NewCelica.org Forum banner
1 - 20 of 38 Posts

·
Premium Member
Joined
·
18,200 Posts
Discussion Starter · #1 ·
Last weekend when I was on vacation, I did a little messing around with VVT tuning right at lift engagement. Since my car is turbocharged, it is extra sensitive to VVT tuning in that area, similar to but much better than the way the C2 GT-S turbos were.

Even with NA cars, people seem to find that they have trouble tuning out the slight bog or hesitation at lift transition. On my car, this is an extremely noticeable stumble and momentary loss in power. I have gone through several ideas of how best to deal with this situation.

1. The first generation was based on tuning with the default-style maps with the 500 rpm incrememnt breakpoints. People including MWR tried to lower the lift engagement with no success.

2. After people showed that lowering the lift engagement point could be done successfully, I tried an idea where I used three closely-spaced breakpoints, with the center one being right on the chosen lift point. For example, at a chosen lift point of 5700 rpm, I might have breakpoints of 5500, 5700, 5900. Using this method, I found tuning right on the exact lift point to be virtually impossible.

3. The current method I've been using is to have two breakpoints, one immediately on either side of the lift point. On my car, I've been working with a 5800 rpm switch and have breakpoints set at 5700 and 5900.

Even with this method, I'm still having trouble at the transition. I've moved the VVT at 5700 rpm up to 55 and had increased the VVT at 5900 as high as 35, with no very noticeable effect. Once you get past 5800 rpm, everything works great, even with the cam way more advanced. This tells me that there is something odd going on at lift transition, something that is being massively magnified by the backpressure my turbo causes. This brings me to my idea about how the VVT system works and the mystery of the 6000 vs. 6200 rpm lift switch ECU's.

As far as I can see, there is one of two things taking place at lift that is causing the bog/hesitation. The first is that the engine doesn't behave well if there is a transient VVT condition at lift. Meaning that something bad happens if the VVT system is moving the cam while attempting to switch to the high profile. Most people write maps with the small cam very retarded pre-lift and very advanced post-lift. This would cause the VVT mechanism to have to be moving at essentially its full speed through the lift transition. I'm not sure mechanically what effect this could cause that could impact driveability, but it might cause confusion with the sensors. The lobe centers of the two cam lobes are completely different, and moving the cam while switching its inherent timing could cause some odd issue with the camshaft position sensor, causing the running problem.

The second idea is much simpler, and probably makes more sense: the cam needs to be very retarded for a short period of time after lift. Switching cam profiles would cause a major change in resonance in the cylinder. You can hear this with the different sounds coming out of the engine. If you advanced the cam too quickly following or during lift engagement, you could cause some strange airflow behavior in the cylinder that would lead to poor running.

Either one of these ideas could explain the 6000 vs. 6200 rpm mystery. A revision in the tuning could basically hold the cam either still during lift transition or keep it reatrded for a short time after the lift transition takes place. I am going to test this idea by switching my lift engagement and testing some new VVT ideas.
 

·
Registered
Joined
·
634 Posts
Makes sense, but I thought they DID find out how to make lift transition smooth? I'm going to be getting some dyno tuning with Hydra EMS and we're going to spend the entire day on VVTL-i crossover and trying to get the maximum power from it.

I'll post up what the dyno shop finds when we do it.
 

·
Registered
Joined
·
11,169 Posts
Could one of you gents show a graph that plots changes you've made in in VVTover RPM? I'm curious to see the direction and magnitutde of change you folks have made, especially as concerns various lift crossover points. I think seeing it visually would be instructive both to PFC as well as camcon tuners :)
 

·
Premium Member
Joined
·
18,200 Posts
Discussion Starter · #4 ·
jlitman said:
Could one of you gents show a graph that plots changes you've made in in VVTover RPM?
Just look at any of the numerous Power FC maps floating around.
 

·
Premium Member
Joined
·
18,200 Posts
Discussion Starter · #5 ·
Ryude said:
Makes sense, but I thought they DID find out how to make lift transition smooth?
Not that I'm aware of. On all the dyno charts I've seen, there tends to be a small dip, which can be seen on a log as a dip into one load row lower across the lift transition.

Adding the turbo into the exhaust dramatically affects exhaust scavenging from the cylinders, which will accentuate any hesitation at lift engagement.
 

·
Registered
Joined
·
634 Posts
Maybe toyota did that on purpose to make the transition feel powerful? I won't know until we do the dyno and get to tuning it out. The shop guaranteed me they could get it smooth or they will charge me $0. So we'll see how that ends up.
 

