Alright... so I've had this idea of making a valve that would help a twin scroll turbine spool faster. One guy made one using a butterfly valve design and it spooled 700 rpm faster but I wanted something that would flow much better than that, so I set out to make a sort of swing valve design. Here's the buildup process of what I have so far:

Here's the start of it (my work table is a bit of a mess... I need to quit working on multiple cars at once ):

http://www.machzracing.com/Maxima/QSV/P1020113.JPG

http://www.machzracing.com/Maxima/QSV/P1020114.JPG

http://www.machzracing.com/Maxima/QSV/P1020115.JPG

http://www.machzracing.com/Maxima/QSV/P1020116.JPG

I almost have it finished but messed up a few things (fixable) so I'm probably going to finish it tomorrow, but here is the current:

Welds are decent. I didn't want to get too close to the side so that it didn't weld the pieces together.
http://www.machzracing.com/Maxima/QSV/P1020155.JPG

Took the welds down a bit to hopefully assist in making it flow a bit smoother (pictures a little blurry)
http://www.machzracing.com/Maxima/QSV/P1020156.JPG
http://www.machzracing.com/Maxima/QSV/P1020157.JPG

Got the other side tacked on
http://www.machzracing.com/Maxima/QSV/P1020159.JPG
http://www.machzracing.com/Maxima/QSV/P1020160.JPG

Here I basically matched one side of the flange to the valve assembly to try keeping flow decent and turbulence down. Its not the straightest match but should work pretty good.
http://www.machzracing.com/Maxima/QSV/P1020161.JPG
http://www.machzracing.com/Maxima/QSV/P1020165.JPG
http://www.machzracing.com/Maxima/QSV/P1020166.JPG

I hit two corners but it shouldn't be too bad. Since I'm not really sure if this will work or not I'm not working too hard to make it absolutely perfect.
http://www.machzracing.com/Maxima/QSV/P1020171.JPG
http://www.machzracing.com/Maxima/QSV/P1020172.JPG


http://www.machzracing.com/Maxima/QSV/P1020174.JPG

Here's what it'll basically look like:
http://www.machzracing.com/Maxima/QSV/P1020175.JPG
http://www.machzracing.com/Maxima/QSV/P1020176.JPG
http://www.machzracing.com/Maxima/QSV/P1020178.JPG

The valve open for top end flow
http://www.machzracing.com/Maxima/QSV/P1020182.JPG

Closed for faster spooling. I still need to add a separator at the top and bottom to keep both passages split, but this is basically what it'll look like.
http://www.machzracing.com/Maxima/QSV/P1020181.JPG

 

Wow... I really like that... how long did it take you to make it? Could you make more possibly? lol

 

This is good stuff. Reminds me of Honda's variable flow turbo on the RDX.

 

I'm not sure exactly on hours... I'd say maybe a little less than a full day of work on it all together. I don't get to go out and work on stuff too often anymore so the times are all spread out.

If it works I'm hoping to be able to make more but I need to invest in some better equipment to really make a lot of these. I'm eyeing a bandsaw as my next purchase, that would make a huge difference in making these things, but I can't decide. I'll let you know if I do end up making some more though.

Yeah, this thing really just turns it into a variable area turbine basically. By closing off half of the turbine you're essentially cutting you're a/r roughly in half. I measured the displacement of the HX35 turbine as well using water to see roughly how it compared to a stock Z31 T3 and an old Air Research T3 I had laying around. The Z31 T3 I believe is a .63 and the AR T3 is something like 1.06 (can't see any numbers on it, but comparing it to ones online, that's what they were).

I can't remember exact numbers off hand but the HX35 displaced only slight more than the Z T3 while half of it was less than the Z T3. The AR T3 was much bigger, I believe at least 150 to 175 milliliters more than the HX35 and easily double that of the Z T3. I have some pictures of the measuring process on my camera if anyone is interested in seeing them. Its kind of funny how I did it.

 

I forgot to mention how exactly I was going to make it work. I'm going to set the valve up so that it is constantly closed. Basically once the turbo spools up close to the pressure I'm aiming at the valve will open for top end flow. I'll change that around a bit to where I have the best spool up with the best flow I can. I'll likely do this by getting the valve to open at different points while on a dyno... hopefully that'll work.

Since it will be constantly closed that'll definitely increase back pressure a little bit so when cruising I'm going to actually have a vacuum signal to the other side of the actuator so that the valve opens under vacuum, much like most external wastegates are designed. That way, when cruising, back pressure will be as low as possible with the valve setup for as much fuel efficiency as I can get.

I didn't get a chance to work on it tonight so hopefully I'll be able to tomorrow... but knowing my luck it'll be the weekend.

 

yea this thing looks great. That sure going to be a fast spooling turbo. Im starting to make boost on mine at 2900 on my dek with out the valve

 

Really innovative. I love it.

So which unit are you planning on using this with? Largest twinscroll I know of is a GT30. With this awesome spool going bigger would be advantageous; no?

 

Holset HX35, which is pretty close to the GT35. It has similar power peaks but the Holset has a much wider map.

