[TheMadScientist]

Does a 0 in this map = 0* advance = 0* overlap ?

TMS

[Christian.]

0 in this map means 0 advance of the stock intake camshaft for that particular RPM and Load point.

[daemon]

I think the question is more along the lines of... is zero advance equivalent to no overlap and positive advance equivalent to increasing overlap?

[Christian.]

I have not analyzed/dialed the stock cams on a cam degree machine, so I do not know what the overlap is at the various degrees of AVCS advance. I apologize I misunderstood his question.

[TheMadScientist]

I guess it was sort of both.
I just figured 0* advance would = 0 overlap. But obviously the start position could be any amount of overlap and 0* advance could have some amount of overlap built in. I wrote an email to SOA to see if they will tell me.
In the AVCS thread on wrxforum some said that 0 in the map was max advance/overlap. So that was part of my question too.
Thanks

TMS

[TheRipler]

Originally Posted by TheMadScientist

In the AVCS thread on wrxforum some said that 0 in the map was max advance/overlap.

That would be me who suggested it. That is what I was told by someone familiar with ECUTEK, which is a different product. Seperately, I have come to the understanding that some overlap at higher RPMs will increase volumetric efficiency (VE) of a motor, resulting in more power. If that is true, then the provided maps make more sense as cam retard.

Keep in mind, I'm just learning this stuff, and I may have missed a major factor somewhere.

[TheMadScientist]

Originally Posted by TheRipler

That would be me who suggested it. That is what I was told by someone familiar with ECUTEK, which is a different product. Seperately, I have come to the understanding that some overlap at higher RPMs will increase volumetric efficiency (VE) of a motor, resulting in more power. If that is true, then the provided maps make more sense as cam retard.

Keep in mind, I'm just learning this stuff, and I may have missed a major factor somewhere.

I was thinking about that last night. With a small turbo like the VF39 the pressure in the exhaust could be much higher than the intake at higher load. So you would want to dial out some of the overlap to prevent reversion as much as possible. That could explain the reason the maps look the way they do.
My understanding is that on a N/A car the way you described is correct as far as I underrstand. The reading and theory I have been reading has all been about N/A cars.

TMS

[TheRipler]

That's an interesting point. [Edit: ~Almost~] everything I've found written about variable valve timing systems has been on NA cars as well. Once again, I'll provide a disclaimer. This is my first car with variable intake valves, and only my second with a turbo. Still, I'm inclined to believe that VE is VE, whether turbo or NA.

While there will be pressure on the exhaust side, there is also pressure on the intake side. Both will be in pulses, as the valves open and close. Moments before the intake opens, pressure will be building on the intake side. The exhaust gasses leaving will still have momentum, though it may be less from the back pressure of the turbo.

Last spring, I found this Turbo Calculator. That was before I started wondering about AVCS. If you play with VE, you can see what a difference it makes with power output.

There is also the Winter 2005 Drive article from Subaru Explaining AVCS.

At heavy engine loads: When the engine is used aggressively for greatest performance, AVCS advances the intake valves further to open even sooner during the exhaust stroke. This produces a scavenging effect – that is, intake airflow helps clear the cylinder of exhaust gas. It also closes the intake valves sooner on the compression stroke.

This results in improved volumetric efficiency and helps to generate higher power output.

All this makes me wonder about the length and diameter of the exhaust and intake, and how that effects the velocity and pressure of the pulses in either on the motor side of the turbo. ...but that's probably a different thread topic, and delves too far into thermodynamics for my limited understanding of the subject.

More importantly, we need to stick to the current question, and figure out how this map works. It would also be interesting to know the values of max advance and retard of the intake cam. Similarly, it would be interesting to know the min/max effective values in the AVCS map, and why the provided maps have all the values in multiples of 5.

[daemon]

Anyone want to set the entire AVCS map to zero and drive around for a while? Noting and logging. Then set the entire map to 25 and log and note?

[TheMadScientist]

Originally Posted by daemon

Anyone want to set the entire AVCS map to zero and drive around for a while? Noting and logging. Then set the entire map to 25 and log and note?

I have been switching between a 0-20 and a 20-0 AVCS maps. I haven't done any logging yet but I can't "feel" a difference between the two. I was trying to get the Idle to change by playing with the AVCS to determine how the map worked. You know that cammed lumpy idle, I was thinking that if I dialed enough overlap into the map that I would get a lumpy idle. But it seems like the AVCS system uses a preset for off throttle like the 23*timing and injector shut off when decel. I am going to have to log some of these things to find out what is really going on.

TMS

[Christian.]

I will summarize some of my findings with AVCS in an effort to help out.

AVCS is generally advanced in the lower RPM range and not in the higher RPM range. Here is another perspective...I run a road race Integra Type R (I know, I work at a Subaru performance facility...weird). We do not have the ability to advance or retard the camshafts other than mechanical means, adjustable cam gears. We would usually advance the cam gears for road races to get more area under the torque curve at lower RPMs to allow for better power out of the corners. We would do the opposite if we took our cars to a drag race; we would retard both intake and exhaust camshafts to extend our torque curve in the upper RPM ranges, which would allow HP to grow with RPM more than with the previous settings. Having too much torque down low in a Honda does nothing but spin the wheels. Enough of Hondas...I hope the point was made. Changing camshaft timing allows you to move your torque curve...not make more peak torque per say. Now back to Subarus.

