Reducing internal EGR and throttling loss with VVT
 

Hey all,

I've been playing around with the VVT settings for a week or so, mostly to get rid of some internal EGR effect and try to squeeze a little fuel economy out.

Looking at the stock maps, there's a massive amount of overlap at low load which sends a bunch of exhaust into the intake to reduce throttling losses and emissions. This is great for lowering emissions, but it makes combustion less stable and raises the charge temperature considerably, so I tried to see if I could reduce throttling losses even more while reducing charge temperature at the same time.

Throttling losses can be reduced with late intake valve closure, and less overlap gives you less EGR effect. I logged a bunch of parameters but the things to pay attention to are MAP, load, and VVT angle. Higher MAP = less volumetric efficiency and lower pumping loss, approximately. Only doing later intake VVT will somewhat reduce the EGR effect since that's partially driven by trapping exhaust gases in the cylinder and partially from exhaust backpressure blowing out the intake valve.

This is the intake VVT map I'm using:
http://i60.tinypic.com/2uopie9.jpg
(-10 in the 800, 0.15 cell was because I was trying to see if it would improve idle smoothness, which I think it did but it's hard to say lol)

I tried removing 10 degrees from the exhaust VVT map but manifold vacuum actually increased over stock (MAP was lower), so I set it back to the stock map, and got a slight improvement overall. Tuned is just the intake map changed, stock exhaust map:
http://i62.tinypic.com/110wtc1.jpg

The data points are ~60mph cruise, 2600 +/- 10 rpm. You can see the MAP tends to be about 0.5psi higher with the intake at -10 deg instead of 0.

0.5psi * 1L/rev * 2600rpm =~150W in throttling losses saved, which is not a whole lot of power but charge temperature should be down a little increasing efficiency a tiny bit as well. It takes around 8.5kW to do 60mph so this probably increases fuel efficiency by 1.5-2%.

If you drive through hilly areas where you'll hit 0.7-0.8g/rev loads going uphill for a significant amount of time, I imagine taking 10 degrees off in those cells would give you a teensy bit more efficiency.

A note on higher load cells, the OFT tune has the 800 and 1000rpm high load cells set with a ton of intake cam advance which I'm sure gives more torque, but I decided to not use those maps because it's probably best to limit torque at low rpm where the bearings aren't getting much oil.

I'm fairly sure that the AVCS tables aren't in operation at idle.

Good work though!

You are correct, I just looked at my logs and they say 0 deg at idle not -10 haha.

I recall seeing that the AVCS has 55 degrees of adjustment, but I think you can't go more than -10. Since the max value in the table is 37 I wonder why they wasted that extra bit of adjustment range. Makes me want to degree the cams 10 degrees from stock, you would have just as much power after adding 10 to the high load cells but the engine would be a little bit smoother and efficient at low load, although since the cams would have to move 50 degrees rather than 40 to reach maximum torque output at some speeds it would have slightly slower response.

This is a bit counter productive since EGR reduces pumping losses, increases fuel efficiency, and lowers EGT's...

Decrease in vacuum = decrease in pumping = increase in fuel economy.

Less EGR = better combustion.

EGR is using inert air in the combustion chambers to allow less fuel to be burned without it being lean (as there's less O2). It's cooler combustion and helps resist knock.


I don't have more time to get involved in this tonight, but lower MAP would mean more EGR effect theoretically.

That's correct, but the purpose of EGR is primarily to reduce pumping losses, and that means increased MAP. The end goal is increased MAP, you can get there with more EGR or more LIVC, and my logs suggest LIVC is more effective. It would be better to have a longer duration intake cam which allows both maximum EGR and very late intake valve closure, but we have just the stock cams to work with. LIVC is better than internal EGR in the sense that the mix isn't being diluted so there's no problems with unstable combustion.

I flashed a map with -15 deg on the intake and logs suggest the AVCS is cooperating with me. Romraider lets me input -40 but I doubt it will actually go that far, I'm going to try -20 next but I don't know if the engine will actually do it.

Interesting discussion. Without actually being able to tell the mixes/pressures/temps before/in/after the combustion chamber, we are making simple educated guesses on a complex combustion chamber system based on simple assumptions. I don't have any real answers.

EGR is inert air but it is hot, inert air; potentially at quite high pressure if there is a restrictor like a turbo in the way. EGR may cool combustion in some situations and make it hotter and more unstable in others.

Except we do have a pressure sensor in the manifold telling us how the pumping losses are on the throttle side (which is by far the greatest proportion of pumping power). Basic principle of VVT is retard the intake to cut volumetric efficiency at part load.

EGR will always cool combustion as less fuel(energy) is used. EGR's effect is never great enough to cause combustion to be unstable, at least on a properly working system. On old cars with EGR valves, if they got stuck open or you activated them to test their function at idle, the car would stumble or stall. At idle is about the only time it's going to be unstable as there simply isn't a whole lot of fuel and air entering the engine anyway.


Tuning for a richer condition (a classic Toyota tactic in years past) also cools the charge, as does direct injection, though this is getting off point :bonk:

That's only true for factory settings. They add the maximum amount of EGR they can before combustion stability falls below some threshold.

As an experiment I added 5 degrees exhaust retard to the low load cells yesterday and revving it in neutral at 2000rpm produced stutters and pops from the exhaust. D4-S engines are designed to have a maximum of 21% charge dilution with EGR or so (you can see this in the 2GR-FSE development paper), vs. ~10% on a port injected engine.

By the way, the intake cam's most retarded position is -25 deg. I put -40 into the tables and it only gave me -25.

Right now I'm running -25 with 5 degrees extra on the exhaust in the 0.3 - 0.5 g/rev cells. Gonna try 2 degrees extra in the 0.15 and 0.2 cells maybe just for the hell of it, it probably won't really save me much gas lol.

SHEAR US YOU TABLES

all what i know is lowering the VE as mach as we can to set ECO-FUEL SYSTEM but we lose the power

Well you don't lose power, because the cam is returned to an advanced position at high load to restore VE. If efficiency is actually being increased, then you get a little more power at low load because for the same airflow there's less pumping loss.

I'm still trying to get more data on these, I'll share once I'm totally happy with a map. I drove a long distance yesterday and noticed the mpg gauge was reading a little lower (I highly doubt there was a headwind the whole time lol), so perhaps even that 5 degrees exhaust timing was too much.