What's the 'disadvantage' of a hybrid turbo?

[cheekychimp]

I've been doing a 'lot' of reading recently on the performance of different turbos. It's funny how as you begin to understand and the penny drops, that you begin to crave more information.

It has started to dawn on me that the term 'hybrid' is a bit of a misnomer. People (well turbo manufacturers to be precise) have been experimenting with different turbine and compressor wheel sizes since turbos were invented, in an attempt to produce the perfect blower for different engines and applications. It seems a 'hybrid' is little more than a shortcut in that it simply uses two commercially available housings that can be bolted together rather than redesigning the whole turbo.

I'm just wondering now, where the point of diminishing returns comes. I can see that if you have two sized wheels the smaller being X and the larger being Y, the concept is that a hybrid X/Y turbo should spool as fast as an X/X configuration and flow as much as a Y/Y. Obviously you don't get something for nothing, so I'd imagine, all things being equal that the X/Y would spool faster than a Y/Y but slower than an X/X and flow more than the X/X but less than the Y/Y.

There do however appear to be situations where a Y/Y configuration actually outperforms an X/Y in terms of both flow and spool up.

Is that just down to better tuning or could a particular setup actually spool up faster with a larger inducer?

I realize housing design has a huge amount of influence here, but assuming that the housing remains the same, is there any reason a larger inducer would spool faster than a smaller one on the same setup?

Paul.

 

[PJGross]

I don't think there's a straight answer to your question. I've tried to "read" my way through the science of turbines and compressors and how they match up, but haven't been very satisfied with any conclusions. (this is after learning how to size compressor for industrial applications) You can read some corky bell and others, but the secrets will remain secrets from a design and application standpoint and are tweaked by running a bunch of combos on a particular setup and settling on the one that meets the most performance criteria.

There are so many variables even after choosing the turbine and compressor wheels, like which housings on hot and cold sides, porting, not to mention header design, etc., that I think there is only one way to select the appropriate turbo for your application:

Write down what you want performance wise out of your setup, and then duplicate a combination that has proven itself to meet your performance needs. Current turbo offerings are pretty well dialed in and you might get a little more of a uniquely designed setup by stating your exact use and expectations to any number of tuners/turbo builders, and letting someone pick the best combo for you. You still might end up tweaking your setup after tuning and running on your setup as every setup will behave a little bit different... especially yours since you've got a wide variety of one-off parts on there not to mention some interesting driveline bits and a weight that I'm waiting anxiously to see where is settles in at.... /ubbthreads/images/graemlins/smile.gif

-PJ

 

[jepherz]

I don't know much about the topic either, but I don't think it should be considered that a hybrid turbo is just a cheating way to get around a bigger problem. Keep in mind that most of the turbos out there are designed for use in industrial applications. I don't think something like a gt42 was ever meant to be run in a 2.0 Litre Mitsu, etc. I guess the best you can do is try to find compressor maps (yes, it's sometimes hard!) for your application. A larger hotside is going to increase your max flow, but hurt your spool a bit as the wheels are heavier. A larger coldside is going to also push more up top, but take more to get it going, delaying that point in which it really kicks in.

However, I personally would be more comfortable running a turbo that came from the manufacturer (mitsu, garrett, etc.) in it's intended configuration, rather than trusting something that a smaller company to do a bunch of machining etc. Once again, that's just my opinion.

 

[turbowop]

When I think of "hybrid" turbos, I think of turbos like the FPgreen or FPred, which use Mitsu and Garrett parts. I don't think mixing and matching different wheels and housings from Garrett makes them hybrid, it just allows you to choose the best combo for whatever given setup.

My definition of hybrid turbos came along so that people could bolt Garrett parts onto a factory Mitsu exhaust manifold. Basically for simplicity. That said, there are big quality differences between all the different manufacturers of "hybrid" turbos that you have to watch out for. You don't really have to worry about that when using a straight up Garrett or Mitsu turbo.

