The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support STM Tuned
Please Support STM Tuned

Billet compressor wheels

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

I'm not so sure.

The 7 blade wheel doesn't surge at all mated with an 05h turbine in an 8cm² housing on my car. I've seen 30psi (Pressure ratio of 3.45 at my elevation) at 3500rpm in 3rd gear briefly with zero hint of surge. Granted, the 7 blade has the advantage of higher blade count (14) over the 11 blade. You can see in the models below the effect that blade count has on slip factor and efficiency (Z= number of blades).

Part of the increase in efficiency means more exhaust is wastegated, which reduces the overall restriction of the smaller wheel. At lower boost, this setup makes the same power figures that my old TD06-20g did, but with significantly faster spool.

There are two things about all these newer wheels that will help prevent surge:
1. Extended tips.
This acts like a wheel with a slightly larger exducer. This reduces the radius ratio (RR) of the wheel, and improves its slip factor:
You must be logged in to view this image or video.

The 20g is already one of the largest RR wheels out there at ~0.77, and it's already operating in the region of the model where efficiency is dropping off quickly. Extended tips move the RR to the left on the model. In fact, I think the 20g wheel probably benefits from extended tip technology more than any other wheel out there, for this very reason.

2. Larger blade angle.
The OEM MHI blade angle is around 28° and these new billet wheels about double that. Smaller blade angle helps generate more velocity off the exducer tips, which translates to pressure in the diffuser and scroll more quickly (i.e. quicker spool), but as the linear speed of the exducer increases, it also results in massive shock losses as the air exits the blades.
You must be logged in to view this image or video.


The larger blade angle doesn't build pressure as quickly (though I believe these wheels make up for it in volume flow per revolution), but it eases the transition off the exducer and into the diffuser, allowing higher wheel speeds before shock losses begin to choke the compressor. When the 20g spools up too fast, the shock losses result in the pressure breaking down, which starts the surging. You'll notice all the other MHI wheels that use the same blade angle have the same rightward dip of the surge line above 2.0PR.

Great information!!!:thumb:
 
The less blades there are, the better a certain compressor performs at high boost (generally). The reason the 11-blade wheel doesn't have staggered minor/major blades is because there's an odd number so they couldn't be spaced evenly.

To go a step further, the MAP EF2 uses a standard-size billet compressor that has 10 total blades- 5 minor, 5 major:

You must be logged in to view this image or video.
 
The less blades there are, the better a certain compressor performs at high boost (generally).

I don't think it would be correct to call that a "general" rule, as it's pretty clear from the models (which Busemann derived by observing and cataloging actual compressor behavior) that more blades are, generally speaking, more efficient. The efficiency gains that come as you add blades decays logarithmically, but the gains pretty clearly exist in the regions we're working with.

I think the trend toward lower blade counts is an attempt to increase the size of the volute's throat for a given wheel size with the intention of getting more total airflow without changing wheel sizes. This does have an efficiency trade-off, in that it will require more wheel speed to get the same fluid work out of the wheel. This introduces other compromises as well in order to minimize the shock losses at the exducer at higher speeds which tends to affect low-speed behavior (spool-up). It's a compromise that really only makes sense once you get near the limits of the higher blade count compressors. From an engineering perspective though, it's probably better practice to move up a wheel size at that point to keep safety factors in check, for reliability reasons.

What I'm surprised at not seeing on these taller wheels is an attempt to soften the blade angle of the inducer's leading edge. The outer tip of the inducer actually has the highest relative velocity of any part of the compressor, and using large turning angles at that point is a significant factor in determining at what speed the compressor chokes. I would have expected to see the taller wheels taking advantage of that added blade face to work that angle better, but they don't seem to be doing much with it.
 
I don't think it would be correct to call that a "general" rule, as it's pretty clear from the models (which Busemann derived by observing and cataloging actual compressor behavior) that more blades are, generally speaking, more efficient.
Are you referencing higher or lower boost levels? Higher blade count may have a broader map, but I look at the T04B V-Trim compressor which was a poor high-boost performer....it had (16) blades total- (8) major, (8) minor.

