Practical VVTi tuning questions

Hi,

Im tuning a 2jzgte vvti engine on an M150. I was just curious if others are getting similar results as I am when it comes to the drivability areas of the VE map.

Do you always want to adjust the cam angle to achieve the highest VE value? The advance during cruise is meant for better fuel economy so it has a negative effect on the VE numbers.

This is a performance street Supra so Im not ultra concerned with fuel economy but Im wondering if having a major dip in the VE table is a typical result, also wondering if I can improve the drivability by removing the advance at cruise so the VE table doesnt have as much of a drastic change

pictures attached of my VE table and Cam Advance table.

There's more than one way to do this, but here are my recommendations.

First, recognize the complexity of tuning VVT on speed density. In that era (late 90s early 2000s), Toyota was using intake VVT with a MAF sensor on a lot of engines (1ZZ in the early 2000s for example). A MAF sensor lets you get away with a lot. VVT adds a lot of complexity. Hondas of the same era had different VE maps for each cam position. Modern engines with intake and exhaust VVT are even more complex; look at a Ford Ecoboost engine for example. They have over 10 VE maps to account for the different possible intake and exhaust VVT positions. This is primarily a fuel economy thing.

Second, remember that as you increase the overlap (due to more intake cam advance) in non boosted conditions, you partially "dethrottle" the engine. What that means is, you get more residual gas in the combustion chamber, and you have to open the throttle more to achieve the load. This is where the fuel economy benefit primarily comes from. So as you add the VVT advance you are physically changing the VE and this is important-- the MAP shifts as the VVT angle changes. You can find yourself chasing your tail if you're not careful. That's why the OEM speed density + VVT systems are so complex.

Here's what the OEM's do when they tune VVT speed density during development. They zero out the VVT tables first. They tune the VE and spark tables on the whole engine with 0 VVT advance. Then they retune it for each VVT position. I'm not saying you do that level of work. I'm saying, this is how these systems were designed to be tuned on speed density if you are trying to do part load fuel economy and keep driveability and such.

Here's what I recommend you do as a practical matter:

Zero out the entire VVT and get the VE table tuned at part load with 0 cam advance. Make sure the VE table is safe in boost as well, although it doesn't have to be perfectly optimized. It won't boost as well without any VVT. Save that tune and keep it in your back pocket so you always have something you can go back to.

Then add VVT starting at 125kPa MAP and up and tune the boost area. Get the VE, boost control, and spark dialed in. Add VVT at part load ONLY at the very end if you are looking for a project. At each speed and non boosted load you get different internal EGR rates, which slows down the burn, and can make for unstable or slow combustion. You can actually hurt fuel economy by adding VVT at part load and not accounting for the slower burn with more spark advance. This spark advance effect is totally non linear and time consuming to tune. It's a LOT of work for tiny benefit on a project car.

There are two effects you need to keep in mind from adding the intake VVT. The first is the effective compression ratio change, and the second is the scavenging effect. The effective compression ratio increases when you advance the intake cam and close the intake valve near BDC. See the attached image. This is optimal at low speed, where you want to trap as much air as possible as long as it's not knocking.

The second effect is the scavenging effect. Adding intake VVT when you are in boost will help spool the turbo more because you will have more overlap due to earlier intake valve opening. The pressure at the intake valve is higher than the pressure at the exhaust valve, and fresh air blows through the combustion chamber and spools the turbo better. The practical impact of this is a possible change in boost control behavior.

With regards to your actual VVT map in boost (skip VVT in vacuum for now), you will want to test to see what the best map is for your hardware . I suspect that your map would work better if it is shifted to the left. You may want to try 35 degrees at 1500 rpm, tapering down to 0 degress at 4000rpm. Your maps is keeping intake VTC into much higher rpm. It depends on the grind of the cam, but that is likely going to hurt your actual breathing.

These older engines have a more advanced centerline than newer intake + exhaust VVT engines. Usually newer engines with intake and exhaust VVT tend to keep intake VVT to higher rpms like you are doing now, due to the intake VVT having a 50 or even 70 degree range. In the end though you really will have to test for yourself.

The attached pics explain the principles of overlap for fuel economy, scavenging in boost, and the actual valve timing. Intake VVT.png is a somewhat generic illustration of cam positions when intake VVT is at max. Intake VVT fuel economy.png illustrates the principle of intake VVT for fuel economy. Scavenging.jpg illustrates the principle of using VVT to spool the turbo through scavenging in boost.

Thanks for all the insight!!! This actually explains why I’m getting oil in my intake manifold! Too much overlap due to too much advance at low rpm and load.

You gave me a lot of info to consider

Bob

@DentGuy - There are several webinars in the archive that deal with VVT optimisation which will be helpful to you. With a simple single cam VVT system where you're only controlling the intake cam you'll find that you don't want anywhere near as much advance as you currently have at light load. It's typical that your cam advance in the cruise areas (2000-3000 rpm and 40-80 kPa) are going to want to be closer to 10-15 deg.

https://www.hpacademy.com/previous-webinars/113-dual-vvt-tuning-strategies-motec-m1-2/

https://www.hpacademy.com/previous-webinars/081-vvt-cam-tuning-aem-infinity/

https://www.hpacademy.com/previous-webinars/continuously-variable-cam-control-link-g4-plus/

Just rewatched the linkG4 webinar. Hopefully I can get back on a dyno before wintertime fully kicks in. Thanks again guys!!

