Hypergear Turbochargers and High flow Services Development thread

[Lithium]

Is that a parallel twin turbo setup on that RB25? Doesn't look like a staged twin turbo setup to me, if it is then it hasn't necessarily be done overly well - as Stao implied, there are single turbos which would provide comparable spool with nicer power delivery and at least as much power.

[hypergear]

Just my opinion that Compound didn't work on paper.

Mainly because of the exhaust side as small turbo's rear need to satisfy the flow of larger turbo's compressor.

I can think A way for this to work, that is making the smaller turbo's rear to have identical flow as the big turbocharger.

For example if the larger turbo is a 3582 then the smaller turbo has to be a 3571 sort of thing.

But if we use above method to cool down the exhaust end, then the compressor end will have issues as the smaller 71mm compressor wheel can not flow as much as the 82mm compressor feeds, its probably going to cause Turbo B to surge.

For 3082's power and stock turbo's response, A sequential system will do. So before 14psi engine pumps gas into a small 2860 lets say, and after 14psi, the 2860 goes to sleep and 3582 wakes up. The 3582 during the time before 14psi needs to be always in a "half sleep" state or its going to create a dip in power curve if its coming from stand still. Just needs to put abit of smart thinking into manifold, dump and discharge pipe. I think this will work in a much better manner.

[GTSBoy]

Stao,

A proper compound turbo system cannot be designed "in your head". The problem is essentially what you fear - trying to balance the flows through the various stages.

But the reality is that given a decent database of turbines and compressors and housings and so on, such as a Garrett or even an engine manufacturer like Caterpillar or Perkins would have, you can determine how much air you need to move and compress through each compressor stage which then tells you how much shaft power you need to drive it and hence what the turbine has to do. You then simply size the turbine sides so that they are each able to provide the required amount of shaft power. The pressure drop across each of the two hot stages is one of the variables that is somewhat freely selectable. If the turbine closest to the engine needs to drive a compressor with a smaller power requirement, then you will set the downstream pressure of that turbine (which is of course the upstream pressure of the further turbine) at a higher level than if you need more shaft power. So you might well end up with strange looking mixtures of really quite large looking turbine housings connected to not very large compressors, or vice versa, depending on how it all works out.

The other way to do it is the way that all the truck racers used to do it, which is to either spend a hell of a lot of time mix and matching, or peek over the garage wall at a successful competitor and copy.

[afb312]

Just look up "mad Mac evo". He has successfully run a compound turbo supercharged car. An evo . Which is a winning machine!

[GTSBoy]

Yeah, but turbo-super is easy by comparison - provided you understand a couple of key points before you start (like not trying to make the turbo responsive and that the two compressors multiply the boost, not add the boost). Compound turbos are nasty to design.

[Lithium]

A guy I know is doing an interesting twin turbo setup with his Supra which sortof fits into this topic - while I am not convinced by his approach I am watching with great interest, this conversation (and the potential outcome) may be interesting to people here:

http://www.supraforums.com/forum/showthread.php?720625-Kevin-s-Compound-Sequential-Twin-Turbo-Design

[hypergear]

Stao,

A proper compound turbo system cannot be designed "in your head". The problem is essentially what you fear - trying to balance the flows through the various stages.

But the reality is that given a decent database of turbines and compressors and housings and so on, such as a Garrett or even an engine manufacturer like Caterpillar or Perkins would have, you can determine how much air you need to move and compress through each compressor stage which then tells you how much shaft power you need to drive it and hence what the turbine has to do. You then simply size the turbine sides so that they are each able to provide the required amount of shaft power. The pressure drop across each of the two hot stages is one of the variables that is somewhat freely selectable. If the turbine closest to the engine needs to drive a compressor with a smaller power requirement, then you will set the downstream pressure of that turbine (which is of course the upstream pressure of the further turbine) at a higher level than if you need more shaft power. So you might well end up with strange looking mixtures of really quite large looking turbine housings connected to not very large compressors, or vice versa, depending on how it all works out.

The other way to do it is the way that all the truck racers used to do it, which is to either spend a hell of a lot of time mix and matching, or peek over the garage wall at a successful competitor and copy.

I don't see shifting shaft speed or what ever happens between the two turbos would be a problem.

The only reason I would considering compound system is gain response while make heaps of power which means reaching the maximum flow capacity of turbo B. By enlarging turbine end of turbo A so to suit turbo B's flow I will lose response. I'm not going to argue the effects of compound system on Petrol engines to a dead end as I've not trailed it my self, but point out areas that could be considered problematic.