Also using the HX40 as an example....there are three 60mm wheels offered- a 8-blade (16 total), 7-blade (14 total), and 6-blade (12 total). The 8-blade was originally designed to work better at lower boost levels on non-wastegated constant-RPM engines like generators, compressors, and marine applications while the 7-blade and 6-blade models are found on wastegated HX40's that operate at much higher boost levels.

Not saying you're wrong- just trying to get a better understanding from experience as well as what other manufacturers have done in the past. Fluid dynamics was never my area of expertise. LOL
 
Not saying you're wrong- just trying to get a better understanding from experience as well as what other manufacturers have done in the past. Fluid dynamics was never my area of expertise. LOL

All good examples, thank you. In contrast, most of the EFR compressors are 7 (14) blades, while the Toyota/Hitachi CT26 variants were all 5 blade, similar to the MAP wheels, and were notoriously inefficient at high boost. I think it would be prudent before trying to figure out a 'general rule' to try to examine other aspects of the wheels too in order to determine if there were any other pieces of the wheel that changed along with the blade count. It's possible the wheels designed for lower boost used smaller blade exit angles, etc.

For example, if you look at the differences between centrifugal supercharger wheels and turbocharger wheels, there are some pretty big differences (many of them use lots of blades too) in design due to lower boost, but due mainly to the fact that the wheel isn't designed to see speeds necessary for higher boost. One of the big differences you'll see is that they have zero blade exit angle- the exducer is perpendicular to the axis of rotation. This builds pressure more quickly at low wheel speed, but also chokes at lower speeds. If the higher blade count wheels were designed for lower boost and wheel speeds, there could be bigger compromises built into them. This is just speculation, but it would be worth investigating if we wanted to come to a general conclusion about blade count.

I'm also really only looking at slip factors at this point, but slip factor is essentially the indicator of the wheel's ability to turn linear speed into fluid work. There are obviously other aspects of compressor design that could play a part. I do get the feeling that a lot of these new billet wheels are designed by trial and error and not necessarily by following the engineering principles behind centrifugal pump design, at least in the case of the non-OEM suppliers. There are probably exceptions to this too, but I doubt "snail turbo" is one of them.

The KTS spec sheet I downloaded a while back even lists their new 20g billet wheels as MHI style (6-blade), EFR style (7-blade) and GTX style (11-blade), though they do seem to all use significantly larger blade exit angle than their original counterparts. The point is that I'm not sure whether these were engineered by themselves or just adapted from designs that were already shown to work fairly well. Either way, at least the 7 blade has shown to be able to spool fast without surging AND make decent power. The 601whp it made on that evo suggests the ability to flow at least 55lb/min, taking into considering the thermal efficiency of the high compression motor it was on.

Now we're just waiting on some results from the 11-blade. That will actually be a good comparison for blade count, since the rest of the wheel's aero is pretty much identical. The trick will be getting enough usable data, rather than just comparing power numbers.

I'm also beginning to believe that the weight of the wheel really has little to do with spool-up, considering the weight differences in some of these wheels account for 3-4% of the total weight of the rotating assembly and the fact that most of that weight is close to the axis of rotation (the weight saved has very little rotational inertia anyway). For reference, a TD05h turbine shaft weighs 183g, and a TD06 shaft weighs 268g.
 
Last edited by a moderator:
I'm totally nerding out on all this tech info, please keep posting! It's almost impossible to find info about how blade count, angle, depth, etc effects compressor efficiency and such (at least when related directly to turbo charging), so your posts are both enlightening and refreshing!
 
Well i had my car out a couple of times with the new kts 6 blade compressor. It is a little peppier at low rpm and i think a tad more power up top. Makes 2 more psi on the spring, 27 vs 25.

The logs are showing 50-100 hz more airflow and 2-3 % more duty cycle from my 27 psi dyno run. The dyno run was in 15-20* warmer but dryer air.

I can't turn the boost up because i am out of injector. Never had the kinugawa 20g over 27 either.

Keep the technical stuff coming guys.
 
Last edited by a moderator:
Fail on my part, but a well deserved bump for this amazing thread!
 
