219 replies to this topic

[tmh983]

That has gotta be the single most complicated set of headers i've ever seen! haha nice work.
What sort of butterfly valve are you planning to use as an ecv? I've been thinking about this myself for a while.
You can buy exhuast butterfly's off the shelf, but they are mainly designed for noise control, I'm not sure how they would handle the turbo exhaust temp both in terms of the MUCH higher exhaust temps when on boost, and the constant thermo cycling. I think the off-boost exhaust temps would be somewhere around 300-400 deg, but on boost they could easily go >700. I would imagine this would severly shorten the life of the exhaust butterfly, or possibly make it sieze solid due to thermo expansion...?

[Tim.]

JunBL does some butterfly valves

[acres]

That has gotta be the single most complicated set of headers i've ever seen! haha nice work.
What sort of butterfly valve are you planning to use as an ecv? I've been thinking about this myself for a while.
You can buy exhuast butterfly's off the shelf, but they are mainly designed for noise control, I'm not sure how they would handle the turbo exhaust temp both in terms of the MUCH higher exhaust temps when on boost, and the constant thermo cycling. I think the off-boost exhaust temps would be somewhere around 300-400 deg, but on boost they could easily go >700. I would imagine this would severly shorten the life of the exhaust butterfly, or possibly make it sieze solid due to thermo expansion...?

Thanks, was certainly a pain in the sphincter to make it fit without contacting anything or being too close to any plastic or hoses. Idea is to try and better even out the pulses to aid in transition and top end flow as well as making better use of the crossover pipe to help balance pressure differences between the 2 different sized turbine housings.
I'm going to have a crack at making my own butterfly valve out of some flat steel and a steel rod. I think it was 3mm flat bar that I bought. I'm just using steel on the pipework too so hopefully less chance of seizing from expansion? I'm not really sure though I think it will just be a trial and error thing to see if the valve can hold up.

JunBL does some butterfly valves

I think I may have looked at those once upon a time. I will look into them again though, thanks.

[acres]

So, annual update time. This project was once again placed on the back burner for the better part of the last 12 months however I got motivated again over the Easter break and made a bit of progress.

Made a butterfly valve ECV from a 63mm exhaust brake from an isuzu truck. I cut the bracket off the subaru actuator and made another bracket and welded it to the actuator to suit the new valve.


Tacked together a set of dumps. 63mm off both turbos and expanding to 75mm at the merge.



Was originally planning on running the internal waste gates on both turbos with a divorced setup and separate 50mm pipes from the dumps. But after cutting the flanges to suit and adding the splitters I have now decided to scrap this idea as when test fitting it is too difficult to fit the actuators into the bay without contacting anything. It is also difficult to fit the extra waste gate pipe on the secondary side because of the location of the ecv. I now plan on running an external 38mm turbosmart gate on the primary side only. This should work better anyway.


Bought this front mount a couple of years ago for this project. Dual core and vertical flow to ensure even cooling across the cores. Only had to cut minimal material from the rear lip of the reo bar to make it fit and I should be able to keep the fogs too.

[Hypnotic]

Good to see this is still going.

[acres]

Yeah, I'm pretty keen to finish it this time too.

[Hypnotic]

I would love to see you finish it also, considering the time and effort you have put into it.

[sbv]

Lots of wow! Stuff I've only ever dreamed of!

Well done so far mate!

[Jimbo]

Im trying to get the mechanics around my head.

How will it work with the ECV in the dump?

[Bumpty]

Mmmmm spaghetti..!
(Most intelligent think I could thing of)

[Cam.]

Best talk with your tuner before throwing that on.

Those headers are going to do funny things.

[acres]

Im trying to get the mechanics around my head.

How will it work with the ECV in the dump?

 

It should work better with the ECV in the dump than in the factory spot as there won't be that horrible restriction in front of the turbine entry which would surely cause a lot of turbulence and impede overall flow through the turbo. The factory ECV is also tiny in comparison so would have much lower flow capabilities than the truck valve. The other benefit Im hoping to achieve by using the butterfly valve is to have greater control of prespool. I plan on using solenoids in the black box to hold the wastegate shut until high in the rev range when the secondary is fully online. So while on primary only I plan on using the ecv for boost control. This serves a dual purpose of giving greater exhaust flow to the secondary to help bring it online faster and increases total exhaust flow while running on primary only, thereby reducing backpressure. I believe this will go a long way to removing the VOD. Through some experimentation I have found that the factory ecv is rubbish for this purpose. I found that it would basically be either full open or full closed. I would imagine this is why so many others have had success with the ecv mods involving disengaging the positive line to the ecv actuator and manually adjusting the rod length.