·
Registered
Joined
·
11,169 Posts
Jesse IL said:
jlitman said:
Could one of you gents show a graph that plots changes you've made in in VVTover RPM?
Just look at any of the numerous Power FC maps floating around.
I found one, but I'm not clear on whether it;s tuned -- do you have a recently tuned map you could post a screen shot of?

Ryude said:
Maybe toyota did that on purpose to make the transition feel powerful?
I think you are right -- the jarring transition seems deliberate and unecessary.
 

·
Premium Member
Joined
·
18,200 Posts
Discussion Starter · #8 ·
I tried an experiment today that seemed to help somewhat.

My current map has rpm breakpoints at 5700 and 5900, with the lift transition set at 5800. I set VVT at both breakpoints at 50, ten advanced the cam somewhat at all higher rpm points. The problem went 90% away, but there still seems to be a slight dip in power, but not the massive misfire issue there once was.

Tomorrow I'm going to try setting the switch at 5700 and seeing if the extra hundred rpm makes a difference.
 

·
Premium Member
Joined
·
12,227 Posts
interesting theory Jesse. i look foward to your conclusion...ready for a trip to CA?
 

·
Registered
Joined
·
430 Posts
I might have something useful to contribute to this thread. Jesse you are right the engine does not perform the best when engaging the high cam while the VVT is still moving.

The problem is a dynamic one. Consider:

~.2 seconds for the VVT to move through it's range (a 40 degree cam angle change). It's actually faster than I thought.

.15 S in 2nd gear to accelerate between 5600 RPM and 5800 RPM

.3 S in 3rd gear to accelerate between 5600 RPM and 5800 RPM

The large cam wants to have a low VVT setting (max overlap) once the high cam is engaged.

So you can see that if you set up the VVT and lift so the VVT completes it's movement range just before lift in one particular gear you'll have no dip in power. But in another gear the VVT may not quite complete it's transition before lift, so effectively a higher than wanted VVT setting is in place when lift engages - causing a dip in power. Conversely, there could also be a situation for say 100 RPM or so when the cam completes it's transition before lift, so a effectively a lower than wanted VVT setting is being used before lift. This is not optimal for power either.

In other words, I am convinced there is no perfect VVT/lift setting for all gear selections. The best thing to do is minimize the dip by setting breakpoints before and after lift at RPMs that can best take into account the dynamics on a particular car - acceleration through various gears, and the VVT timing. I have some measurements I can post if I can get on my other computer to show the dynamics of what's going on.

Here is a suggestion:
VVT: 43 at 5625 RPM
VVT: 3 at 5675 RPM
Lift: 5700 RPM
VVT 4 at 5800 RPM

With this set up, the VVT completes it's range just before lift engages (but there is still 75 RPM of non optimal VVT settings before lift engages). All cars are a little different too so you'll need to experiment.

BTW, I run 5W 20 Royal Purple Synthetic now because it helped speed up the VVT and lift engagement due to it's improved flow characteristics. It's used in Acura/Honda cars, and some folks say it's optimal for VTEC cars. I don't know if I would recommend this for turbo cars as these cars may need the additional shear strength a higher weight oil provides.
 

·
Premium Member
Joined
·
18,200 Posts
Discussion Starter · #11 ·
Interesting, you have basically done exactly the opposite of what I've done currently. I have the car holding the retarded cam past the lift enagement, but I will experiment with the opposite way as well.

With the turbo car, the acceleration times through rpm windows would be smaller, so probably the breakpoints would need to be spaced farther apart. The one thing that doesn't make sense from your above info is either the time needed for the VVT system to move or the breakpoints.

You say it takes 0.2s for the VVT system to move. Next, you say the time to pull through 200 rpm is 0.15s in 2nd and 0.3s in 3rd. You breakpoint recommendations don't even seem close for 2nd and still not on the money for 3rd. Are you optimizing that for 4th gear? I understand your point about having the breakpoint 25 rpm before lift, to get the cam to basically full advance, but the first breakpoint seems like it should float depending on the gear you're optimizing for. Unless its fueling related, I don't also quite get the point of the 5800 rpm breakpoint from a purely VVT point of view. It doesn't seem totally necessary to have two breakpoints 125 rpm apart when your gradient is only 1, again unless its purely fueling related.
 

·
Registered
Joined
·
759 Posts
Jessie, I've been playing with my vvt #'s too the past week, i tried keeping the #'s high on both sides of the lift transition, but it feels way better when i use high values before lift, then low values after lift, i had higher AFL V's that way too. I'm hoaping to get some dyno time in within the next couple of weeks, Is MWR using a Power FC on the gt-s shop car??? if so, i wonder what his vvt #'s look like?
 

