God I hope not. I've rechecked timing over and over. I'll see tomorrow.

 

Does the VVT tuner have a in and an out, or is it just in parallel. I don't know if there's a latency issue to be aware of, but after it's all in and working and warmed up, you might do the timing-light-to-redline thing, to make sure there's no drift. If so, you may need to add some hardware latency compensation (either from before, or from the VVTuner's delay).

Your timing shouldn't jump otherwise from this - I'm just saying with additional cylinder charge with the more efficient cam timing you may get bigger gains by retuning that map as well.

I'm really curious about your thoughts - I also ran without VVT for months and was happy when I added it.

 

the VVT wont change peak power output at all. It will make 3000-5000 much fatter.

And there's really no reason to advance before 2000 RPM. It will only serve to make the car fail at idle nad low engine speeds. There's just not enough flow rate to run smoothly at the lower RPM range. Not sure why, but possibly related to low engine speed early cam advance and inadequate passage of the exhaust gases out of the cylinder into the exhaust ports rather than letting them migrate upstream.

 

So.... Do you like/dislike the posted table? You'd drop it to ~0 till 2k?

I dunno if it's my new wheel or just getting the PID tuned, but I've got much better cam phase following than I did before, it runs a lot better.

The filled in mid range makes the top end seem less impressive. :-)

 

the little exp I have seems to agree with y8s on this: below about 2500 I really couldn't measure anything noticeable w/ advancing the cam
In the midrange, however, it was quite significant with between 0-29* feeling lazier and a full point richer with the former vs the latter

 

I've seen significant gains advancing the cam before 3000. Enough to have to balance timing against MBT to find the sweet spot. My tables all jump to 20° or so at around 1500rpm on street cars. On track cars I'm a bit more conservative as that step off torque isn't ever used. This is all N/A and Rotrex mind you, no turboz.

 

Mine was a street setup so stability and driveability over a range of conditions was critical.

Also I'm digging this up from memory since I haven't had a miata for a while. I saw obvious gains down to maybe 2500 but that's only a few rpm above where the pull started. From my few runs (not made of money/race shops/etc.) it was unclear if maximum advance produced max power or slightly less advance did. The two lines seem to crisscross a lot below 4k.

Also also I had zero head work and different intake and exhaust parts, which could make a difference.

Also^3 many other people have been doing a lot more with VVT than my early caveman VVT days. I could be wholly ignorant of the new hotness.

 

I didn't realize that VVT tuning would be so YMMV between users. Perhaps I'm in over my head again and should source this out to a more competent tuner with a dyno.

 

I used Emilio's table and noticed it did a TON of good over not controlling it at all and better than the other table posted that I tried.

I think it's one of those cases that "perfect" takes dyno tuning, time, and understanding, and putting in someone's proven numbers will get you 90% of the gain with little to no thinking.

I'd recommend set and forget, and next time you're on a dyno with hours to kill, play around with it. WAY better than almost any other approach - even with YMMV, the most you'll be off is a couple degrees this way. :-)

 

/\ Yep, a good known table, and making sure your car actually outputs commanded values, and making sure (and this for me is the most important part) that the car is stable and completely consistent doing it.

 

On that note, I briefly tried the EVO8 sensor in place of the stock one - car ran really smooth, but seemed down on power? It's kinda inconsistent in general, just an MS thing as best I can tell. The list of reasons why it might run funny and there's NO way to tell would take a lifetime to review...

Anyway, the point is, it'll run in semi-sequential and not let you know, and not control cam timing. So if you're having MS do it, make sure you check on screen for full-sync (and/or cam phase being controlled).

On MS-Droid there are some gauges that are left/right split, and I use cam target and cam actual on there - if you see them lined up, you're good, if you see one jump ahead and the other catch up, it's easy to spot the lag.

 

This is how I check the PID and VVT settings. Good idea to check with engine cold and hot as relative oil viscosity affects those settings. On the race cars I focus on hot oil temp accuracy and ignore cold oil. On street cars, I still optimize for hot oil temps but usually commute in the car a few times to check cold oil temp VVT response/accuracy, WUE and a few other temp related tables.

 

How much lag is acceptable for VVT control? Mine have a touch of lag, but I have never been able to get it better so I assumed they were about as good as I could get. I'm talking like when the target shoots from zero to 45 at an RPM threshold or something, the actual hits 45 too quickly to count any actual delay, but with stacked bar graphs you can see the lag.

 

At 2,000 RPMs, mine has about 0.2 seconds of lag to go from full retard to full advance. Once it gets there, tracking my table is not a problem/no perceivable lag (looking at both target and actual gauges).

I'm planning to change my VVT table anyways so it will run advance during cruise for higher vacuum anyways (a supercharger issue, higher vac to hold bypass valve open at cruise), so this 0.2 second lag at 2K won't matter soon.

 

Question not directed to me but I'll chime in anyway. I look for about a quarter sec to stabilize above 3500 or so. Below that I'm not too concerned if it wavers. Again, this is N/A with engines to conservatively for knock. On a boosted engine running gas, you would would want that stabilization even tighter, maybe ~100ms at anything above 2500. Since the cam timing has a profound effect on dynamic cylinder pressure, you could easily get little flutters of det in transient throttle conditions around town at lower RPM.

 

Thanks for chiming in - I'll have to check mine, I'm not sure it's that tight, and I occasionally get what I suspect is ping down low - but it's rare and not strictly repeatable when I get it.
-Abe.

 

If it's difficult to repeat, that points to VVT transient errors. Injection timing is keyed off VVT target. Ign timing is completely independent of VVT angle so a momentary bump in dynamic compression results in that flutter.

On Deviate (300whp Rotrex), I had the luxury of corn so it was fairly aggressive on bottom. Had I ever run it on 91, I would have either pulled a bunch of that sub 3000rpm cam advance out or spent another hour on VVT settings to sharpen response down low.. and still pulled a bunch of timing to guarantee no transient det.

 

After some good support from DIY, it seems the only way is to hook up the Vvt tuner as a completely standalone unit to control the cam. I will have the oem CPK setup, CAS and OCV exclusively controlled and wired to the tuner with ign switched 12v supply and its own grounds.

I have already noted the best base cam angle etc mapping tables from earlier posts. I assume I only need this data as it it is not connected to main engine EMS?
Any help much appreciated

Thanks

Peter

 

This one seems to be a thread worth resurecting. Any updates on VVT functionality in the past year and change? Does anyone have a good base VVT timming map they want to share? I am currently in the process of tuning my 2003 engine and would love some recent input. Pic just because.

 

This is the VVT map im running in my haltech. Advancing any farther did not make any more power for me (non turbo)