 

[jepherz]

I'm not sure if that's what is meant by a 'hybrid turbo'. I think the FP green and red may even use their intended Garrett turbine wheel, correct? If so, I wouldn't even consider them to be hybrids. There are a lot of people running combos of the same brand hot and cold side. Something like a 42r cold side and a 40r hot side would be a better example of a hybrid turbo, I think.... /ubbthreads/images/graemlins/dunno.gif /ubbthreads/images/graemlins/dunno.gif

 

[Polish]

Red/Greens use TD06h Mitsubishi Turbine wheels.

 

[Hksvr4]

I think FP is considered to be real "hybrid" turbos. An example is the FP3065. In order for FP to make this turbo, they used two garrett turbos. The gt30r exahust wheel and a gt35 compressor wheel. They have to machine the 65lbs/min wheel to a fit the gt30 shaft. Then they put their FP exahust housing on them that changes the a/r on the exhaust side. Also the FP3575, has a gt35 exhaust side and a gt40 compressor side. They managed to put a T4 wheel on a T3 shaft! All FP turbos uses a 360 screw down thrust bearing. They are the only ones to do that. Of course there turbos are more money than any other brands. Companies like PTE, AGP, or Slowboy, they just change the exhaust housing ( majority of their turbos) to fit a mitsu. flange. In essence, it's still a straight garrett or turbonetics.

 

[cheekychimp]

Thanks for the input here guys;

Quote:
Write down what you want performance wise out of your setup, and then duplicate a combination that has proven itself to meet your performance needs.



The thing is, this was sort of what I did when choosing the Holset but I got flamed for that choice quite a bit. I really wanted a very fast spooling turbo that would give full boost at around stock and my experience on the stock engine was boost coming in around 3500 rpms.

I was aware of other 'tried and tested' turbos but I wasn't aiming primarily at track or 1/4 mile times. I had actually heard rumours that the 'problem' with the Bullseye Housing was that it was too small and choked which stopped guys getting good power and mph figures but was responsible for crazy spool up.

Quote:
A larger hotside is going to increase your max flow, but hurt your spool a bit as the wheels are heavier. A larger coldside is going to also push more up top, but take more to get it going, delaying that point in which it really kicks in.



This also seems relevant. I am wondering if the larger coldside is allowing more flow on the full HX-40 in the bullseye housing which is resulting in the higher trap times. I feel this is relevant as I'm beginning to think that even if the housing is flawed that the straight HX-40 configuration still works better in that housing than the hybrid. What doesn't add up here however is that spool up on the HX-40 is faster even in the same housing as the hybrid HX-35/40. This basically makes me feel the guy running the hybrid had tuning issues.

As a comparison, the straight HX-40 was run in a Bullseye Housing on a completely stock motor (head and bottom end) with an AEM. Car was full weight with stock suspension and stock intake, but had intercooler and fuelling mods including alky injection. The car ran an 11.53 @ 125.4 on 27-28 psi. Full spool came in at 4200 rpms.

In comparison the HX-35 was run on a built bottom end with ported head, SMIM and cams, stock IC with lousy piping (lot of 1" and 2" sections largest was 2.25"), and no fuel mods (ran a Keydiver chip with reduced timing to allow 25 psi on 93 pump). He ran an 11.68 @ 119, but couldn't get full boost until 4500 rpms.

Quote:
I don't think mixing and matching different wheels and housings from Garrett makes them hybrid, it just allows you to choose the best combo for whatever given setup.



I tend to agree as well Mark. So I did even start lookin at guys that were running Holsets on the Holset housings off a T3 manifold to see if this made a difference. This starts to get complicated. There is 9cm^2 housing which many think would be better suited to our motors but it isn't designed for a split header that separates pulses from the pairs of cylinders. Another 12cm^2 whilst bigger has provision for split headers and would no doubt work well with something like an HKS manifold on a T4 to T3 adapter. Thing is, that neither of these housings appeared to make a great deal of difference to spool. A 9cm^2 produced 20 psi at just under 4,000 rpms on a 2.0 litre whilst a 12cm^2 was almost identical getting 25 psi by 3,800. Most surprising was a guy running a 2.4 litre 9.1:0 compression who made 25 psi at 3,500 rpms on a straight HX-35 in the bullseye housing.