·
Registered
Joined
·
419 Posts
you shouldnt carry over the exact same VVT position before and after lift. The way the lift profile is the peak of it compared to the regular profile is not on the same angle. The peak of the cam lobe the exact moment of lift engagement should always occour at the same point in the down stroke that the peak of the regular profile occoured just 2 strokes previous. The reason they are not exactly the same peak is because of VVT. Toyota has VVT moving during engagement and that slight movement is compensated for my the off-set peak. Without VVT, the secondary profile woul dhave a peak that was the same as the standard profile. This is evident when looking at any Integra GSR or ITR camshaft.

Assuming you have changed the VVT position right before lift, you need to have the VVT moving at the exact same rate through lift engagement as it would be if the car was brand new and stock. I am not familiar with the stock advancement or retardedness so asking me what that rate is wouldnt help. You need to go see snakes on a plane. And you also need to look at the movements around lift engagement on a stock engine and make those movements match in change degrees around your lift engagement.

Example:
Stock
5950 rpm -> 5 degree retard
6000 rpm -> lift engages
6050 rpm -> 2 degree advance

Yours
5750 rpm -> 10 degree retard
5800 rpm -> lift engages
5850 rpm -> 3 degree retard

I hope this makes sense, ive had an AMF (1/3rd vodka, 1/3rd gin, 1/6th blue raspberry schnapps, 1/6th club soda or sprite or tonic water in a 36 oz glass half filled with ice) and 6 16oz tallboys and then some crappy sangria at dinner... started drinking around 9pm, now its 3am.. anyway i hope this helps

also keep in mind even stock it laggs a little bit unless you adjust the VVT slightly, but i dont know those adjustments, but i assume it is because you are in a high overlap mode for non lift and need to change that when you engage a different cam lobe because the crank speed and torque hasnt caught up to teh volume of air and fuel being detonated

later
 

·
Premium Member
Joined
·
18,200 Posts
Discussion Starter · #15 ·
SlamminS85 said:
you shouldnt carry over the exact same VVT position before and after lift.
This was obvious before I started the thread, and has been long understood. Not to bag on you or anything, but what I was concerned with was practical limitations of the VVT mechanism itself and the impacts that would have on power. The idea was that these limitations might cause a map to be written in such a way that isn't completely intuitive.

Back to the point, I have verified exactly what Torqued suggested: the VVT must move to full advance BEFORE lift transition in order to eliminate any bogging. However, my results were very interesting as far as effect on power. Now don't apply any of these below numbers directly. Not only is a turbo VVT map completely different than an N/A one, but my switch point is not even close to optimized.

Breakpoints: 5250, 5700, 5900, 6350

Condition 1: VVT: 5700 = 55, 5900 = 35, VVL point = 5800.
This was the worst, causing a noticeable misfire and stumble right at the VVL switch point. I was thinking it was a lavle overlap issue, so I had continued to retard the cam on both sides of the switch. This was also causing the cam to be moving during the switch.

Condition 2: VVT: 5700 = 50, 5900 = 50, VVL point = 5800.
This was noticeably better, keeping the cam retarded through the lift point. The major stumble and misfire went away, but there was still a noticeable dip in power. Cam help still through lift point.

Condition 3: VVT: 5700 = 50, 5900 = 16, VVL point = 5900.
Problem solved. The key was to get the cam advanced before the lift transition occurred. This would explain why people initially thought that it was impossible to make power by changing the lift transition. When working off of 500 rpm breakpoints, its impossible to have the VVT set properly at the right place to make this work. In order to tune a car properly, you would first have to determine the optimum VVL change point, which on an N/A car I've heard alternately as 5500 and 5600 rpm. You would then need to pick several breakpoints in your map to get the VVT system to make the most power at all points. You would want to hold the cam retarded as long as possible (this breakpoint would vary depending on which gear you're trying to optimize for), then have the cam at a "lift" VVT seting either right before (as Torqued suggested) or right on the VVL change point. The location of the next highest breakpoint is debatable, although Torqued has more tuning experience and may be locating it just post-lift for fueling and timing reasons.
 

·
Registered
Joined
·
419 Posts
doesnt advancing your cam right before lift cause your engine to bog before lift instead of right as lift hits?
 

·
Registered
Joined
·
3,732 Posts
Advancing VVT right before lift and then retard significantly has prooven to be effective up here in canada too ...
 

·
Premium Member
Joined
·
18,200 Posts
Discussion Starter · #20 ·
jlitman said:
danGTS said:
Advancing VVT right before lift and then retard significantly has prooven to be effective up here in canada too ...
Something like this?
Not exactly, that's a CAMCON graph. It is only showing advancing the cam relative to the stock position. We're talking about absolute cam position.
 
1 - 20 of 38 Posts
Top