AVCS Tuning

Right. I've been contemplating whether to open this debate up or not, given that I have received advice that is of a confidential nature regarding the tuning techniques.

So upfront - no, I am not going to discuss anything that has been disclosed to me in the course of that debate, or the process suggested. What I would like to do is get a debate going that can broaden the understanding of what AVCS is all about, the implications thereof with respect to tuning for power (and potentially fuel economy - there are threads out there on this matter), and how a self-tuner could approach this 'mysterious' component of the tuning process. There are so many views out there on this issue that a moderately knowledgeable self-tuner, let alone a newbie, can get frustrated and confused very quickly.

I'm going to reserve a couple of posts upfront to allow for the expansion of this discussion, and in order to get some real value out of this - please, value the 'elite' nature of this sub-forum. What goes on camp, stays on camp. Otherwise, this will simply revert to the million and one other debates out there - as Fuji has said before, like watching kids with crayons.

Some provisos as well:

• Do not take the process of AVCS tuning lightly. You will blow shit up. With timing, as you iteratively add timing, you get to the knock threshhold and can then take it back a little for a safe timing map. With AVCS, minor changes can lead to massive knock. I speak from experience here - think 1 degree change causing 4* to 6* timing being pulled due to knock. At low RPMs. Baby steps.
• Whilst all the various tuning variables are interrelated and affect one another, this is particularly so for AVCS, as it fundamentally affects volumetric efficiency - and hence pretty much everything else, particularly timing and fueling.

Ok... I could some assistance in structuring this debate - I can always go back and edit. I'm not going to be able to do it all tonight (well, night here) so I'll reserve some posts - please feel free to contribute. I was thinking:

1. A short primer on Subaru's variable valve timing system.
2. Some approaches in the public domain, I've seen a couple:
- tuning for peak volumetric efficiency at each load / RPM areas (I've seen this approach on Romraider, with some excel tools)
- Broader approaches - I've seen a couple out there as well - and the pros and cons:
- extending the intake advance and retard out for maximum overlap
- tackling the lower RPMs and higher RPMs separately
- more targeted approaches (here I cannot elaborate - I will leave it to the discretion of those in the know as to whether they join the debate or not)
- implications of and limitations of various approaches (e.g. over extending intake advance from what I gather can actually reduce power at the top end due to not maximising the VE
3. Possible tactics - fueling / timing / intake / exhaust
4. Please add!

Also - I'm not an engineer or a mechanic - there are people on here who have far greater knowledge in these areas - so I'm not stating a position - but opening up a discussion. I do have an aptitude for working with numbers however

Reserved 1

Reserved 2

Reserved 3

Reserved 4

Yeah I'm not sure what I'm gonna be able to bring to the table here. What ever I have come up with over the years I have derived purely on my own. This is an area of the tune that I admittedly need some guidance on or at the very least some confirmation that what I am doing is correct or at least "A viable method" I would hope that Fuji and the like would chime in and give some advice.

So what ever I bring to this thread are my own findings and such, and it may not agree with what anyone else is doing but it is what I have come up with without the benefit of a dyno or any guidance. The other handicap I have is that I only have access to a car with intake AVCS and very limited access to ones with exhaust so everything I have observed is with respect to intake AVCS only. Now I do understand the dynamics of the exhaust AVCS and it's implementation but I have not been able to confirm my understanding into real world trials. For that I will rely heavily on what I read in this thread moving forward.

So please tell me if any of this makes sense...

I have always been very interested in the relationship between valve timing, fuel and ignition timing and what effect each one has on the other. I haven't come to many conclusions except for one. I have found that increasing the valve timing in the low load and rpm range is not very effective and that running peak VE in those ranges is counter intuitive as well. So when I tune AVCS in the low load/rpm range I tune it for drivability and I ignore VE. Now that with respect to ignition timing, I have noticed that decreasing valve timing in those ranges has also caused me to reduce ignition timing in those same areas. Not so sure why that is so and maybe I will comment further once I have heard from someone on that. The only thing I noticed with respect to fuel is that the AFR's differ slightly and I just end up rescaling the MAF as a result. I haven't found if the AFR's swing either rich or lean as a result of reducing valve timing because I get different result every time.

Now with respect to the WOT and higher load/rpm range I do this a bit differently. This range I try to tune for peak VE and to do so I gauge what the MAF is seeing and what the engine load is doing. Not sure if that is correct but I will tell you that my ass feels a nice big difference when I do it this way. I have noticed that more advance in this region is very effective for my setup... and I stress "my setup" and I am noticing that adjustments here have much more discernible differences in fuel and not as much as I though for ignition... Again I'm not so sure why but my ignition timing remains the same in these regions. I noticed that back when you were adjusting your AVCS this was not the case with respect to ignition timing because I can remember you have some knock issues after advancing your valve timing. This may be because of the additional dynamics introduced by the exhaust valve timing... who knows.

So in turn I will adjust my AVCS for peak VE and then take a look at fuel and ignition and make the necessary adjustments and then I will repeat the process till I can go no further.

I know others will disagree to what I'm about to say next but when you are dealing with intake only AVCS you can forget about utilizing that for the purpose of spooling your turbo for two reasons because it is a trade off as to how you make power. One, you can adjust your intake valve timing to aid in spool but you loose VE or two, you could tune it for peak torque and VE and loose some spool time... I would rather tune for VE and torque and then employ the rich spool method of Timing and Fuel to improve spool. (Which works amazing BTW). Now I know this is not true when you factor in the exhaust AVCS because you can use that to evacuate the cylinder of gases faster and improve spool.