Bob

I've heard of an argument for intentionally using 'too much' VVT advance at light load on roadrace cars that will experience lots of on/off and off/on throttle transitions. For these situations, I've heard tuners suggest to use a 1D-looking VVT map similar to your original one but without the zeros in the 20kPa row. The justification was setting the VVT to the ideal full-throttle position for all loads eliminates inconsistencies you might experience due to VVT response time. Your application is obviously different, but I thought it was an interesting concept.

The VVT response time was good enough for the OEM to meet emissions and driveability, at least on a stock car. You have to tune the gains though, and VVT can be sensitive to oil temperature and weight.

You're right that if you don't vary the VVT with load the transient behavior is less of a thing to deal with. However now you have more internal EGR and potentially need to run richer to stabilize combustion and prevent surging, or you need to spend a lot of time tuning spark to see if you can account for the slower burn.

@Scott, you're right, this is a sensible approach if you're using VVT on a race car as the VVT tracking on constant throttle changes can cause inaccuracies although some systems are better than others. I always try to keep relatively smooth steps between individual cells for this reason but when you transition from maybe 30 kPa to 100 kPa at 3000 rpm on a quick WOT application you're inevitably going to see a large change between the optimal VVT points. I think for a road car the negatives of reduced fuel efficiency at cruise would potentially be more significant than the effects of momentary inaccuracies. Always a good idea to log the VVT position vs target and see how the system is actually performing though.

I've just tuned a 3.4l VVTi MK4 Supra on a Syvecs ECU last week. Try to target way less advance in the off boost and cruise rpm areas.

Depending on the turbo and cam combination you'll also find, that with higher RPM and higher boost levels you also want to taper down the cam advance.

For reference, between 30 to 36PSI and above 6500rpm, I was only targeting around 4-10 degrees while slowly tapering down to zero till 8500rpm.

It's always best to tune this on a dyno, as this will give you the torque feedback your looking for.

Sorry to revive an old thread, just got my VVTi Supra back from the tuner. Only able to get 25 degrees of advance at 2500rpm and nothing more below that, everything below 2500 rpm is zeroed out. Tuner states there isn't enough oil pressure to advance the cam past that (25 degrees) below 2500rpm. Car was tuned on a brake dyno, could this have affected anything? From what I've read, VVTi cars should be tuned on a load bearing dyno. With that being said, can the car be tuned on the street?

Unless you're having problems with oil pressure in general, you should be able to achieve the full range of VVTi movement from pretty close to idle. Certainly you should be able to achieve this well below 2500 rpm. Basically with the VVT system you can either achieve full cam travel or you can't. In order to optimise the cam timing you need to use a dyno with a load cell that can actually give you torque feedback based on the changes you're making. It's not possible to properly tune VVT on the road.

Are you sure you'd say it's not possible Andre? I mean obviuosly the load cell on the dyno is going to give you 100% the correct results, but on the road you can still get a pretty decent result by looking at the AFR as you advance or retard the cam. Leaning of the AFR would indicate increase in engine VE which is in essense what we are trying to achieve by varying the cam position. Sure it's not going to be 100% accurate but it ought to still get you in the ballpark vs just guessing numbers.

@chris1388 - yes, the AFR will 'guide' you in the correct direction, however often you'll see very subtle changes in the AFR, particularly in the off boost regions of the map which make it difficult to be accurate. If you have no access to a dyno then you're obviously limited to what you can do and in this case looking at your AFR is almost certainly better than nothing. Just understand that you'll still be leaving some potential on the table when compared to mapping the VVT on a proper load bearing dyno.

Thanks for the responses. Being as VVTi tuning is fairly new here in the states, what am I looking for when tuning on a load bearing dyno (just found one in my area) so that I can pass it on to the tuner. Checked oil pressure and it's spot on. What could possibly be causing the issue? Tuner said that cam wouldn't advance below 2500. I know nothing about tuning and VE, but is there something I can do to check and see if it's possible without drastically messing up of the VE of the engine?

Put it on the dyno in 4th gear and load it by engine rpm. Start logging. Go to 2500 rpm. Set entire VVT target angle table to 25 degrees. Adjust dyno to 2400, 2300, etc in 100 rpm steps until you see the VVT stop working. At each rpm sweep the VVT duty cycle to check VVT operation. Bonus if you have an electric oil pressure sensor that you can datalog. Also make sure the oil is up to temperature.

4th gear or 1:1 ratio gear? What am I looking for as far as AFR at partial and WOT? Talked with a few people and they are clueless on this stuff due to the rarity of VVTi and the lack on info online on tuning it. If you could explain a bit more in depth on how to tune and what to look for, that would be awesome. Would have messaged you for this info, but it doesn't appear as if PMs are an option. Thanks!

Watch a few of the webinars on the topic. Andre does a great job of explaining things there.

awesome, thanks!