Divided turbine housing spool-up idea[discussion]

[Mr Peepers]

I've been kicking this idea around for a little while now. I've seen people block off one side of a twin scroll turbine for spool up, and then go one step farther a put a butterfly valve on one side; opening at a certain boost level or rpm. None of these retain the benefits of the divided housing such as keeping the exhaust pulses from cross contaminating each other, and the exhaust slams into a closed plate when you're trying to spool the turbo.

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

The basic idea is that you have all the exhaust go through one side of the divided turbine(stock Holset turbine housing etc), in effect cutting the a/r in half, and generating much more velocity to get the turbo up to speed. Once you reach a certain boost level, the plates will become vertical, and flow just as any twin scroll setup would. The plates would be controlled by something like a wastegate actuator and the cracking pressure would be controlled by a MBC. The plates would move at a 1:1 ratio in respect to each other with rod connecting them, similar to a locomotive wheel or primaries/secondaries on a carb.

With this setup, there is decent flow when you have one side blocked off for spool, and very little obstruction at full boost/high rpm when you want all the top end you can get. The best part would be keeping the exhaust divided when it's fully open, something I haven't seen for this type of idea.

What I'm worried about is the temperature of the plates. They will be directly in the flow of the exhaust without much to dissipate the heat, and molten metal/chips etc aren't good for turbine wheels. SS 321?

Any input?

 

[aero_sallee]

CHeck out the "MST" turbine housing that ATS has on their site.

Not sure if this will take you to it or not.
MST

 

[Mr Peepers]

CHeck out the "MST" turbine housing that ATS has on their site.

Not sure if this will take you to it or not.
MST

Yup, sure did. Nice find, I've been having trouble finding anything on the subject.

It's hard to tell if that retains the divided exhaust flow when fully open with only 1 valve plate; either way I'd like to hear some real world results

 

[turboking15]

you would want the butterflies to open gradually though instead of a sudden open like a wastgate other wise you will loose exhaust pressure when it opens hurting the spool. with a gradual open you will get a more linear power curve.

i have no proof that this would happen, but it seems like it would. just trying to throw out ideas for you to work with. but this is a very interesting design, i would like to see what you come up with.

 

[Mr Peepers]

you would want the butterflies to open gradually though instead of a sudden open like a wastgate other wise you will loose exhaust pressure when it opens hurting the spool. with a gradual open you will get a more linear power curve.

i have no proof that this would happen, but it seems like it would. just trying to throw out ideas for you to work with. but this is a very interesting design, i would like to see what you come up with.

If the butterflies opened suddenly and affected boost, they would immediately close to regain spool since they are controlled by boost, similar to a wastegate setup. This would keep happening (most likely very suttle changes though) until you were able to maintain the same boost with the butterflies fully vertical. It would be a feedback type system.

I wouldn't be trying this for a LONG time if I were. Hell, I haven't even gotten a good start on my rebuild because of the cold weather/snow If I were, it would be for an HX-35 or similar.

 

[ceedawg]

CHeck out the "MST" turbine housing that ATS has on their site.

Not sure if this will take you to it or not.
MST

Man that's a great find bro With this turbo,one would not even need a twin scroll divided mani!

 

[Steve93Talon]

The Acura RDX works similarly to this.

 

[trick engine]

I was thinking about this as well . I was looking to try it on my new GT42R set up at some point . But I just want to get the 42R in for know to see where the spool will be.

I found these pics on another board and they got some ideas going.

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

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

 

[Mr Peepers]

^That's one of the first ideas I had on doing it, and I like the simplicity, but I just don't know how much the "off-center" valve will be affected by exhaust gases. Enough to overpower the wga trying to flip it back up? Who really knows...

Intriguing to find out how well/if any of these ideas work.

 

[trick engine]

Yea that is a good point . I guess some testing would have to be done to see what happens.
But with your idea like you said having the vavle in direct flow may cause issues . I would hate to see a turbo get killed during testing .