It'll definitely help out the mid-range power when going with larger turbos but you are still limited in the size you can go depending on the surge line. I don't believe it will spool fast enough to hit the HX's surge line but I'm not sure.

If I have the compressor map correct and am able to spool it quite fast my limits for instance would be 15 psi at around 2,500-2,600 rpm and 7 psi at about 2,100 rpm. If I could spool the turbo that fast, which I don't think I can but if I could, I could basically set the wastegate to aim for those pressures at those rpm points (and in between) using say an AVC-R and it should ride just to the right of the surge line then keep say 15 psi up until redline. The surge line is pretty linear so I could ride it up to whatever pressure I'm aiming at, though I think once you start aiming for higher pressures like that the turbo should be spooling fast enough by then to really hit it fast. I know some guys running these turbos say that once the thing starts spooling it'll hit 30 psi very fast. You definitely wouldn't be able to use the valve aiming for high pressures like that (keeping it closed) as the exhaust pressures would probably be too high but like I mentioned above, once its spooled up it should hit those higher pressures without trouble.

I don't think I or the majority of the people on this board would be aiming for pressures quite that high though so it shouldn't really matter. I'm not wanting a ton of power but wanted to have decent low and mid range power while retaining a good deal of top end and this should definitely do the trick.

I forgot to post that I have finished the valve. I haven't got pictures of the complete unit yet but will get that asap. The clearances are a little on the tight side so I need to do some modifying to keep smooth operation. Other than that its ready to go. I now just need to figure out where to put the turbo and valve in the engine bay. I have a pretty good amount of room in the engine bay but at the same time it doesn't seem like it.

 

Finally got the pictures...

I still need some more welding practice but the new welder definitely makes a difference over the cheapo welders. Now I just need to get the Maxima running so I can test it and check for leaks in the valve itself.
http://www.machzracing.com/Maxima/QSV/P1020193.JPG

http://www.machzracing.com/Maxima/QSV/P1020192.JPG
http://www.machzracing.com/Maxima/QSV/P1020195.JPG

http://www.machzracing.com/Maxima/QSV/P1020194.JPG

Threaded the holes to make for an easier time bolting it all up. Due to it being square a nut or bolt wouldn't have fit and this looks better anyways.
http://www.machzracing.com/Maxima/QSV/P1020201.JPG

 

looks great! them welds dont look that bad

 

Looks awesome. Welding 1/2" flanges is tough unless you have some pretty beefy welding equipment. What are you using?

So the HX35 is a true twin scroll? This is a huge advatage over the GT35 then. I guess I need to look into the holsets a bit more. I was looking at twin GT28RS's. I wonder if they have anything in that range with a fatter efficiency island.

 

I have a Miller 140. It definitely won't do 1/2 without multiple passes... and even then it probably wouldn't turn out the best. I was thinking about going to their 175 (or 180 now, can't remember), which might pull it off, but for the majority of the stuff I'd be doing the 140 should be more than enough. Plus I didn't want to screw with running 220 wiring across the garage.

They do have a few smaller turbos. HX20, 25, 27, and 30. I actually have an HX25 and its definitely a small turbo but it should support between 140 and 360 hp, which surprised me. The HX30 should support between 180 and 420 hp.

The HX25 and HX30 are on eBay every now and then but I've never seen any 20's or 27's. The 27 has the same compressor stats as the 25 so I'd say its probably just a turbine difference.

As for efficiency, the wheel designs alone should make them more efficient than the Garrett's and such. The HX30 is when they start using the map width enhancement groove which, as the name implies, helps widen the compressor map. I have seen HX30's with twin scroll and single scroll turbine housings. Holset actually markets all of the twin scroll housings as symmetrical twin flow turbine housings.

I also just noticed on their site it looks like the 20-27 range is available with an anti-surge compressor housing as well. I'm not sure on the 30's but if they're available on the smaller range I would think there'd be some 30's out there with them.

Here's some pictures of my HX25:
http://www.machzracing.com/pictures/HX25/

 

Um how would 1.5L spool a HX20 or 25? I think you know where Im going with this

~Alex

 

I've got a Hobart 140 and I have the same troubles. I think stepping to a TIG would make more sense but thats just my thinking.

Thanks for the info about the Holset units. I need to look into that a bit more.

 

I was actually thinking about trying this out and am even eying another 25. Just basing solely on the looks of the unit, the thing should spool pretty quick on 1.5 liters. They're actually designed for diesel engines up to 4 liters so if you base it the same way the HX35's are, they should be good for up to 2 liter gas engines.

TIG would definitely be awesome. I want to get one down the road but... that'll be way down the road.

No problem. There's a lot of info on the HX/HY35's out there but very little for the smaller units. Compressor maps are even harder to find, which really sucks. I'm not even sure why Holset keeps them so secret... its not like their competitors don't have the equipment to map them on their own.

As much as I hate Borg Warner, you may actually look into their turbos as well. They're designed a lot like the Holset turbos, putting a lot of work into the wheel and housing design. They're actually aiming more for the aftermarket with their new turbos though, much like Garrett. I'm still sticking with Holset either way as I dislike BW and I can get Holsets much cheaper.