I don't know... This AVCS stuff hurts my brain which is why I never comment on it. I'm still experimenting and probably will be for quite some time unless someone wants to mentor me.

As soon as I get a chance, I'll start updating the OP - but as I come across interesting information, I'd like to post it up.

This discussion really caught my attention - unlike many other iterations of the intake AVCS, this one is very aggressive, and more interestingly - it is a JDM Sti stock map for an '06. Have a look at this thread - but here's the map. Look at the shape:

Wow. You say that is the stock JDM map? I have seen many JDM maps and I never saw one that looks quite like that. I also compared the USDM map and the JDM map for the 06 because I have an 06 and I'm pretty sure the load does not go to 5.60. :-o This also looks like it would create way to much overlap in the low load/rpm range.

I have the maps at work so when I get in on Monday I will post them up.

Yeah - that's what I found so odd - but apparently off an Australian '06.

Yeah, it looked too 'blocky' / lacking in resolution to be OEM. I know the JDM maps tend to be more aggressive, and have much less emissions related concessions, but the overlap at the low end surprised me. I'd like to see if a model name is referenced in the thread to check the base map for comparison.

What did interest me aside from that is the tapering, increase and further taper at the peak torque areas. What's that about...

Ok, I misread - that is an altered map. What was being said was that the JDM maps run intake advance of 45*.

Still interested in what was being done at the top there, seems counterintuitive. From everything I've read, you want to taper it down to zero beyond the cam roll off point, which on stock cams is around 5200.

Oh, and Holy, what do you use as the yardstick for VE - engine load over MAF?

Td_d said:

Yeah, it looked too 'blocky' / lacking in resolution to be OEM. I know the JDM maps tend to be more aggressive, and have much less emissions related concessions, but the overlap at the low end surprised me. I'd like to see if a model name is referenced in the thread to check the base map for comparison.

What did interest me aside from that is the tapering, increase and further taper at the peak torque areas. What's that about...

Click to expand...

To be honest with you it looks like a practical joke. 5.6 load, 49 degrees, the dip in the middle... I don't believe the AVCS mechanism itself is even capable of more than 40*. I believe that is it's mechanical limit even for the JDM units because it is exactly the same hardware as the USDM stuff.

Td_d said:

Oh, and Holy, what do you use as the yardstick for VE - engine load over MAF?

Click to expand...

I look at them both together but since VE is directly related to flow I look at MAF a little more for VE and use load more to gauge torque. Again I don't know if this is right... It just makes sense to me in my head.

Yeah, it looks extreme - but - look at this legacy intake map http://forums.nasioc.com/forums/showthread.php?t=2221402 - see the taper up and down at the high loads?

I really wish I had some kind of theoretical process to approach this.

Hmm... That's why I'm curious - if you read the rest of the thread, the guy is legit, and the results are seemingly there. It looks like one also needs to factor In the difference in cam initial position between the JDM and US / EUDM.

What I'm keen on learning is _why_ .

Man, I gotta get to sleep. It's 3am here...

Maybe the profile, but the load and degrees are unrealistic for an STi unless he is running some un-godly turbo and some after market timing mechanism. Even then the profile doesn't make sense to me unless you could elaborate. :tup: I'm just trying to understand this

Plus will an Evo profile be good on a Subaru? I've had a look at some Evo profiles and none of them seem similar but I only sampled a few. I have been comparing AEM maps lately.

Well - I started experimenting with the intake AVCS at the lower load areas - I previously had high advance in the cruise areas (25*). Tapered it down from the 25 I'm running in the spool to high load areas, right down to 10*. Without a doubt, driveability is greatly improved. For lack of a better description, it feels more 'peppy' at the botttom end now - more responsive.

I'm also up to 28* in the high load / up to 4400 RPMs load (after which I gradually taper down to zero by 6000 RPMs). The car is pulling like a freight train, you can feel the torque, it's stupid fast. Haven't analysed the WOT logs yet to see where torque is peaking now, but I suspect earlier. Also, I am sure that VE has improved, based on higher MAF figures. Get this - in 3rd gear, I hit 4.78 g/s :shock:. Didn't do any 4th gear full WOT runs, too much traffic to do it safely, but I know with lower intake advance I was already hitting 4.84 g/s - leaving only 2 cells to maxing the MAF out. I would not be surprised if I'm about to do that...

When my power comes back on I have to post my map... this sounds like the same formula I ended up with.

Sent from my Desire HD using Tapatalk

Oh and I think you mean 4.8 volts not g/s lol.

Sent from my Desire HD using Tapatalk

HolyCrapItsFast said:

Oh and I think you mean 4.8 volts not g/s lol.

Sent from my Desire HD using Tapatalk

Click to expand...

Chuckle, woops, yup 4.8v - more like 360 g/s :tard: Yeah - the map is similar to the one you PM'd me, also nearly as aggressive in terms of advance at the top. I suspect I can push it to the maximum 30* intake and still gain. Then I need to start thinking about Exhaust AVCS. Looking at some other maps out there, I've seen advances up to 32* in the mid RPMs regions - mine only extends to 20 currently, so looks like there is a lot of scope to push it there as well.

You need a larger tube. Also, keep playing with her. Increase the max AVCS advance and play with her until you don't get any more results.

Click to expand...

Definitely going to continue trying to advance at that RPM band as far as I can push it. Interestingly enough, fueling barely changed this time, pretty spot on. I'm running an 80mm intake now, and it's obviously not enough... Should probably move to a 4" and be done with it. Who knew you could squeeze so much out of a Dom3 xtr - makes me wonder what I could do with a Dom 5 and some decent tuning to get better spool.