MIVEC tuning

[JohnBradley]

bump for road test?

[jcsbanks]

I was planning it today, but we have lovely weather and I just had to get the alcohol out and do a BBQ

[DB Performance]

Originally Posted by JohnBradley

Its funny how no matter what size motor, or how many cylinders, some things are constants. I have been building engines for 16 years and tuning EFI for 6, and it all comes back to common sense sometimes. Retarded cam at high RPM=The Loss.

Actually, on turbo cars at very high RPMs, they usually like the cam retarded. You see the same thing on the Honda K20 motors when you turbo them. They will like some advance in the very upper RPMs all-motor or with a supercharger, but you slap a turbo on there making all kinds of backpressure, they don't like advance. I have played around with a bunch of different MIVEC settings on the dyno. Most of the gains that I have got with MIVEC have been in the lower RPMs. In the mid and upper RPMs there is a little bit that can be cleaned up. Tuning MIVEC on the Evo is much much easier than dealing with the 26 fuel, timing, and cam angle maps that you have to tune on a K-Series Honda. I wouldn't want to try to fine tune MIVEC without a chassis dyno.

[240Z TwinTurbo]

Originally Posted by DB Performance

Actually, on turbo cars at very high RPMs, they usually like the cam retarded.

You are correct, but it is relative to your starting point. For the old school 280zxt motors I used to play around with, retarding the cam timing allowed for considerable gains on the top end. I once tuned a turbo honda and we jacked around with all combinations of intake/exhaust cam timing and ended up with ~+20hp on the topend, but ~+80hp throughout the midrange. Point being that cam timing can play a significant roll in power.

The only good way to tell if your mivec changes are making positive results is by using a dyno. You can datalog AFR'S, but I am unsure if you can detect 10-20hp change via AFR.

[Ted B]

Feel free to correct me, as MIVEC (and tuning thereof) is new to me, although variable cam timing is not:

- MIVEC retards the intake cam only
- Each step of "4.8" in the table = one degree of retard

Yes?

[jcsbanks]

Did logging of three maps today. I tried to control the conditions as best I could by using the same stretch of road in the same direction, starting from 2000 RPM in 3rd gear and lifting at about 7400 RPM. In between I reflashed and then let the car cruise for a while to cool down from the heatsoak from the stop and reflash, and I also let it run through the revs twice to hopefully do any "learning" it had to do. Every time the AVC-R showed a peak of 1.69 kg/cm^2. I only changed the VVT map each time, so the ignition and fuelling may be best altered once this has been done, but I left them alone. There were some knock counts usually during spool up of up to 3, these were less marked with more VVT advance (ie RS or RS+ maps). The JDM GSR map felt most sluggish to spool, otherwise I couldn't tell any difference.

JDM GSR
JDM RS
JDM RS+ with 28.8 instead of 24 at 3500 RPM, and +4.8 from 5000-7000 inclusive

Acceleration times (average VVT angle through this interval) in 3rd gear:

3000-4000 1.20s (14.4), 1.13s (24), 1.10s (28.8)
4000-5000 1.03s (24), 1.00s (24), 1.02s (24)
5000-6000 1.09s (19.2), 1.08s (14.4), 1.08s (19.2)
6000-7000 1.25s (0), 1.24s (0), 1.22s (4.8)

Total time 3-7000 4.58, 4.46, 4.41

[Ted B]

Originally Posted by Ted B

Each step of "4.8" in the table = one degree of retard

Ok, no need for correction, as I've been informed this is not the case and the table refers to crankshaft degrees. Needless to say, it certainly seems like a great deal of adjustment.

[JohnBradley]

So JC, it appears that I have lead you in the right direction as far as high rpm advance yes? I am running 5.4 now in the 7000+ (instead of the 4.8 that I have posted in that map) and it made little if discernable difference. It does however feel ALOT better over stock, as is evidenced by your quicker acceleration as well.