On the same manner, the sequential twins are much more manageable, and theory in practice suits petrol engines better. Which It will be the last project to be trailed on my Skyline.

[GTSBoy]

Compound turbo (or any other series turbo system) only makes sense for one reason, and one reason only. Massive boost. A-la diesel truck racing engines. Petrol engines can't take more boost than you can make with a single compression stage, therefore there's very little point in thinking about it (for petrol engines).

[GeorgesR34]

...

Other alternative for huge power with stock response is by using a welly tuned sequential twin system. I do have drawings to make that work and is probably the next thing on the trail list. Its going to get very expansive how ever, and I'm not sure how things might be fitted under the bonnet.

Do you mean similar to whats on the 2JZ Supras?

[hypergear]

Not really. quite opposite actually. They start with one turbocharger and the other turbocharger weaks up later.

I want mine to start up as a twin with only the smaller turbocharger generates boost, when the larger turbocharger in pre-spool (or lag) state. Probably an ATR43SS1 with a ATR43SS5.

So in this setup once certain boost level reaches, take 20psi for instance, which on paper appears to be @ 2800RPMs for the SS1, It will carry the engine out to 4500RPMs in which the larger turbocharger also makes 20psi. At that point the smaller turbocharger goes to sleep. and the bigger turbocharger carries on.

There will be a couple of switches involved, how ever If every thing works out right, I shouldn't have any dips in power curve and it will behave like a tinny turbocharger makes 490rwkws.

[Super Drager]

Sorry but this is incorrect. Have a read of this thread.

http://www.yellowbullet.com/forum/showthread.php?t=216811

Theres a lot more to it than high pressure ratios.

To those who want to see how well a compound turbo kit can work on a petrol engine, go to youtube and lookup streetfighters z31. 145mph trap and full boost well below 4rpm cant be too bad.

Compound turbo (or any other series turbo system) only makes sense for one reason, and one reason only. Massive boost. A-la diesel truck racing engines. Petrol engines can't take more boost than you can make with a single compression stage, therefore there's very little point in thinking about it (for petrol engines).

Edited January 10, 2014 by Super Drager

[hypergear]

This is Important:

I've just had a customer brought a set of Kando's oil feeding line kit in dummying up with his newly high flowed R33 turbocharger.

I've noticed the benjo bolt supplied with the oil feeding line has a 1mm oil restrictor in it. That must be drilled out to at least 3mms.

I definitely recommending of buying our own oil feeding line to work with our turbochargers. How ever if you've got the Kando / Ebay oil line kit, Please make sure all fittings and the hose has a inner diameter of at least 3mms before using them on any of our journal bearing turbochargers.

[Josh080691]

Hey stao i got my lines from u with the newest version highflow are those lines all good?

[hypergear]

All the hoses from My self are restrictor free they are good to use.

[GTScotT]

Stao sent me a banjo bolt to suit and I used my own kando lines (restrictors are in the banjo so staos non restricted banjo fixes that).

Definitely make that standard procedure if it's feasible!

[colourclassic]

Just bought an R34 OP6 turbo to highflow for my R33. Tao, do you include an actuator with your STD high-flow turbos? Or do I need to use the stock one?

[hypergear]

I will need the factory actautor, which I can set it up in a way to hold 20psi. But if you don't have one then we can supply an high pressure actuator for $100 additional with fitting bracket.

[craigy32]

wow this latest R34 high flow looks cool! do you think you would get any gains from putting on different housings?

[hypergear]

Well any thing bigger will need to work with a even larger compressor wheel which will be looking at the SS3 and SS4 series in .82 rear. How ever those wheels are too large for the factory compressor housing so as far as high flow concern the 376rwkws E85 result is about the limit for now.

And here is a quick peak of a brand new ATR28SS1.5 Alpha prototype, which took me a weeks of spare time to complete. Note this is the internally gated version.

It is currently a bolton turbocharger for SR20det motors. After a little bit of driving around it is definitely more responsive then the SS1, even the CHRA configuration is larger.

There are alot of reasons on paper that it will not work properly, and I will end up with very little power plus either a massive boost creep or a massive boost tapper.

And also many reasons it should work as pressure drop by any exist will result a pressure drop off all exist. This is like the argument that we've had earlier with few members on where the external gate should be fit thread, of pressure V flow.

If it does work then by using this method we should be able to knock some lag out of GTR motors that runs on large twins.

Dyno tune tomorrow morning. Hopefully the working theory proves its ground.

[hy_rpm]

Any update yet ?