forgetting about actual timing chain problems - ie jumped tooth or chain elongation, because I think the ECU will pick up on both these events...

I think the problem is the MIVEC system just can't react fast enough...

I will have some more data tomorrow.

What does oil pressure do under WOT? Does it go up or down or stay the same??

 

Goes up with engine speed (its just a gear driven pump).

 

Looking at my WOT logs, in most part the intake mivec is less advanced than the target (target 30*, actual 28*). However, in some areas of the same WOT pull, the intake mivec is more advanced than the target (target 30*, actual 32*). There seems to be some fluctuation but I don't think it has to do with the reaction speed of the mivec system. I don't know how oil pressure is being regulated to the mivec valves but it may have something to do with uneven oil pressure.

 

yeah I see that in my logs as well... slight change in advance/retard even though the map doesn't change...

im assuming oil pressure diffs would be the answer..

 

On a perfect working mivec system on Evo 9 cam angle change speed on intake is 12-13ms for 10d cam movement with oil pressure more than 3 bars above 2000rpm.For comparison on an avcs system is about 10-11 ms for 10d cam movement and about 10ms on an ivtec system.Haven't checked the Evo X yet.
Most of the times the reaction time of mivec is affected from oil quality and especially from old contaminated oil leaving sludge inside the cam pulley.I have seen cases where cam angle change speed has dropped to 1 sec for 10d.If it goes slower than that then the pulley starts to get stuck.
About target cam angle and actual cam angle differences can be found in many occasions e.g. on 2 stock Evo9 engines I have found 3d offset change for fully retard reference.This can be from production line tolerances and has nothing to do from missaligned timing marks on the belt.Same thing can happen from different cams or from modified engines where the engine block or the head is much sorter from resurfacing.The worst example will be in avcs applications were you can find differences from left to right cam more than 10d.
Tephra have you tried to lock your exh cam lets say to 15d retard and check your knocksum?Mine is not affected.

 

well for some reason the cams arn't zero'ing in on their target properly... the exhaust cam seems to be more problematic... seems to be 5* out when its in fast transistion load cells, and then out 2* at peak torque...

anyways i will have a log to post up tomorrow morning..

we need someone with a service due to test this out as well, and then test after the service to see if oil quality makes any difference..

 

So, what about the opposite situation? What if someone put in thicker oil? Would we hit our targets better, or over shoot? (I'm also thinking that super cold weather could cause the same thing, its -5*F / -20*C )

 

It might work... but also thicker oil might be to viscous for the small oil passages to properly lubricate everything. I use a 0w-40 oil currently... I need to check my MIVEC and see what it is doing. So finding something in the middle is usually best. I think a 0w-40 /0w-50 would work well. However, these exotic oils also come with a higher price as well. I know Redline makes both of those viscosities, M1 makes a good 0w-40.

 

So after looking at my logs it does seem that exhaust retard actually retards past the target value... I have it set for -10 Target and I get -11.2 for Actual.

So it seems that setting it a bit lower is a good idea if you want -10 WOT to set the cam to -8.8 for the Target to get a actual of -10.

 

watch the mountains!

Maybe we have the scaling or something wrong on the mivec-e cam table.

Maybe there is meant to be a 2-3* offset??

 

so it sounds like you guys found the addresses to log actual intake and exhaust mivec then, what are they? can someone give me a quick "how to" modify the hex so i can log them please?

 

You just add the definitions to Evoscan... select proper logging byte and formula.

 

Have you tried to log while just cruising? Because if it levels out to the same value, ( Exhaust_CAM = Exhaust_CAM_Target ) then your scaling is correct.

Trying to find the correct scaling during a WOT pull seems like adding too many variables into it.

 

I did log during cruise... the intake and exhaust target and actual are very very close to each other... it is only in WOT where there is a difference in the values maybe because the system is always trying to overshoot to meet the target value...

 

yeah good point...

/checks logs..

...

yeah its the same at cruise... so the formula is right.

so it begs the question why intake cam timing usually gets there but exhaust cam timing doesnt..

how does the oil passages work? any difference in intake vs exhaust cam?