[jcsbanks]

My start of 4.8 degrees was very gentle, I did wonder if it felt smoother at the top, but I'm a sucker for placebo so now I'm quite cynical.

The only difference I would say was signicant is the 3000-4000RPM range where I saw higher airflow readings, earlier spool up and 5% quicker acceleration from doubling the cam advace from 14.4 to 28.8 degrees.

Thing is, the RS map was IMHO pretty good already, and all of them were fine from 4000 RPM upwards.

It is interesting to compare acceleration times from working out my gearing/tires etc:

3rd gear, 1.65kg/cm^2 mid, 1.40 top. 98 RON full tank (55 litres) no additives, 30kg luggage.

30mph 0s
40mph 1.25s
50mph 2.17s
60mph 3.06s
70mph 4.00s
80mph 5.07s

The 10mph increments are about 1 sec at the top end, it took a lot of mods to my Subaru to get that far and eventually ended up at 0.8 sec in a lighter car but nothing like the power band for similar octane/cubes. It is nice to see what standard performance cars this would also make mincemeat of

The stock version of my IX at 345 BHP takes 4.8 to do 30-70 in 3rd whereas I'm at 4.0. With just exhaust, pump and reflash I'm quite pleased with that.

[JohnBradley]

Next time I make it to a dragstrip I will be able to get a little more reliable information as far what gains have been found. Like you I am running 28.8* in the lower range (marked improvement) and really have only altered from 6500rpm up where it went from 9.6* to 0*.

Last time I ran at the strip I couldnt help but notice the car felt dead above 6500rpm. I was going thru the lights at 7300 (4th and 109.1mph) and kept wondering if I would have been better off to shift one last time. I have access to sufficiently secluded and straight roads where I can say without a doubt the car DOES NOT feel like it did above 6500rpm in 4th. It rails like a freight train now. How much advance, in what loadcells, these are variables that a dyno and Evoscan would work very well in determining. I just purchased EvoScan the other day and really havent had much time to play with it (or get software to map the data..like DLL) so after I get some logs we can debate gains some more. Butt dyno however says gain has been achieved.

The current hypothesis that this is all crank degrees of timing (and therefore twice cam degrees) while I am not ruling it out has me interested what a little more might do. Like I said though a dyno and/or EvoScan can probably tell us what we want to know. When I get home today after playing, I will upload a copy of the new MIVEC map for you to look at. If you want PM me email addy and I can send you the hex to copy as well.

Cheers

[JohnBradley]

OKAY, so we have an interesting development. Besides the fact that I have unconfirmed specs for the IX cam (not going to disclose till I have hard data in hand), Ted B is 100% correct that the timing map operates in crank degrees not cam degrees. This means that our maps are half of what they need to be. When I get home tonight I will post new and improved MIVEC map.

[Ted B]

Originally Posted by JohnBradley

This means that our maps are half of what they need to be.

Be careful, advancing or retarding the intake cam some 15+ crank degrees or more seems like quite a lot. 20 deg in the map is like 10 deg on cam gears!

[JohnBradley]

Noted, however factory JDM maps have 28.8* crank degree advance across the board near peak torque in all load zones. Refer to page 1 for JDM RS map. So in the interests of keeping people from doubling everything on the map, which is not what I meant, please observe the following:

AT YOUR OWN RISK YOU MAY FIND THAT INCREASING FACTORY MIVEC TIMING ABOVE 6000 RPM ANYWHERE FROM 2-4 CAM DEGREES (4.0 to 8.0 on MIVEC map) RESULTS IN MORE TOP END POWER.

Please do not do this if you are not willing to take risks, have problems focusing, are pregnant or nursing or are trying to become pregnant. Sexual dysfunction resulting from broken car is totally the responsibility of the driver and in no way my or anyone elses fault cause we didnt hold a gun to your head.

[Ted B]

I saw that. It would greatly enhance perspecting to know the static specs of the IX cams.

[JohnBradley]

Check your PM's on the "other board" where we also have this conversation going.