So I'm confused. It is undeniable that there are far too many variables to make any decisive conclusions other than 11 second timeslips appear attainable with this turbo which really was what I was hoping for. If I can get 25-28 psi in under 4,000 rpms I'll be happy even if that means I make 80 hp - 100 hp less on the top end. 25 psi by 3,500 rpms on a 500 hp engine would be insane !!!

I want to try this out and see what happens. I have quite a few options available in terms of configurations if I pick up some different wheels and it would be interesting to see how the Holset Housing would change things but I have my doubts if changing the housing will have as significant an impact as some believe. I personally am beginning to think the 35/40 matching my be poor and that a 40/40 match would be better.

Again there though are too many variables. HX-40 wheels came in a number of different sizes with 6,7 and 8 blade versions. I guess at the end of the day by taking the hard route I will undoubtedly learn more and if it all fails I can always come back and try what has been 'tried and tested'. I just think there is more potential here than has yet been realized.

 

[jepherz]

Quote:
Red/Greens use TD06h Mitsubishi Turbine wheels.



My bad, I was thinking it was just a custom turbine housing.

 

[Polish]

No biggie, just figured I would clarify. /ubbthreads/images/graemlins/cool.gif

 

[PJGross]

Since you clarified your question to Holset setups only, this limits the discussion quite a bit. I think variables already mentioned above by various still apply, but you are working at a disadvantage since you are using technology from a totally different power application strategy. You are piecing together truck parts and trying to optimize them for your application. Nothing wrong with that, but you don't have much to work with from a theoretical design standpoint. The parts are limited enough that you could quite possibly mix and match and drive and tune until you're happy.

I think you could take your goals to someone like FP and they could set you up with a much more optimized complete turbo and give you a wider, greater area torque curve for your setup in your intended rpm range than something you piece together from a Holset setup.

I really don't understand your situation. People run Holsets because they're cheap and can roughly match the needs of certain performance applications. Also because they're cheap. In fact, many people run them because they're cheap. I fail to see how this fits with the rest of your buildup. (Did I mention they're cheap?) I would think you would spend your time talking with HKS or FP or company X and have them custom spec, build, balance a turbo that matches your intake, cams, displacement, exhaust manny choice, operating rpms, use. Not only that, but would better match the quality of buildup you are in the middle of putting together.

-PJ

 

[Hksvr4]

From my personal experience, the turbo should be the very last thing you'll buy.

 

[cheekychimp]

Quote:
I think you could take your goals to someone like FP and they could set you up with a much more optimized complete turbo and give you a wider, greater area torque curve for your setup in your intended rpm range than something you piece together from a Holset setup.

I really don't understand your situation.



I was likely misinformed. When I asked, no-one was able to give me the name of a turbo that could get more than 400 whp on pump and produce 20-25 psi between 3000 and 3,500 rpms. Quite frankly I think I would have been happy with a 16G and a good tune but I was told the "G" was too small for a 2.3 litre and that 400 whp was it's absolute max. Guys are running HX-35s and getting spool up under 3800 rpms and irresepective of the so called design flaw of the bolt on housing seem to have more room up top than the 16G.

I think people 'don't understand where I'm coming from' because the vast majority of people on here compete on 1/4 tracks where power is clearly important. I really think given all the money I've spent on the drivetrain and handling mods, I should be aiming at 350-450 in terms of power which is why I think my mistake is not necessarily going with a Holset but going with the larger 40 option instead of the straight 35.

I realize that everyone on here thinks I should have got an FP or a PTE or Garrett or whatever. Who knows maybe someday I will, but right now I'm more interested in trying to understand how some people are getting such good results bolting on these turbos that use "technology from a totally different power application strategy". The reason being that I'm sure what I learn taking the harder route will pay dividends later if I ever custom spec a turbo. FWIW I do honestly believe that NOTHING and I mean NOTHING off the shelf is ever going to match an application perfectly. Custom specs is where it is at.

One thing I do kind of have difficulty understanding here though is peoples unwillingness to discuss the issue as soon as the name 'Holset' comes up. It's like someone making a post on here asking about good piston, rod, intake and cam combinations to rebuild a motor and being told "You'd be better off just asking Buschur or Magnus to build it for you".

