Tuning AVCS without a dyno (AVCS for Gruppe-S header post 12 and 24)

[Kodename47]

@Tor thanks for validating the tool works. I haven't been able to test on my car so was really good to see it working in a practical fashion.

Am I assuming that the yellow/green are the new calibration? If so that's a good job well done.

[solidONE]

Originally Posted by Tor

I'm not sure this will be better on an OFH, after all Shiv tuned the cams specifically for that header.

But sure you can try. Maybe the OFT cams are set conservative because it needs to run over a broad range of cars. Only Shiv will know how he decided on the OFT cams. I'm just saying this might not be an improvement for you.

If you run a lot of ignition timing, it might be a good idea to take out -1.4 deg from load 0.9 and 4000 rpm up. On my car, Vicent's and a 3rd one (all Gruppe-S) it doesn't cause FLKC. On my car it infact looks like there is room for more timing now.

Regarding gains. I did some messurements of the stock tires today vs. the 18 inch I have been running over the summer. As it turns out the circumfrence of the tires are practically the same, give 2 or 3 mm.

So the actual gains look like this:
Highest and lowest of 6 pulls today, and 2 highest and lowest out of 6 with my pre AVCS tune:

Fantastic gains!

Assuming you did not change your fuel tables, it looks like the afr readout is leaning out possibly due to better scavenging on the updated cam settings. Berry nice! I suppose you can compare the elapsed time between the logs to determine if it took a shorter amount of time to go from 3000 to 7300rpms with your new cam timing compared to baseline.

[Tor]

Originally Posted by Kodename47

@Tor thanks for validating the tool works. I haven't been able to test on my car so was really good to see it working in a practical fashion.

Am I assuming that the yellow/green are the new calibration? If so that's a good job well done.

Thank you. Yes the Yellow/Green is the new tune. Thanks.

I dismissed the low results from a few days ago as the road being damp/wet. I had to change wheels on the car since the previous tune. Again today the result was similar low using VD to calculate wheel diameter from profile/wheel size.

So I decided to measure the circumference and they are within 3 mm and calculates to 24.97 inches. My 225/40R18 comes out as 25.09 in VD and 215/45R17 (stock Primacy) comes out as 24.62. So it's way off. So for simplicity and compatibility with my previous VD logs, I will use 25.09 for both sizes.

Here all 6 runs from today and they all look consistent. Most importantly around 4000 rpm where they diverge from the previous tune. And it's actually noticeable, like Vtec kicking in. I don't know if that means there is more improvement to be made in the torque dip. I really focused most on higher rpm:


(Run 4 and 6 was taken out of the compare due to wrong shape at high rpm).

Log:
https://datazap.me/u/tor/tor-2371?lo...zoom=3688-3803



This is with a modified exhaust cam too, which I prepared for making the same procedure as with the intake.


Your AVCS excel tool is fantastic! I used it to make some more changes in the cams in low load which I tested out today and it works great.

The changes are based on the overlap phase and getting a smooth overlap map instead of two smooth cam maps with inconsistent overlap.

I make a post on that in the next days when I have time. It requires a bit of thinking and time to write.

[Tor]

Originally Posted by solidONE

Fantastic gains!

Assuming you did not change your fuel tables, it looks like the afr readout is leaning out possibly due to better scavenging on the updated cam settings. Berry nice! I suppose you can compare the elapsed time between the logs to determine if it took a shorter amount of time to go from 3000 to 7300rpms with your new cam timing compared to baseline.

No, this is due to removing ELC compensations, because I didn't want it to interfere. That hump was in the previous tune too, when I didn't compensate it. I will restore some of the compensations because my logs say it needs fuel there in CL too. I don't plan to correct it fully again since some of it is probably fake lean.

Good idea with the log compare. I will look at that.

Edit: 2 random pulls 9 vs 9.5. Probably need to check more, but I'm out of time.
https://datazap.me/u/tor/tor-2371?lo...mark=6869-6960
https://datazap.me/u/tor/tor-235?log...rk=12141-12236

[makinen]

Hi,

I've reached similar form after doing AVCS calibration with Kodename47's sheet and vgi's tool.



As you wrote, cam angle movement should be adjusted manually because that AVCS are not able to trace the raw values that picked up by vgi's tool when those values are scattered.

[Kodename47]

It's also worth noting here that changes in the AVCS impact VE later on in the rev range, by that I mean that if you increase/decrease the intake advance at low RPM this may impact the performance at higher RPM due to changes in resonance in the exhaust and/or intake manifold.

