One of the things that we feel is a common mistake that people do when they are building street cars running on pump gas is to go overboard on turbo selection. We would rather have a responsive 400whp or so with lots of low-end torque than a laggy and soggy 430 hp over a narrow powerband for driving around town. It’s more fun. We should know, we spent our youth driving cars on the street with big turbos with gas that was too poor in quality to use the boost they were capable of.
We happened to have a brand new HKS GTII Turbo laying around the shop for a few years that we sorta had forgotten about. The HKS GTII is a mild upgrade over the stock turbo that was designed to have faster response, less lag and more power than the stock unit. How does this turbo have all of these diametrically opposing qualities?
One of the cool features that the HKS GTII turbo has is a true dual ball bearing center section. The bearing cartridge floats in oil to damp out potentially damaging vibrations while the ball bearings give super low friction for much faster spool, about 10-20% faster all other things being equal. The ball bearings also handle thrust loads better.
The HKS GTII turbo also has a water-cooled center section like the OEM turbo to help prevent coking and after shutdown overheating, a large cause of turbo failure.
The compressor side of the HKS GTII turbo is moderately larger than stock. The compressor wheel is a 49 Trim with a 74 mm major diameter with a 52mm inducer. This is a decent step up from the stock 68mm diameter wheel with its 48.3mm inducer. The compressor wheel lives in a largish .82 A/R housing with a large diffuser.
On the exhaust side of things, the HKS GTII turbine wheel is an 81 Trim with a 60mm inducer and a 54mm exducer, in a .74 A/R twin scroll housing. This is more than a decent upgrade from the stock turbine’s 56mm inducer and 49mm exducer. By keeping the factory twin scroll turbine housing configuration, this should be a very nice moderate turbo with a zingy response.
Strangely by looking at HKS’ Japanese website, it seems as though they had stopped making parts for many popular cars like all EVO’s and most WRX’s! However many places still have this turbo in stock so it is easily available in both new and used configurations.
17 comments
You mentioned that your experience is turbos liking a bigger plenum, but how big do you aim for? Is there such a thing as too big?
A good rule of thumb is 1.5 times the engine displacement. Too big results in poor throttle response and lag.
Great, thanks! The one I’d already planned to use is 1.45 times displacement, so that’s awesome.
Some other stuff if you can package it is to make your manifold walls nonparallel and try to have 4″ from the end of your intake runners to the opposite wall of the manifold. Most of the time you don’t have room to do this well but its good to shoot for it if you can.
That’s awesome info. I’m not sure about the gap from runner to wall because I haven’t started the build yet, but the pictures look like the TB end may be around that.
https://www.semmotorsports.com/parts/18t_transverse_intake_manifold.html
So I’ve heard of big intercoolers and long piping causing a perception of lag as it takes time to pressurize all the volume, despite the turbo already at full chat.
Would a bigger manifold also contribute to that feeling? Granted, the volume seems much smaller than big IC’s and piping.
You are correct. Too big of a plenum would do that. A general rule of thumb is a plenum 1.5 times the displacement of the engine but there is usually not enough room to do even that.
Have heard some say larger throttle bodies might make it less driveable, and harder to modulate power.
Been involved in building a mild 2jz with wide powerband and early spool, we used a lesser quality manifold looking like the hypertune (didn’t know of them then) and a 90mm throttle body, would there be any benefits to going smaller or bigger?
I know with a Turbo engine that having a really big throttle body results in a poor ability to modulate the engine output like there is almost no difference between 1/2 and full throttle. Some of my friends in Pro Drifting have gone to smaller throttle bodies for this reason.
While I no doubt think that JDM turbos (HKS, Trust, Tomei) are of good quality, I cant help get the impression that they’ve been “left behind” by USDM brands, namely by Garrett and Precision, in terms of tech and innovation. Recent JDM turbos just seem to be an “updated” version of pre-existing turbos all the way back as early as the late 90s. And the recent rumor in reddit that Tomei and HKS dont make their turbos in house anymore.
No doubt that an upgrade over stock is always good specially if its just found lying around! but yeah, just cant help get that impression with JDM parts in recent years.
I agree with you. When this turbo came out as it was groundbreaking being ball bearing, OEM fit and reverse rotation. Now the aero is dated but I still think it is one of the best direct drop-in OEM replacement turbos for this car.
