least move around in a not so terrible way.

 

Then the rest of it in open loop by disabling closed loop (to avoid AF correction and learning from screwing with the fueling), reducing timing and richening up the top end (for safety), and calculating the "new" maf g/s by taking the logged (wrong) maf g/s and multiplying by [Wideband O2/Target AFR] and simply replacing the old g/s with the "new" value in the table.

Does this seem like a valid method?

 

Why does everyone else tune OL scaling using numerous logs and complicated spreadsheets? I'm going on the (probably wrong) logic that as long as the throttle changes are not abrupt, method above should work fine with just a few short logging sessions.

 

Am I missing something?

 

Or can someone be nice and point me towards a kstech 73 scaling I can use as a start to save me some time?

 

Then the rest of it in open loop by disabling closed loop (to avoid AF correction and learning from screwing with the fueling), reducing timing and richening up the top end (for safety), and calculating the "new" maf g/s by taking the logged (wrong) maf g/s and multiplying by [Wideband O2/Target AFR] and simply replacing the old g/s with the "new" value in the table.

Does this seem like a valid method?

 

I think the one missing piece the interpolation between the maf voltages table on the map. So yeah, you can log and make adjustments to a specific Maf V you hit, but you also need to "bridge" the change from say 3.5V to 3.65V on the table. The spreadsheets essentually do what you describe above but they also bridge the difference between Maf V so that the scale is nice and smooth. Obviously, this can also be done manually, but the OL maf scaler excell sheet available on the romraider forum is pretty easy to use. You should be able to scale your highend very closely on 4-5 logging runs.

Forgot I had tuned my friends 04 STI with KStech 73mm, scaling attached.

 

He had 850cc DWs, walboro pump, FMIC, and FP HTA Green.

 

Hope this helps.

 

KStech 73 mm on 04 STI.csv

Thanks for your help!

I'll get started on it this week and will post up my scaling when I'm done.

I found LittleBlueGT's scaling on the romraider forums (http://www.romraider.com/forum/viewtopic.php?p=61789), and since the intake design is different on the LGT vs GD, I went with his scaling as a base. Thanks though, Spec B.

 

Anyway, I drove around and scaled the closed loop portion. and LittleBlueGT's scaling was only very slightly off, requiring minor changes.

 

I'll be doing open loop over the coming week or so.

Should I reduce timing up top to be safe? Initially make the OL maf scaling "more" so it's "flowing" more to richen things up? What do most people do?

Yeah, just to be safe I turn down boost.

 

Glad it's going smoothly.

A quick update on my maf scaling adventure... California 91 btw.

 

I reduced timing and ran wgdc 0 which on my AVO 380 is about 1bar (15psi actuator).

The CL portion scaled wonderfully and was fairly close to LittleBlueGT's scaling, but the OL portion is almost 9% off at higher loads.

 

I was targeting mid 10s but hitting lowish 11 afrs.

It started raining so I gave up after one pull.

 

I guess I'll scale what I can to fix the discrepancy with the one pull I have and then make the higher end of the maf scale where I don't have data so it'll run rich - maybe apply a 10% correction to be safe...

Sounds good - I would add 12% accross the board (to be more safe) on the high end and rescale again.

Still working on scaling and now my errors are all within 2% as of the last run.

Of course this is still on WG boost so I'm only hitting 4.4 maf V. Anything higher is just an estimated curve.

I'll start turning up boost, timing and leaning it out to hit the higher flows.

 

The only problem I see right now is that the curve isn't exactly smooth yet, but I guess I can sort that out when I get the top end dialed in and just do another logging run to smooth the entire curve.

 

No idea what's with the CL/OL transition area on the maf scale.

 

The CL portion was scaled in romraider's maf tab, so maybe my wideband reads differently from the front o2?

Any updates on the MAF curve? Am getting the same intake on a blouch 1.5 and would like to see your results. and compare scale #. Thanks.

Haven't driven the car much since and haven't worked on the scaling either for reasons unrelated to the car, but I'll post an update when I get a chance.

Any updates on the MAF curve? Am getting the same intake on a blouch 1.5 and would like to see your results. and compare scale #. Thanks.

 

I have the same setup. Nearly finished scaling my MAF.

Still working on scaling and now my errors are all within 2% as of the last run.

Of course this is still on WG boost so I'm only hitting 4.4 maf V. Anything higher is just an estimated curve.

I'll start turning up boost, timing and leaning it out to hit the higher flows.

 

The only problem I see right now is that the curve isn't exactly smooth yet, but I guess I can sort that out when I get the top end dialed in and just do another logging run to smooth the entire curve.

 

No idea what's with the CL/OL transition area on the maf scale.

 

The CL portion was scaled in romraider's maf tab, so maybe my wideband reads differently from the front o2?

 

I'm going to say this once, then go away, but your intuition about the "curve" is deeper seated than you think. I once played chase-my-tail with this, but there is no end to that story. It doesn't make mathematical sense. Air flow is air flow and it doesn't make "curves within curves" in an engine.

 

What I decided to do was, to create a perfect mathematical curve... then find the means to enable my engine to relate to it. That is because there are influences that affect the MAF CAL that most do not take into consideration. One important influence is the AVCS mapping. Change the AVCS and the MAF CAL changes. Change timing and... and...

 

My thinking is, the fundamental parameter that governs all else is the MAF CAL. Therefore, start with a mathematically "perfect" standard. By following that idea, despite the tuning hardships that it sometimes brings, I am rewarded with results that vindicate the process in the long run.

 

YMMV

Thanks for the info seeeeeya, I'll probably start working on my scale again soon.

It's running pretty rich across all of open loop righ now. Better rich than lean though!

Sorry to bump an old one but I wanted to check to see if anyone has a "finished" maf scale for this intake. I'm retuning my car (bad dyno tune...) and need to get the maf scaling sorted first and thought I might check here for one that is a good start. Thanks!