I dont think the exhaust flow would pull the valve down when trying to pull it to center . I would think once it craked open it would move easier at that point cause air would move around it and help push it to center ?

 

[Mr Peepers]

It would start to move easier once it's initially "lifted" from its seat with that setup, but it wouldn't be in a linear fashion. I just like the idea of no forces being imparted onto the butterfly valves and them opening when and how I want them to

 

[DSMunknown]

Matt, have you also considered the effect of the extra heat on "half" of the turbine blades? What about the extra force that will be applied onto the end of the turbine blades (assuming that the dividers will be directed onto the outer section) and how that relates to the balance of the shaft/spline and the compressor blade wheel (as well as the turbo bearings)?

Honestly, I don't think what I've just mentioned will be much of anything to worry about. I'm just throwing thoughts around.

 

[Mr Peepers]

The plan was to direct the exhaust towards the volute closest to the center section so it would be forced to travel through the entire turbine wheel, hopefully aiding in spool up even more. I honestly didn't think about the balance or heat issues, and although I agree they may not be much of a problem, it's always wise to think about everything that goes on to troubleshoot in advance.

The ideas are always appreciated

 

[DSMunknown]

Don't forget - the turbine housing "generally" absorbs a couple hundred degrees of heat when the exhaust gases are rushing through it. Now in this example you have proposed this heat will be concentrated onto part of the turbine housing (instead of being shared throughout the entire housing). See if this will have an effect on your example.

EDIT: You responded when I was still tying and somewhat answered my question. Good.

 

[BeatupVR4]

Already been done boys. VATN (Variable Area Turbine Nozzle) technology had been around for a while. I'm sure it's used on other OEM vehicles but I know for a fact the newer 7.3 and all 6.0 Ford Turbo diesels are equipped with such a turbo. Though it's oil pressure opperated and not vacuum/boost accuated.

 

[Mr Peepers]

VGT turbos(VTG, VNT, VATN etc) are more at home on lower EGT diesel engines. They tried a VGT, t25 actually, on the '89 Shelby CSX-VNT but had problems with it sticking after carbon build up and overboosting since VNT acted as it's boost control.

IIRC the most recent 911 has it. Very expensive materials/process to get it to work properly with gas engines.

Trust me, I would love to throw a VNT turbo on my DSM

p.s. I love all the 3 letter acronyms

 

[coltboostin]

I have dreamed up an All- DIY, all cheap solution to this. Not as efficient or as fancy as what you have drawn up-but it looks good on paper.


Stay tuned...

 

[ilikespeeding]

Sorry to bring this thread back from the dead, I just happen to notice a few things.

First, is heat. You're going to have to make this out of something that can withstand 1,400 degrees with only 1/4 inch of thickness. Right after the collector is where it is the hottest. I'm thinking iconel 625. It's acid resistant and can be welded unlike SS 321. If you still like SS...how about 310 to cover if the motor runs lean.

Reliablity. If your right diverter gets stuck in low boost and then the center opens for high boost, two of your cylinders die. I say remove the right divider. You'll still direct 75% of gases to only 25% of the turbo blade. Also, have the 75% be directed on the side of the turbo near the center bearing. There is oil and water there to help cool the housing.

 

[tkelly27]

1/4"!? The middle of the turbine is less than that and it does not melt. I think 1/8th-3/16th would be beefy enough.
http://www.calphad.com/graphs/Metastable Fe-C Phase Diagram.gif
Seems to me if you are melting steel, your tune is off. 900C is the target, right? My internet isn't working all that well. I dunno, I think you could get away with mild steel for a prototype. For corrosion purposes you'd probably want a stainless type of material.

Lol, TLAs (three letter acronyms) are your thing eh?

I think sealing the shaft would be a PITA.