I mean, a lot of the older HKS turbos were Garrett CHRAs in custom housings, and on their site they’re outright saying that the GTIII series are Mitsubishi CHRAs though it’s not easily apparent which series. Greddy/Trust and Tomei were always Mitsubishi CHRAs too as far as I know, and again as far as I know are still using some of the pretty old families. In bolt-on Subaru sizes for example, a Trust T518Z is a TD05-18G, and a Tomei M7760 is a TD05-20G. Tried and true, but not too new. But then again as far as I can tell, the McLaren MP4-12C came with TD04s so it’s hard to come down too hard on the TDxx MHI turbos.
Interested to see your results with the large Hypertune intake manifold. The stock Evo IM has proven to be very efficient and most aftermarket manifolds lose tons of mid-range on sub 700whp setups. Ported versions of the stock manifold have supported high power setups.
Looking forward to the HKS turbo and E85 results as I’m planning a similar build.
Mitsubishi doesn’t make ball bearing center sections. It looks Garrett but Garrett severed their relationship with HKS some time ago.
The results I have seen is a slight loss in mid-range. Back in the EVO VIII days we cut off the plenum of the stock manifold and welded on a crude bigger one and saw 40 more hp on the top on a mild bolt on motor with no cams with no other changes.
Your articles are great and the stuff you bring to the community is amazing. I am just confused why you choose some of the mods you do. There is endless R&D by the Evo community already. I guess there is definitely an upside to new content and it will be interesting how that manifold performs, especially with the velocity stacks at each runner.
Funny thing is unless you are looking for over 850HP and by the article talking about mid range power and quick response, you are not, a stock Curt Brown Racing ported IM will provide you the flow and power you want without any downsides to mid range torque / driveability. Curt Brown Racing reported no loss in top end and 50 ft lbs more of torque over the second best tested skunk2. http://www.curtbrownracing.com/manifold-porting.html
Then with the turbo selection, I’ve read tons of articles and it seems the reason people choose FP turbos is because they work well when utilizing stock location. From looking at turbo after turbo for stock location, FP seems to have the best and most reliable. There definitely are better, newer / more advanced turbos but they all require new custom hot parts and LICP. There is also apparently a reason that there are a lot of the HKS GTII Turbo’s happen to be “laying around”. Any extended track use has apparently lead to EGT’s climbing, flapper arms loosening and destroying the turbo, and they’re reported to be difficult if not impossible for servicing / rebuilds. From the way people talk about the HKS GTII on Evom, its like strapping a $2000+ grenade to the engine.
Hopefully you can share some good results though and have good luck tracking the car with the turbo. Any chance you can get numbers with just the IM change then add the turbo after?
A long time ago, I was doing some dyno testing for a now-defunct magazine on 4G63 intakes and the Hypertune came out on top of the 3 I tested which were stock vs modified stock, vs Skunk vs Hypertune, I was going to do more but the magazine went defunct with the publishing group owing me a bunch of money. Me getting pissed at them for stiffing me is a big reason why MotoIQ is around today.
The car was my Blue Evo VIII. Stock turbo and cam. The Hypertune had the most top end sacrificing a little mid-range, the skunk was decent all around and the modified stock manifold was good all around too. The modded stock manifold had the plenum cut off and a larger fabricated plenum welded on with ported runners.
So the Evo IX has variable cam timing, Cosworth M2 cams, and a ton more midrange and bottom end than the VIII did so I think the Hypertune intake might do pretty well. I had the Skunk on there and the results were not as good as the close to stock VIII. I am not sure why the Skunk had so so gains on an engine that is cammed out with a better turbo. I am thinking that the variable intake timing makes the engine less sensitive to intake manifold tuning. That still doesn’t explain why the engine didn’t respond much to a larger plenum volume.
So that’s why I choose the Hypertune, I had it, haven’t run it on an Evo IX, the Skunk didn’t do much on this engine, my modified stock manifold was on the Evo VIII when I sold it and the Hypertune looks cool. I also don’t listen to forum people much because much of the time, I feel that their testing and tuning are flawed.
As for the turbo, message me on FB messenger if you want details why I used this turbo vs FP. The HKS uses am OEM level CHRA and housings cloned off of housings designed for the wheels used. The turbo is reverse rotation so at the time the wheels on the market were all aftermarket.
I doubt this turbo causes EGT’s to climb. That is a fault of the tuning which in the Evo market is all over the place. In any case, the turbo would cause less of an EGT issue that the stock turbo. High EGT’s are caused by compressor to turbine wheel size mismatches, too small of a turbine and a big turbine carved into a way too small exhaust housing being forced to run at high power levels.
One thing about this turbo, it is critical to run water to it as it’s bearing separators are plastic, like Garrett and can get damaged if the CHRA is run dry. That could be another reason why there are failures.