I'm sure I read somewhere that once you have what appears to be the ideal settings then you should try the whole rev range +/- x degrees to see if that nets any further performance/VE gain due to the above.

I know this is basically what you did @Tor and I doubt that any further changes are likely to net any obvious improvements with road tuning but it's worth noting never the less, especially if anyone were to get any dyno time for those last few horses. I would personally just do +/-2 degrees on exhaust and then intake to see if there are any gains to be had.

[Kodename47]

Thought I'd add this little quote I found from HP Academy:
"Start with the intake cam, then go through and get the exhaust cam to we think it needs to be. Then go one step further though and go back and revisit the intake cam timing one more time, perhaps at this point you should be already very close. I'd do one final set of tests where I'd advance the cam timing five degrees, perhaps retard it five degrees to make it all optimal.

So by going through the process of doing the intake cam, following that with the exhaust cam, and then finally going back and revisiting the intake cam, this is going to give you 98/99% of the potential results that are there in a time efficient manner. Finally, once you've actually got our cam timing maps 100% to your liking, you can then go through the process of optimizing the fuel and ignition and making sure that's all completely accurate."

[R2]

This is exactly how I approached it. It took several iterations to get close. I found that when thugs got close I had better results in looking at the VE plot and picking my own optimal cam settings rather than blindly picking what the tool calculated. The plot is usually noisey and a large noise swing can make it pick an obvious non-optimal cam position.

It's worth noting that if you use Virtual Dyno, the only way to get constent and repeatable results is to do it on a zero wind day, probably less than 10km/h can be ok too, and using fourth gear to maximize sample time. This will get you +-2lbft, at which point the cams cannot be dialed in any further on the road.

One thing I haven't explored is whether more torque be had via knock resistance with a slightly less VE optimized cam setting vs. Max VE. The torque increase of 1degree of additional ignition timing could be worth more gain than a 1% reduction in VE. So, if compromising VE a bit in certain load/rpm ranges makes for a more knock resistant cylinder charge, then it can be worthwhile. We don't have a tool for that though.

Originally Posted by Kodename47

Thought I'd add this little quote I found from HP Academy:
"Start with the intake cam, then go through and get the exhaust cam to we think it needs to be. Then go one step further though and go back and revisit the intake cam timing one more time, perhaps at this point you should be already very close. I'd do one final set of tests where I'd advance the cam timing five degrees, perhaps retard it five degrees to make it all optimal.

So by going through the process of doing the intake cam, following that with the exhaust cam, and then finally going back and revisiting the intake cam, this is going to give you 98/99% of the potential results that are there in a time efficient manner. Finally, once you've actually got our cam timing maps 100% to your liking, you can then go through the process of optimizing the fuel and ignition and making sure that's all completely accurate."

[Kodename47]

Originally Posted by R2

This is exactly how I approached it. It took several iterations to get close. I found that when thugs got close I had better results in looking at the VE plot and picking my own optimal cam settings rather than blindly picking what the tool calculated. The plot is usually noisey and a large noise swing can make it pick an obvious non-optimal cam position.

This is true, always worth a visual look at the VE plot to confirm.

Originally Posted by R2

One thing I haven't explored is whether more torque be had via knock resistance with a slightly less VE optimized cam setting vs. Max VE. The torque increase of 1degree of additional ignition timing could be worth more gain than a 1% reduction in VE. So, if compromising VE a bit in certain load/rpm ranges makes for a more knock resistant cylinder charge, then it can be worthwhile. We don't have a tool for that though.

This is an interesting theory and not one I had explored. However this is definitely something you would need time and likely a dyno to test and prove.

[Tor]

Thanks for all the input.

My idea was to follow the same method as in @mad_sb 's thread, just with the modification to stick around known good values to reduce time.

I might give the exhaust a go as well later on, but I think it will be difficult to see a difference without a proper dyno. And if I have to revisit the intake, it's a fairly lengthy process for maybe a minimal gain. Maybe later on...

I did make both small and large changes to the exhaust already. Small up high to prepare its rpm-curve shape for something I can add and subtract from. And a large change at low rpm. Basically, I reverted low rpm to near stock.

Thanks to Kodename 47's excellent AVCS excel tool, I noticed the resulting overlap from intake and exhaust was a bit bumpy. Also, I was wondering why stock and EL stg 2 maps had high exhaust retard at low rpm (incl mad_sb's findings) but UEL didn't.

When looking at the stock map with Kodename 47's tool, it becomes obvious they fit together when looking at the overlap.