The side flapper is the simplest idea. I think having it on a cam style system with the WGA would also be very simple. High torque to pop it open, and the rest of the cam would cause it to have a long throw to open it up the rest of the way. Like a throttle body. You know this, and you reject it constantly! : D

Umm, umm, something else... Oh yeah! It's already been shown on another forum that running a divided housing with a block off plate is fine for the turbo. A few people have been doing it with divided turbine housings that are grossly oversized and having success for a good long while now. Extra damage to the turbo has not shown up. That doesn't mean it can't, but I think most of the people reading this thread go through engines and turbos so often it won't matter.

This also sounds like a job for the $150 blue light ebay special turbo.

 

[andy4g63]

Besides Shep's car, I've seen a couple more turbos, one of them in person with Tial style flange on the turbine. Especially the one I saw with my eyes was from a PEUGEOT rally car. Magnesium housing/turbine/ extremely light.Anyway if the divided would matter so much, even with the butterfly it should've been there.Personally I think this thing/divided / is great but a little overrated. The new 911 turbo has a Variable turbine geometry and there are Variable Vane turbos, both of these works kind of the same way like the one posted in the beginning of this thread. You can get them, but they are a little pricey. The turbo on the RDX as Steve 93 mentioned is kind of the same as the one on 911, besides size and manufacturer. BTW 911 turbo gets around 450lbs/ft of torque @ 1950 RPM.
I would love this too.

 

[Mr Peepers]

The thing that really sparked my interest was if it would work or not. After reading some more about divided turbine housings and that certain exhaust pulses only see one side of the volute anyway, I'm having doubts at how effective this would be.

Is a single volute block-off plate as effective in spool up as utilizing the twin scroll design? It seems like most people who use a block off plate were previously using a standard(non-divided) collector, which would severely hurt spool since the two volutes combined are larger than a conventional scroll.

Ilikespeeding: The plates would be connected with a linkage so if one got stuck the other would not move.

I won't be trying this anytime soon anyway...

 

[tkelly27]

Is a single volute block-off plate as effective in spool up as utilizing the twin scroll design? It seems like most people who use a block off plate were previously using a standard(non-divided) collector, which would severely hurt spool since the two volutes combined are larger than a conventional scroll.

As far as I know, this idea all started on theturboforums.com

Faster spool and more boost

They were using a twin scroll with a single collector manifold and found it worked really well. Then someone was like "Butterfly valve!" and someone else was like "OMFG, NO, with a WG" which I think is kind of dumb, but it looks like they are doing it.

They each have their own individual AR's that combine. Two AR .5's create one AR of 1. I think that's the way it goes, feel free to correct me.

Magnesium turbines?

"Incendiary use: Magnesium is flammable, burning at a temperature of approximately 2500 K (2200 °C, 4000 °F), and the autoignition temperature of magnesium is approximately 744 K (473 °C, 883 °F) in air."

Maybe an alloy? Or inconel? I don't think I'd even use straight magnesium for the compressor, if the blades got a little friction under full boost you could catch the whole car on fire!

 

[dsm-onster]

Is a single volute block-off plate as effective in spool up as utilizing the twin scroll design? It seems like most people who use a block off plate were previously using a standard(non-divided) collector, which would severely hurt spool since the two volutes combined are larger than a conventional scroll.

Yep, exhaust gases are pulsed. The volute hones/directs the individual pulses. One pulse going through one half of the twn scroll housing has a TINY volute directing it. Lots of work is exerted on the blades. Luckily there's plenty of time before the next pulse because the number of pulses entering that half of the housing is also halfed. So there's good flow.

Not to mention the ability to exhaust a cylinder without contaminating another during blowdown/overlap of 2-1 or 3-4. TS systems allow the engine to actually benefit from valve overlap like n/a. It's like dual exhaust for your turbo car, but with only one pipe out the back

 

[turbosax2]

Anybody thought of using one of these?

http://www.spracingonline.com/store/Sound_Performance_Quick_Spool_Valve/3643

 

[aero_sallee]

Anybody thought of using one of these?

http://www.spracingonline.com/store/Sound_Performance_Quick_Spool_Valve/3643

Did'nt know anyone was making one. Pretty expensive for as simple as it is. IMO