Aerocharger

[Prigor]

Does anyone have any experince with the Aerocharger 53 series turbocharger? We are choosing a turbocharger for our single-cylinder 450cc engine and this one looks the most appealing since it is self contained and also has variable vanes. I am wondering if it really is as good as it looks on paper or does it have any hidden issues? Are there any problems with the self contained lubrication system because the throttle body is on the inlet side?

I was also looking at Garrett GT12 and IHI RHB31. Garrett is a good alternative because there are carbon face seal kits available. We don't have the time to develop and test new seals for our application so the IHI turbocharger is probably out. Are there any other turbochargers that we haven't consider and would be fit for a small engine, keeping oil leakage in mind?

[Warpspeed]

Originally posted by Prigor:
Does it have any hidden issues?

I would advise very strongly against wasting your time and resources with a variable vane turbocharger on a throttled petrol engine.
But they do work fantastically well on an unthrottled diesel which has some very different mass flow characteristics.

It sounds great, close the vanes right up and get full boost pressure at very low rpm. But to do that (at low mass flow) the exhaust back pressure must rise to SEVERAL TIMES the boost pressure, and a petrol engine will not be happy with that situation. The exhaust turbine effectively blocks flow and will cause massive reversion.

All the big multinational car manufacturers have tried to make it work and eventually given up. Honda made it work quite well on a Formula One engine, but failed to get it to work on a road car. The huge resources of Honda engineering plus a vast fortune were spent trying. It must have been devastating for the engineers involved to eventually admit defeat.

Garrett make both the VNT turbo for diesels, and the GT ball bearing turbo for petrol engines.
They manufacture millions of turbos of both types.
Not even Garret could get a variable vane turbo to work anywhere nearly as well on a petrol engine as the plain fixed housing GT ball bearing type.

It is why they still make both types. If the variable vane turbo was superior in all applications, the GT series of turbos would have become obsolete long ago.

Drag racers, hot rodders, and some of the finest well funded power train engineers in the world have all been trying to do this for over thirty years and given up. People are still trying...

Only Porsche have had some success by using VNT turbos combined with normal wastegates and a VERY complex control algorithm.

I have personally tried it and given up.
Don't waste your time dreaming about the theoretical advantages of a variable vane turbo unless you are planning a turbodiesel project.

[Prigor]

Thanks for your input Tony!

I've read that only one FASE team had success with a VVT turbo. Wollongong used a GT15V and had some pretty good torque and power figures. They did also use an external wastegate. Maybe that's the right way to go. I'll probably get the GT12 turbo and use that but if I'll be able to get an Aerocharger at a reasonable price I will definitely try it on a dyno.

[RenM]

The problem with VNTs is the maximum exhaust temperature, which is lower because of the additional moving parts. Porsche is the only manufacturer who uses VNTs because they can pay for gasoline VNTs, which need more expensive materials to withstand the higher exhaust temperatures.
Most VNTs are rated for 850° C whereas most Wastegate Turbos are rated for 1050° C.

Of course your exhaust back pressure will be higher in lower revs compared to a wastegate charger, if you are going for maximum boost pressure, but thats because your work output coefficient and your boost pressure is higher. With the same boost pressure your back pressure will be lower with a vnt because the efficiency based on the complete exhaust mass flow is higher.

Drag racers and hot rodders have very different needs for a turbocharger compared to FSAE. In FSAE you need a fast throttle response and a good low end torque. Both things a VNT is rather good at.

The problem with the aerocharger is the vane actuation. They use a spring actuated by the boost pressure, and thus try to achieve a constant boost pressure. But the lack of control is a no go for me. Plus they can´t provide any turbine maps, so you cant see if their turbine is any good.

[Warpspeed]

I ran a VNT for two almost two years on a road car, and reliability of the vane mechanism was never an issue. The vanes are made from a similar alloy to the turbine wheel, and I had absolutely no problem with vane jamming or thermal damage.

Exhaust gas (and exhaust deposits) can and do leak into the vane control ring, and vane toggle area, and I can see how a very high mileage VNT might get jammed vanes that way, but those vanes are also in almost constant motion, which tends to keep them operating freely.

The biggest problem I had was when flooring the accelerator at low rpm, the vanes would close up and create a huge exhaust restriction.
The engine would hesitate, die, then rapidly accelerate. Top end power was also well down.

The reason for the lack of top end power is that all the gas is forced to pass through the turbine. With a wastegate, perhaps half the exhaust volume bypasses the turbine altogether at the top end.

Opening the vanes right up only slows the turbine down, it does absolutely nothing to increase the total available turbine flow area.

Now you could fit a much larger turbine, but that really defeats the whole purpose of the exercise.
So in summary, I had very poor low end response and very poor top end power, which made the whole thing a frustrating waste of time.

Porsche use wastegates, and that should certainly recover all the top end power, but you still need a very well thought out control strategy for the vanes to make it all work.

In the end I replaced the VNT with a GT ball bearing turbo, and that was vastly better in every single respect.

All I can do is pass on my own hard won personal and very bitter experience.
Others that have actually had hands on VNT experience will have similar horror stories.
And I might suggest that building an FSAE car has enough problems and complexity without going anywhere near this particular can of worms.

Just trying to save some of you a mountain of grief and disappointment.

[RenM]

It sounds like your engine wasnt well tuned. The problem with the VNT is that you can not just simply close the vanes to get the best transient behaivour like with a wastegate, because the possible turbine expansion ratio can be very high even at low mass flow rates.
It seems that this was the problem with your turbocharger.