If it sounds like I'm having a go, I'm sorry but I'm not convinced that bolting a GT35R or an FP3065 on this car is going to magically pull in an 11 sec timeslip ... look how long it took experienced members like Mark to do that. So to just bolt an 'unapproved' turbo onto a completely stock car with some engine management and pull a mid 11 1/4 mile is in my opinion "extraordinary". And if the turbo itself is not 'optimised' for that application then there must be something else right somewhere. THAT "something else" is what I'm looking for.

Maybe it's the huge timing advances used or the application of the injection system and where it is injected, maybe it's the oil feed restriction mods that they do. I don't know. But one thing I do know is that getting a car past 12s and into 10s is very, very hard work, and I'm not convinced that 'any' one turbo out there is going to make that easy.

Paul.

 

[PJGross]

I hope you didn't take my post as: grab anything off the shelf. I was pretty clear that for your unique setup you should be getting something custom built, as you agreed. I was trying to get across in my first post that matching turbos and all of their variables to even a known engine setup is a difficult process. Even after reading Corky Bell and others along with getting all of the maps, its difficult to predict what a combo XYZ will do without bolting it in and driving it on the specific car/engine that you are planning it for. I think the reason you're not getting the response you think you should get is that for the most part people on this board are intelligent enough to know that they really don't know how the heck to answer your question. I for one love this board because of that, and fall into that category. However, I think maybe you might not need to dig that far into the situation, either.

No Holset basher, here, either. I think reliable, affordable performance is great! I dread the day my wife's twin K03 turbos go out b/c there is no easy drop in replacement for her car even though certain companies have been promising a garrett solution for years now.... Anyway, I was considering a switch to a Holset myself or sourcing a couple of them for running twins on my Impala. But you ARE stuck with what you can get from them to a certain extent, which is what I was pointing out. And there are definitely compromises to be had with them, too.

I personally have a circa 2000 FP off-the-shelf T3/TO4E 57 trim, .63 A/R turbine housing garrett on my Galant. It was $650 shipped and at the time thought I would give it a whirl. My goals were a weekend driver/road course machine, no drag racing. Due to a lot of circumstances, my car is just getting reliably finished but in the 500 or so miles I've put on it, I can say that with 9:1 JEs and 272 cams that I think it would be very well suited for daily driving and track use. Severely untuned w/ keydiver chip, MAFT, SAFC, and 660 densos, I get full boost by about 4600 rpm or so. From all the talk I had heard, I thought I'd be ordering up a .48 A/R housing immediately to lower spool RPM. However, after driving it a bit, I just don't see the need at this point. Unless you have a severe hairpin to deal with, I have no issue with going a gear lower and using the 5000-7000 rpm range when in "spirited" driving mode. Though I have pulled through 8000 and it pulls like a fiend up there. I think I might try a .48 A/R housing on it at some point and see about autocrossing, but for now I just need to get it on the road.

I guess my question to you is, why do you need spool at such a low rpm? Is it personal preference or what others are telling you? I'm fine with using a higher rpm "window" of operation, but others may not be. Unless you know that's where you want to be, I wouldn't be too concerned about where you make power, just adjust your driving style to fit. I also have a 3000 rpm 9.5" converter on my Impala, which even though it is mild for most setups, would drive a lot of people crazy as a daily driver.

-PJ

 

[Hksvr4]

The GT35R and FP3065 are proven time after time to be good and effecient turbos. They are exactly the same turbo just the FP spools quicker but makes the same power. These turbos has been proven to be great turbos with in the DSM community.
Quote:
guess my question to you is, why do you need spool at such a low rpm


For autocross or road racing.

 

The fp3065 and gt35r are not the same turbo.
They do both use the 56trim compressor wheel (aka GT40)
but the GT35r use a gt35 turbine wheel and the 3065 uses a gt30 turbine wheel.

 

[Hksvr4]

Quote:
The fp3065 and gt35r are not the same turbo.
They do both use the 56trim compressor wheel (aka GT40)
but the GT35r use a gt35 turbine wheel and the 3065 uses a gt30 turbine wheel.



I said that in my previous post.

 

[cheekychimp]

Quote:
Even after reading Corky Bell and others along with getting all of the maps, its difficult to predict what a combo XYZ will do without bolting it in and driving it on the specific car/engine that you are planning it for.