Looking at the stock overlaps, they predominantly want to target 47 deg overlap at cruise loads. Taking part of either table out here will make them missing in the overlap.

Here OEM overlap copied back into RomRaider for a visual representation:



After restoring those patterns, I can actually see an improvement in fueling error were I before had some inconsistencies in fueling. It shows in the corrections in my Engine Load Compensations where the table has now become smoother.


As for the high load/low rpm area, I made 2 tunes. One with and one without the high exhaust retard. I tested the high retard myself and 2 others tested what resembles OFH with some overlap smoothing.

Also, looking at load vs. rpm vs. MAP, I'm of the opinion that it's not necessary to overly smoothen the inlet cam table horizontally. If the load is above 0.9 the manifold pressure is automatically mostly fairly high too (0.80 or more). Hence, I think it makes sense to advance the inlet cam to the optimum max (horizontally in the load scale) from load 0.9 and up. I think it's more likely to cause inconsistencies in LTFT if it sees a lot of variation in cam overlap, especially for WOT LTFT.

Looking at the stock table with squinted eyes, the stock table also seems to draw a line down at load 0.85. These are, as far as I can conclude, two separate regimes, cruise (or low MAP) and WOT (or high MAP).

Anyway, this is how my overlap looks:

High stockish exhaust retard low rpm:



Low exhaust retard low rpm:



Unscientifically, the high overlap pulls noticeably better at low rpm, but from short testing with 50 deg retard (like stock) it seems to knock and drop IAT it high gears. 4th gear and below is fine.

I didn't test the above depicted (40 deg retard max) enough to say if taking out 10 deg help. But that's the reason I made the alternative exhaust, for the two other guys to test. In any case, it's worth for me to keep.

The shapes of the cam tables on their own look quite weird and jagged, but they fit into each other to provide the smooth results above. Before altering these tables, I would recommend to check what overlap results from it and be very careful with applying smoothning to the individual tables or changing scalings without looking at the impact in overlap.

Originally Posted by R2

It's worth noting that if you use Virtual Dyno, the only way to get constent and repeatable results is to do it on a zero wind day, probably less than 10km/h can be ok too, and using fourth gear to maximize sample time. This will get you +-2lbft, at which point the cams cannot be dialed in any further on the road.

Finding a flat road here is difficult. Finding a flat road where I can rev it out in 4th is even more difficult. I have one, but speed limit there is 70 km/h (43 mph). Doing 20 pulls there, I think from the financial risk, it would be better to rent a dyno.

One thing I haven't explored is whether more torque be had via knock resistance with a slightly less VE optimized cam setting vs. Max VE. The torque increase of 1degree of additional ignition timing could be worth more gain than a 1% reduction in VE. So, if compromising VE a bit in certain load/rpm ranges makes for a more knock resistant cylinder charge, then it can be worthwhile. We don't have a tool for that though.

The changes seem to have made it more knock resistant (except as mentioned in low rpm/high load/high gear). I already added a bunch of timing before starting this out and as far as I read, optimum valve timing is also to some extent dependant on ignition timing. The gains I saw adding the last bit of timing was very low anyway. I'm going to leave it as it is and instead try to lean it out (I also run very rich at WOT on purpose at the moment).

[elBarto]

So I tested the AVCS setting last night (2am , 0°c) , not really a good log, it was freezing and felt the car slipping on some places.
Didn't feel or hear anything differently , but was to busy not crashing. I'm using Wayno's 100ron tune with good 98ron fuel.

https://datazap.me/u/elbarto/testrun...ta=1-5-7-11-13

[Tor]

Originally Posted by elBarto

So I tested the AVCS setting last night (2am , 0°c) , not really a good log, it was freezing and felt the car slipping on some places.
Didn't feel or hear anything differently , but was to busy not crashing. I'm using Wayno's 100ron tune with good 98ron fuel.

https://datazap.me/u/elbarto/testrun...ta=1-5-7-11-13

Difficult to say anything then.

Off topic, your LTFTs are a bit high. Maybe I have something that can help you with that. I'll send you a PM.

[steve99]

Originally Posted by Tor

Difficult to say anything then.

Off topic, your LTFTs are a bit high. Maybe I have something that can help you with that. I'll send you a PM.

his afr is a little rich as well :-)

[elBarto]

I think it's thanks to the cold weather my trims are higher, older logs (with higher temps) doesn't show that high LTFT's .
This is a log I made driving to work, temp around 12°c , same tune (without the AVCS obviously)
https://datazap.me/u/elbarto/log-150...ata=1-6-7-9-13

Anyway, thanks for the info appreciated