A VNT has a high potential but will require a lot of knowledge and dyno time. Its not a bolt on and work perfect system. Knowing and understanding the turbine map is very important. If you look at the GT15 map for example you can see, that you dont get the lowest expansion ratio if you have the vanes in the open position but if they are 3/5 open. Thats because you dont only change the flow area of the turbine but also the velocity triangles. You will probably need an external Wastegate for FSAE becauce you will go for very low boost pressure when the restrictor is working.

To get back to Prigors problem:
You are probably right that he should try a normale Wastegate turbo first. If that one is working and he has some experience with turbos he can try a VNT for the next car.

[Warpspeed]

Originally posted by RenM:
It sounds like your engine wasnt well tuned. The problem with the VNT is that you can not just simply close the vanes to get the best transient behaivour like with a wastegate, because the possible turbine expansion ratio can be very high even at low mass flow rates.
It seems that this was the problem with your turbocharger.
Quite likely Ren.

My vane control system was two stage pneumatic, and fairly simple. Any attempt to close the vanes would simply strangle exhaust flow creating massive reversion.
I don't think fuel/ignition tuning can do much when I was seeing 80 psi+ peak exhaust manifold pressures !

You can get this same effect with an ordinary turbo that is much too small, and also with a very small area ratio. Theory suggests that might produce massive low end torque and early spool. All it does is create very high exhaust back pressure and the expected low end torque simply does not eventuate.

No doubt a much more sophisticated and less aggressive vane control system, and better engine tuning would have improved things, but it would IMHO still fall way short of a well matched modern ball bearing turbo.

My experiences were fairly typical of the problems other people have had.

I now believe a wastegate is absolutely vital, but all the people advocating VNT technology never mention that for some reason.
They harp on about the superior efficiency of moving vanes and how it makes the hopelessly energy inefficient wastegate redundant.
But pushing ALL of the peak power exhaust through a relatively small turbine is a disaster, it is not just controlling the vanes that is the problem.

People with far better knowledge, resources, and determination have given up, so I really don't feel too bad about not being able to get it to work myself.

It is certainly all very interesting, and a big project all by itself, which is why I think for FSAE it would just create a whole nest of complex new problems rather than be a solution.

[Warpspeed]

Ah !
Found it.

Here is a press release of Honda's hopes and dreams 23 years ago about a "variable wing" turbo of Honda's own in house design.
That was going to bring the latest state of the art VNT technology to their new top of the range Honda Legend model.
The specifications and actual performance achieved speak for themselves.

It never happened........

http://dwolsten.tripod.com/articles/jan89a.html

[Jon Burford]

VNT aside, you do not need to modify/replace the seals of the GT12 for FSAE use, they are not needed, our car this year is proof of that.
Also they drag a lot and reduce efficiency.

The GT12 is a great turbo and is internally wastgated and fairly light. It's a great turbo to work with (apart from the outlet flange).

Garrett also give a lot of support to FS teams, I would look no further.

[Jay Lawrence]

Prigor,

We had a great deal of success with the GT15V, using a fairly sophisticated control regime involving both the vanes and (external) wastegate. Unless you have a reliable way of measuring and controlling shaft speed, then forget VNT. As for performance, we achieved around 100Nm at about 6500RPM and 72kW at about 7500RPM, with an 80% torque spread from around 5k to 11k. This was with a CBR600 engine, and we never tuned a single with turbo (or at all for that matter).

Stay away from the carbon seals, they are extremely restrictive as Jon said. For oiling issues, just ensure you have good flow back to your sump (we had a suction pump as well).

To start simply, I would buy the VNT turbo and just peg the vanes at a certain spot. Try different spots to see what works at certain RPM and if controlling it is too difficult just peg it to the desired RPM position.

Keep in mind that (as far as I could find) Garrett do not make a ball bearing cartridge for a turbo small enough for our application, so you're stuck with journal bearings.

You should also keep in mind the plethora of requirements to make a turbo FSAE application work. As Tony mentioned, you will have very high back pressures at low RPM in particular. Basically, you pump as much energy out of the exhaust as possible (cam tuning etc), reduce your compression to avoid detonation (run E85 to avoid this), and generally make your wonderful race engine a dog, which the turbo then has to overcome. If done right, it makes for an impressive powerplant. But it's not a plug and play thing.

[Warpspeed]

Jay is definitely on the right track.

Two things to keep in mind.

Without an external wastegate, all the exhaust is forced to pass through the turbine.

And closing the vanes down beyond a minimum effective a/r will kill low end torque from excessive back pressure.

These are two fundamental limitations of variable vane technology that need to be worked around.

[Prigor]

Thank you all for your replies. It has been very helpful. We have decided to go with GT12. A new engine and a turbocharger will be a challenge itself so I'll rather stay clear of a VNT turbo the first year working with them.

I have a question regarding oiling issues. Is it enough to connect the crankcase breather hose upstream of the restrictor to prevent leakage in the compressor or did you make something else?

[Jon Burford]

It will depend on your application, What engine is it?
The oil feed to the turbo is crucial as is the return. Make sure your return is flowing downhill and if it all possible pump the oil back.
We use a free flowing return with no problems though. If a downhill route is not possible. You will either need to attach it to a dry sump pump or an electric scavenge pump.
We played around for a long time. it's a case of trial and error in my experience.