Time based access TPSdot and ms pulse width

[softopz]

Good Read

[seattlejester]

Just to make sure I am delivering values.

I didn't realize I had to buy a megalog viewer thing to view the logs, I'll have to look into it a bit more, but I can at least see the beginning of my logs.

Idle is about 1200-1300rpm, I need a stronger spring to close it a bit more, but that is what it reliably returns to. AFR is 13.6 at this value.

My previous AE curve found from a restore point:

MS added    1        5      10      11.5

TPS DOT     20      400    800  1600

With it running that AE curve, the car was running rich down to 10.X on tip in with an aggressive throttle stab. On a slow roll in the car would lean out to about 15-16AFR.

That would mean I should probably up the first value and lower the second value no? That would flatten my curve a bit, probably indicating that as is I probably have too much fuel in my tune.

 

That post above makes a lot of sense really. So it might indeed be worth while to turn the required fuel down and up the VE since my values are from 16-90 essentially.

What would be a good starting point, halfing like stated? With my settings it is saying a required fuel of 7.6 if I reduced it to say 3.8, would I then double the VE table values?

On the upper end it seems like a generated VE table has double the values and an actually decent value of 75-204, would it be worth while tuning off of that?

Right now I have done exactly as it says, lowering the req fuel and scaling the VE table and saved that. I will have to play with the AE curve as well with the new info.

I am a bit lost on figuring out the timing curve, I think I will have to leave that alone until I get the fuel more sorted or in the hands of someone who knows a bit better.

[Chickenman]

40cc injectors with Batch fire? . Try 4.8 ms Req-Fuel. Set to Injector scaling to alternating ( Smooths fuel rail pulsations ) and squirts to 2. alternating

 

Re-scale VE table via generating a new VE Table. Set parameters as per attached picture. I think this will be a good starting point. You will of course have to retune idle and Auto-Tune. 

 

Edited October 6, 2017 by Chickenman

[Chickenman]

1 hour ago, seattlejester said:

Just to make sure I am delivering values.

I didn't realize I had to buy a megalog viewer thing to view the logs, I'll have to look into it a bit more, but I can at least see the beginning of my logs.

Idle is about 1200-1300rpm, I need a stronger spring to close it a bit more, but that is what it reliably returns to. AFR is 13.6 at this value.

My previous AE curve found from a restore point:

MS added    1        5      10      11.5

TPS DOT     20      400    800  1600

With it running that AE curve, the car was running rich down to 10.X on tip in with an aggressive throttle stab. On a slow roll in the car would lean out to about 15-16AFR.

That would mean I should probably up the first value and lower the second value no? That would flatten my curve a bit, probably indicating that as is I probably have too much fuel in my tune.

 

That post above makes a lot of sense really. So it might indeed be worth while to turn the required fuel down and up the VE since my values are from 16-90 essentially.

What would be a good starting point, halfing like stated? With my settings it is saying a required fuel of 7.6 if I reduced it to say 3.8, would I then double the VE table values?

On the upper end it seems like a generated VE table has double the values and an actually decent value of 75-204, would it be worth while tuning off of that?

Right now I have done exactly as it says, lowering the req fuel and scaling the VE table and saved that. I will have to play with the AE curve as well with the new info.

I am a bit lost on figuring out the timing curve, I think I will have to leave that alone until I get the fuel more sorted or in the hands of someone who knows a bit better.

Change the first value to at least double your idle PW value. After you have loaded up the new Req-Fuel Values and re-scaled your VE table as per post above. Once you get the Tune dialed in and Auto-Tuned so that it's not running overly Rich every where, you will probably find that you have to raise the Slope angle of the TpsDot curve.  Engiens don't need much AE when they are running extra Rich.

 

Make sure that you change the TpsDot threshold down to about 20. 

[Chickenman]

I think your Spark Table is also a bit too aggressive for a Baseline Tune. Knock about 3 degrees off everywhere under Boost to start off with. The last thing you want to have happen is to be Lean on Boost and have too much ignition advance. 

