AVCS Tuning

[HolyCrapItsFast]

Awesome. Duration issue will be combated by the cam profile I wanted anyway: shorter duration high lift.

Longer duration suits turbo applications though... Just back off on the advance and you will be okay. You need the duration for larger turbos... Otherwise you will be going backwards... IMH (edjumacated) O.

 

[HolyCrapItsFast]

I don't doubt you one bit... It just goes against all I learned about component matching... but then again everything I learned was back when variable valve timing was a wet dream.

You know your gonna have to reveal you secrets to me sometime :lol:

For now I'm sticking with old school knowledge.

 

[Td_d]

Yeesh, I leave this thread for a couple of days, and reks and fuji are bashing it out... :tard:

So basically what you're saying Reks is with a bigger turbo / cam builds the AVCS can result in failure - not justifying the potential upside?

 

[HolyCrapItsFast]

For those looking to street tune AVCS, I have settled into a method that I feel comfortable with. You simply do a run with your table zeroed out... then do a run with your table populated with something like 27* or more if you are comfortable doing so. Then use VirtualDyno to compare the two and you will see were, on the graph, the car is making power with each. Just extrapolate the results and populate your table accordingly. Done! Very Simple!

You can add more runs for advance in the middle but I find that you can simple interpolate the mid range because the effect is mostly linear.

I'm going to try this method some more and update the Tuning Guide.

Just to recap my previous method I would compare MAF Voltage and Load between the runs and make adjustments based on that, but I find that there seems to be little resolution and the results are not as obvious as using the VirtualDyno method.

 

[HolyCrapItsFast]

It does and is still a perfectly valid method. :tup: It's just that the results are more obvious on the Dyno graph.

I have yet to really focus on the low load region of the map. I generally focus on mid and high end where I generally want to make power and/or torque. For the low end I leave it stock for drivability and emissions. I should start focusing on that region.

 

[HolyCrapItsFast]

Yeah but what good is spool if your not making power or torque. They are really all realted and in the end the higher I can get my lines to be the better... wether or not it is the turbo that is getting me there.

 

[Td_d]

For those looking to street tune AVCS, I have settled into a method that I feel comfortable with. You simply do a run with your table zeroed out... then do a run with your table populated with something like 27* or more if you are comfortable doing so. Then use VirtualDyno to compare the two and you will see were, on the graph, the car is making power with each. Just extrapolate the results and populate your table accordingly. Done! Very Simple!

You can add more runs for advance in the middle but I find that you can simple interpolate the mid range because the effect is mostly linear.

I'm going to try this method some more and update the Tuning Guide.

Just to recap my previous method I would compare MAF Voltage and Load between the runs and make adjustments based on that, but I find that there seems to be little resolution and the results are not as obvious as using the VirtualDyno method.

Interesting - what about fueling though - surely it throws the crap out of it?

 

[HolyCrapItsFast]

I do keep an eye on afr's and if they are out I will make adjustments before I do the run but I have only observed a 2% max difference ever. Maybe 3% on bad days. I have never observed ill effects with a 2% fuel difference and in those cases I will not make adjustments. Most of the difference comes from the affects of adjusting AVCS and that's what I'm focusing on. Mostly it is enough for me to make my decisions as to what the shape of my map will be. Hell if my afr's are withing 2% that is good enough for me to call it a tune regardless whether I am tuning AVCS or not but once all is said and done I will make a final adjustment.

It is not an exact science and there is enough room for error to drive a truck through. Ultimately you will be +/- just a few torques and power so I'm not going to kill my self tuning something that is going to yield marginal gains in the first place.

Of course I am talking about single AVCS here. It is with the Dual AVCS where things are greatly impacted by afr and in those cases I will take the time to adjust fuel. In those cases I have to make 4 runs in stead of 2 to get my final results.

 

[Td_d]

Hmm - definitely keen on testing this approach out, looks very promising. Thanks for sharing, George!

 

[Td_d]

So how exactly do you approach the dual avcs - zero intake and exhaust - and then do both at 27? or one at 27 then zero, etc.?

 

[HolyCrapItsFast]

I have tried both approaches in that I would ajust one individually of the other to see what the optimal setting is for each...

and I have tried them together...

I find the individual approach more yielding.

I can't be to sure I am doing that correctly though.

 

[HolyCrapItsFast]

Here is an example of the results... This is an 04 single AVCS with Cosworth Big Valve heads and 272 cams on 92 octane. This was only a 20* delta because I didn't want to risk clearance issues. The key here is the torque. Look how low it starts with such big cams and it maintains that right through red line.

So far I'm quite happy with this but I would like to squeeze more power out of it. Not sure I will be able to get any more out of 92 though. I have to go back and look at timing.

All in all not bad for an internet tune :tup:

 

[Td_d]

Ok - so if I'm getting you right - at 3000, you can see that you need much less than 27*, whilst at 4500 to 5000, 27* is the sweetpoint - and then you would interpolate relatively linearly from around 3200 at say 10* (since some advance did give more power) to 27 by 5200 - and then start reducing again to 0 by 5600 or so?


Genius George, some much easier...

 

[HolyCrapItsFast]

You have the jist of it but that example was only with 20* You need to find your sweet spots.

 

[HolyCrapItsFast]

And remember to look at your load versus RPM in those areas

 

[HolyCrapItsFast]

and I'm at zero by 5700

 

[Td_d]

Sorry, 20* - but I see the approach. In essence, you would also want to plot the rpm vs load curve, as you would also need to extrapolate vertically in the table.

 

[HolyCrapItsFast]

Sorry, 20* - but I see the approach. In essence, you would also want to plot the rpm vs load curve, as you would also need to extrapolate vertically in the table.

Yes sir... exactly!

 

[toomes]

What turbo is that?

 

[HolyCrapItsFast]

I think it's a dom 3