Quote:
From my personal experience, the turbo should be the very last thing you'll buy.



Now I feel like the discussion is going somewhere. It would be nice if I was in the States. I would have bought a used 16G and 'driven' across America to go see someone at FP. I'd have left the car there and let them R & D the perfect blower. Unfortunately I live in Hong Kong and even given their extensive knowledge I'm not sure that it is possible for them to pick the perfect match for my car having never seen it. The cheapness of the Holset stuff allows me to play around a fair bit with affordable components before I start looking at a $3000 custom turbo. It also allows me to play with stuff like anti-lag and pre compressor water injection without having to be worried unnecessarily about unwittingly causing any damage to the wheels.

It makes more sense to 'experiment' with something cheap and renowned for durability during the tuning phases. In particular it will allow me to swap housings as well as wheels fairly cheaply. The Bullseye housing is a .49AR and that apparently was verified by FP themselves.

I am unsure about whether I need to spool that low, but my V6 VR4 is twin turbo and starts building boost right on 3000 rpms and I just love that low down torque. It can be adjusted for in driving style true but I use that area of power SO MUCH MORE than the top end in Hong Kong that to me it really affects driveability.

I realize that there are a lot of variables but I really only had one main question which I would have thought was answerable.

If we assume that the turbine and compressor housings remain the same and that the turbo is used on exactly the same car/engine at exactly the same boost pressure, is there any situation where making the compressor or turbine wheels larger would make a turbo spool faster?

I know flow will be affected as will power but I'm trying to understand if a straight HX-40 could ever outspool an HX-35 or a hybrid HX-35/40 in the same housing.

 

[turbowop]

Quote:
I'm sorry but I'm not convinced that bolting a GT35R or an FP3065 on this car is going to magically pull in an 11 sec timeslip ... look how long it took experienced members like Mark to do that.



While I'm flattered that you consider me "experienced", I'm modest enough to come out and say that I don't know sh*t compared to a lot of other guys on this board, which is most likely the reason it's taken me so long to get to where I have with whatever setup I have at this time. While my car isn't solely set up for drag racing either, which hurts my times, my experience isn't *that* extensive. I'm just one dude that's learned how to work on my car through internet forums and help from local friends who share the same passion.

I guess what I'm saying is...you can't really bolt up *any* turbo and magically pull in an 11sec timeslip, but the turbos mentioned will certainly help. /ubbthreads/images/graemlins/smile.gif

 

[cheekychimp]

Quote:
While I'm flattered that you consider me "experienced", I'm modest enough to come out and say that I don't know sh*t compared to a lot of other guys on this board, which is most likely the reason it's taken me so long to get to where I have with whatever setup I have at this time. While my car isn't solely set up for drag racing either, which hurts my times, my experience isn't *that* extensive. I'm just one dude that's learned how to work on my car through internet forums and help from local friends who share the same passion.

I guess what I'm saying is...you can't really bolt up *any* turbo and magically pull in an 11sec timeslip, but the turbos mentioned will certainly help.



Well Mark, at least you can pull your transmission and fix it (pretty well by the sounds of things). I'd struggle to change my oil and plugs myself; seriously!!! This has been one hell of a steep learning curve for me and quite frankly I'm nowhere near the top yet.

If what you say above is true it certainly gives me hope as I feel like I'm struggling around in the dark quite a lot sometimes. You can't learn everything on the internet I guess but unfortunately I don't get much opportunity to hang out with you guys so sometimes the 'search' button really is my best friend. It's hard to always know what to trust and what to ignore. I think you are really lucky to be able to just hang out at meets and see things with your own eyes. Plus it gives you the opportunity to have a talk over a few beers with guys that have tried this and that and can talk from experience.

If you guys really knew me you'd realize I have never been one for taking the easy route, I prefer a challenge. The slightest mention of the fact that something cannot be done drives me to prove that it can.

If I put a turbo on that works well straight out of the box, I'll probably never bother to tune it properly. If it falls on it's face and I'm forced to tune like mad to get half decent performance, I'll probably learn something. In the end that experience will be worth much more to me than the cost of the turbo and will most likely ensure I get thye best out of whatever turbo goes on the car further down the road.