 Auto-Tune does not react fast enough under Boost to correct a Lean condition. It's always better to start of too Rich under Boost ( or even at 100Kpa ) and then let Auto-Tune pull the fuel out. 

In fact, I would take the above table that I posted and multiply everything above 2,700 RPM and 110 Kpa by a factor of 1.2 just to be safe. Always better to pull fuel out than to be too Lean.

After re-scale:

 

Edited October 6, 2017 by Chickenman

[seattlejester]

Thank you for the feedback!

440cc* injectors, I am not sure if that changes the math.

I wish I had my laptop with me so I can adjust things real time.

Could I bother you for how you arrived at 4.8ms? Is that just a bit of an arbitrary decision compared to halfing it? The hp and tq number seems to be quite low compared to even modest estimations. Would it be worthwhile to overestimate in the generator then have auto tune pull fuel? My friend with the older less efficient, slightly smaller model (56 vs 57mm)  turbo on his single cam 2.7L at 10lbs of boost is making an estimated 320hp to the wheels on virtual dyno. My slightly bigger more efficient turbo on my twin cam, 3L is making 13-15lbs at the moment, so I estimate about 350hp with similar torque. 

Currently my tune is set to 3.8ms with the VE tables doubled from previous. 

I will change it tonight to 4.8ms req fuel, and generate a new VE table with an estimate of hp and tq.

Then richen it up in boost (above 100kPa) by adding 20% via the scalar

Spark table I will pull 3 degrees, so retard timing by lowering the numerical value under boost from (19 down to 16 for example), under 100kPa.

Threshold was at 20 on one of my accel graphs I wonder if it got bumped while I was playing with multiple files, but I will verify.

 

 

 

 

[Chickenman]

I meant reduce spark table 3 degrees overall when Boosting IE: over 100Kpa. "Under Boost "  meant when ON Boost.

I used a PW calculator developed for Haltech. It's very good. Numerous other online calculators and they all come up with the same answers. I enter the figures to get a Master PW and then add about 1ms to take care of the additional enrichment values. 4.8ms was a figure I had derived from a L28 Turbo with 440cc injectors.  Another 200cc in your engine won't make a lot of difference in the Req-Fuel calculation . 

The tables I posted were for 440cc injectors. 

Here is the Injector Calculator I use, Note : For Sequential injection use an ambient temperature figure of 25c. For Batch Fire use 35c

Injection Time Calculator V2.0.xls

Edited October 6, 2017 by Chickenman

[Chickenman]

Tuner Studio' VE table generator doesn't always.  handle high horsepower numbers well. It's not a true VE calculator. It can sometimes calculate some truly bizarre VE Tables when large HP and Torque numbers are added. If you use the MS-2 VE Table as generated by TS, it always seems to make the non-boosted area overly rich ( Too high of cell numbers ) . Particularly if you don't have accurate Dyno figures. Of course how do you get accurate Dyno figures if you haven't gotten the engine running properly yet? Catch 22. 

As a result I find it more accurate to enter the stock HP and Torque figures and  generate a table. That is normally pretty close up to 100 Kpa and at least makes the non-boosted area driveable. In boost I'll add an Extra 10 to 20% increase to the rows in ascending order. Just to be sure it's Rich enough. Always better to pull fuel out under Boost than to have to add it in. Auto-Tune is not fast enough to correct a Lean condition under WOT Boost.

Actually, I want to double check that 4.8ms Tune that I posted. That came off a " problem child " car and I really had to pull the Req-Fuel down to get the VE Table even remotely accurate and with some resolution. As TS originally calculated it we were down in the 12 range at idle... which is not good. 

As mentioned though. The Req-Fuel calculation is just a number in an equation. Reduce it to get more resolution in the VE cells at the low end. Leave it alone if the resolution is OK. 

 

 

 

Edited October 6, 2017 by Chickenman

[Chickenman]

This is a very good article on how to adjust VE Tables manually:

http://stinger-performance.proboards.com/thread/1002/megasquirt-tuning-basics-initial-startup