I have also noticed the motor responds much more to AVCS changes on automatic vehicles compared to vehicles with manual transmissions. Along the same line, I have noticed that different dynos load the vehicle differently and record changes in torque differently based on how they load the vehicle. For example, we use a Mustang load-based dyno which does not record large changes in torque when tuning AVCS. Now go to a different dyno (Dynomite or Dynapack) with the exact same vehicle and tune AVCS...the dyno will report huge movements in torque. I am not trying to start a pissing contest, just trying to educate.

If you datalog AVCS you will notice that if fluctuates around the target value...this is to be expected because it is using oil pressure to actuate. With any change in oil quality, temperature, viscosity, oil pump RPM, etc. the AVCS will fluctuate. We tune AVCS in other increments than 5 degrees because that table is a Realtime table for us. I can see how tuning AVCS with other software does not allow incremental adjustments due to time and ECU capacity constraints.

To sum it all up...it has taken me up to 8 hours of dyno time per vehicle to make improvements over the stock AVCS settings. Part of my job it to go through the existing maps and refine the AVCS tables...this is partly why it is taking us so long to update ALL of our maps. I could be slow or an idiot...I will be the first to tell you. Point is, tuning AVCS is difficult and much more is involved than I stated...the rest of that information we keep in house Make a change in AVCS and you will have to re-tune your ignition and fuel curve to accommodate the additional torque, back and forth, back and forth, etc. until several hours later you finally have it...more area under the curve

Christian.
AccessECU Trainer

[WolfPlayer]

More things to keep in mind here ...

Advancing the intake cam will increase overlap and improve emissions because you get an EGR effect (Exhaust Gas Recirculation). Basically, part of the exhaust charge is used to contaminate the intake charge.

On a naturally aspirated engine overlap is great for upper RPM power. DO NOT ASSUME THAT NA MOTORS AND FORCED INDUCTION MOTORS WORK THE SAME WAY! Overlap is used such that you leave the intake and exhaust valves open at the same time. In higher RPMs this means you are opening the intake sooner such that the exhaust velocity and momentum helps to create an artificial supercharge effect. It helps to pull the new intake charge in. However, for forced induction motors you don't want this. Why? Because by leaving both valves open at the same time you will be blowing your intake charge right out the exhaust! Not good. You want as little overlap as possible once you have positive manifold pressure because you want to trap as much of the intake charge as possible in the chamber. Moral of the story: do not advance your intake cam once the turbo is spooled up. You want to treat this motor similar to an NA motor until the turbo is spooled ... and then treat it like a turbo motor after it is spooled.

Another thing to keep in mind here is that we can only advance the intake cam. We can't do the intake and exhaust cams. Generally, advancing both intake and exhaust will always lead to more power and torque earlier. You end up shifting your power curve lower. However, we can't change exhaust timing. I wish we could. Increasing only the intake cam advance will basically allow you to pump up the midrange a bit and help the turbo to spool .... but that is pretty much it.

Definitely pay attention to what Chris said about retuning ignition and fuel to accomodate the new curve ... especially the ignition!

... BTW, I love cams. I had 4 different cams in my old Small Block chevy experimenting. Cam theory is a beautiful thing.

t

[TheRipler]

Thanks, Christian and Wolf. You have raised many more questions for me, but I'll hold off on most of them for the moment. I'm still trying to get my head wrapped around this. Hopefully, I'll be able to formulate them better in the near future.

I didn't find your explanations of why turbo cams are different to be exhaustive enough to be able to balance this out in my head. It did provide me with a few more search terms for google, where I found this page on Turbo Cams. It seems to agree with you for undersized stock turbos, but goes into a bit more explanation of other scenarios, which satisfies some of the confusion I was having. Does it look sensible to you?

Christian brought up an interesting point about engine oil, and the effect it would have on the AVCS system. I was wondering about that this morning after rereading the Drive article I linked. How would a thinner oil like Mobil-1 synthetic 5W-30 effect AVCS as opposed to a 5W-30 dino oil? I may do some data logging before and after my next oil change, just to see...

Is 35 degrees the max advance value, as suggested in the Drive article?

AVCS continuously varies this overlap through an infinite number of positions. Overlap ranges between a slight overlap (“retard” position) through as much as 35 degrees of the crankshaft rotation (“advance” position).

Was the drive article written as looking at a NA motor with AVCS? That would be strange, considering most of the vehicles listed in the article were forced induction. ...but then again, I never trust marketing people.

[Christian.]

Originally Posted by TheRipler

Is 35 degrees the max advance value, as suggested in the Drive article?

No, I have gone up to 45 degrees...with no resulting gains. I do not suggest you go above 35 because of that and other (valve to valve clearance, valve to piston clearance, enormous amounts of time spent tuning, etc) reasons.

[TheMadScientist]

I just got this from SOA, I hope they get back to me.

 Dear Mr. Murray:

Thank you for taking the time to contact us. I will need to consult with our
 Technical Services Department for assistance with your message. As soon
 as I have any information, I will contact you again. Thanks for your
 patience!

Best wishes,

Natalie Cox
Subaru of America, Inc.
Customer/Dealer Services Department
------------------------------ <wbr>-----------------------------
YOUR ORIGINAL MAIL:

I have a couple of questions about the AVCS control on my 2004 Impreza Sti.

1) What is the at rest 0* advance overlap of the cams?
2) What is the limit of advance of the AVCS system on the 2004 Impreza Sti?
3) What is the limiting factor of the system?  Cam gear cavity, or ECU programing limits?
4) Does a 0 in the ECU maps = minimum advance?

Some of these have all ready been answered.

TMS