:DWell I recieved my kinugawa wheel.

Weight Comparison

My Td05h: 196 Grams
Kinugawa 9 Blade Td06sl2: 185 Grams

You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


Notice the exducer is taller on the Td06sl2? I have an extended tip turbine wheel!
 
Last edited:
I have an exteded tip turbine wheel!

People keep misusing that term.

Extended tips refer to the tips of the exducer on the compressor wheel. The term does not refer to the height of the blades.

This is a picture of the extended tips of a KTS wheel:
You must be logged in to view this image or video.



Interesting that the kinugawa wheel weighs the same as my TD05h (Mine was 183g). The taller blades will help recover some of the loss of efficiency from the lower blade count. I wonder what the blade area ratio is between the two? (EDIT: did a quick and dirty and unscientific analysis by selecting a blade on each in paint.net, blade area looks almost identical to the TD05h). Thank you for posting this.
 
I wasnt being serious obviously cause I was refering to my turbine wheel. I wil correct myself....


I have an extended exducer blade turbine wheel.

Also I must have a pretty heavy td05h wheel if yours only weighs 183 grams!
 
:DWell I recieved my kinugawa wheel.

Weight Comparison

My Td05h: 196 Grams
Kinugawa 9 Blade Td06sl2: 185 Grams
Some weights of stuff i have tried. I am liking that 185 g # :cool:

The mhi 06 is about 268 grams
The kingawa 05h about 202
The mhi 05 about 198
The mhi 05h 15* clipped about 192
 
I wasnt being serious obviously cause I was refering to my turbine wheel. I wil correct myself....


I have an extended exducer blade turbine wheel.

Also I must have a pretty heavy td05h wheel if yours only weighs 183 grams!

Weird. I weighed the 05h shaft TurboLab sent me, which was 183g, and an MHI 05h from a 14b, which was 184g.

The Kinugawa TD06SL2 is 225g. My TD06 was also 268g, as sqr said above.
 
I'm running my 11 blade at 26 psi right now. I've never experienced surge before so I don't think I know what it feels like. But at 26psi it runs nice and smooth. I am running an anti surge (holset style) housing and this wheel is paired with a td05h turbine in a 7cm housing.


hey i saw u changed from the billet 6 blade to the billet 11 blade. how do u like it? any difference between those two? im about to get a either a billet 7 blade or billet 11 blade from turbolab but wanted someones imput with experience in them. what u notice as far top end and spool?
 
hey i saw u changed from the billet 6 blade to the billet 11 blade. how do u like it? any difference between those two? im about to get a either a billet 7 blade or billet 11 blade from turbolab but wanted someones imput with experience in them. what u notice as far top end and spool?

I never put the other billet wheel on. I went straight from an 18g to the 11 blade 20g.

as far as the turbo goes, there is a little more lag then the 18g. But still hits hard, just a tad later. I will have specifics with a link log posted most likely next week.

I don't know how much top end I can have with stock cams, but it definitely pulls to redline. I catch myself wanting to keep the pedal down because its still pulling heh.

So far I love it. Sounds awesome in the anti-surge housing too.
 
Update time! I just got back from an Evo VS Sti dyno day and decided to throw my car on to see where this thing is at.

I put down 374/369 on the billet 11 blade 20G, TD05H/7cm. Boost was at 25psi on 91 octane.

Boost does look more laggy then I thought from the dynograph though. I didn't have my laptop with me so I don't have the log unfortunately.

Will post the graph and video in the dyno section soon.
 
I put down 374/369 on the billet 11 blade 20G, TD05H/7cm. Boost was at 25psi on 91 octane.

What is your actual elevation? Google places Layton, UT between 4300 and 4800 ft.

It's pretty obvious that the people in your other thread trying to compare to a 16g at sea level at the same boost don't understand how altitude affects turbochargers.

That's actually one of the reasons the standard 20g wheel was pretty worthless at high altitude- the higher pressure ratio for any given boost level meant that it would start surging at about 16psi on a 2L motor.
 
It is indeed 4800ft. I'm right by the mountain. People definitely have a harder time making power here unless running a 62mm or bigger turbo.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top