 

In regards to the headers they aren't actually very complicated at all, they just look really messy because of all the pipework that's necessary with the layout of the ej. They are just basically a 2>1 setup to each turbo. However I have just paired opposing cylinders to each turbo instead of adjacent cylinders, like the factory setup has, in order to try and smooth out the pulses to each turbo. I believe this will also aid in transition and total exhaust flow. I moved the crossover pipe to just before the turbine entry, instead of in front of the headers like the factory setup, in order to try and balance out the pressure differences that will be caused from different sized rear housings on the turbos. This position of the crossover pipe also allows room for the 2>1 setup. I am now going to run the external wastegate off this crossover pipe too. 

[Jimbo]

But won't the secondary suffer big turbulence problems with the valve in the dump? Even if its open a crack, all the exhaust gas that cant go through is just going to go back through the turbo, that would make it worse than the factory system. Only way i can see that working was if you had an external wastegate positioned between the valve and the turbo exhaust outlet. Then tuning it so that as load increases the wastegate closes proportionally with the butterfly valve opening, to allow for full flow through. 

It's a big balancing act between not losing response on the primary but then still having the secondary pre-spool enough.

I reckon the best way to smooth out the system would be to use a larger capacity engine such as a 2.5 or 2.7. A larger exhaust gas volume would smooth out the system so much.

[acres]

I see what you are saying but I don't believe that this will be a problem, unless I've misunderstood. The pressure before the turbine should always be higher than the pressure in the dump, even with the ecv not open fully, so flow reversion shouldn't be a problem. This is also where the crossover pipe on the headers comes into play by providing a lower resistance path for the excess exhaust gas to follow back to the primary.

I'm not sure what you mean by the addition of the extra waste gate. Wouldn't that just make it operate like a parallel setup?

Totally agree about the extra capacity, it would make a huge difference. I would actually like to eventually build this motor up, sticking with 2 ltr, to handle even more rpm. That's a long term goal though that fully relies on whether this system will successfully work.

[Jimbo]

Ahkay i think i get it now. If the exhaust pressure post turbine is lesser, then i guess the butterfly valve will handle it okay. My concern was that if the butterfly is fully shut ( provided the setup ever has it at a fully closed state ) then the exhaust gas has no where to go but back the way it came, causing the turbine to stall ( Like flutter with boost but rather with exhaust ). Even though the gas will take the path of least resistance, going through the secondary turbine is still a path of least resistance, till it hits the butterfly valve. 

If you do it this way, i can't see how the setup will be efficient if the butterfly valve is fully shut at any state. It will always have to be open at varying angles, depending on load.

I suppose the standard ECV operates the same way, it slowly creeps open as soon as you hit the go pedal. On my old Gen2 GTB. i teed a boost gauge into the SBV to DPS line to see how soon the ECV starts to open, taking a reading straight from the secondary compressor it showed the ECV began to open at any throttle input. Of course it was closed at idle, and VOD was normal.

Going back to the stock system, like i was saying above, it's a fine balanced that needs to be found. To get a better primary response by having the ECV tighter, you in turn increase VOD. To reduce VOD by loosening the ECV, you'll get better changeover but you sacrifice primary optimization. 

[bmhvanbenthum]

Let me ask a question about the new butterfly valve setup, such exhaust or engine brake valve will not shut completely. I'm also busy with a project. I bought a legacy twin turbo 2001, with some strange malfunction. My exhaust control valve is broken, the axle has some play in the valve holder, so the actuator cannot shut it completely. This is resulting in completely no boost till 5000 rpm. Also see this thread on UKLegacy from a guy having the same trouble: http://www.uklegacy....t-until-6k-rev/

 

I don't know the exact cause. I cannot believe that a small leakage to the secondary turbo charger results in a lack of power of the primary. I think the differential pressure sensor will notice a smaller differential and starts to pre-spool the secondary or maybe opens the intercooler valve so that the pressurized air of the primary flows back through to secondary back to the inlet.

 

I'm removing the engine from my car, since the head gaskets are also blown. I inspected both turbo chargers (VF32/33) and both are fine. No radial play and if you use some force a tiny amount of axial play. So they are in a very well condition. All actuator are working properly, and all hoses are fine, no vacuum leakage.

 

Anyway, if the butterfly valve in the downpipe will not be fully closed how do you prevent strange behaviour of the sequential turbo charging?

 

My goal is to put the twin turbo setup into my LHD 2.0 LX legacy, since I live in the Netherlands and don't like RHD cars.

 

I need to modify the primary up and down pipe to fit the steering rack and axle from the firewall. There is no room for the vacuum brake booster so I will stick with a hydraulic booster which is much smaller.

 

I'm also going to remove the cat in the primary downpipe, and remove the center cat, which I'm going to make swappable for the yearly check.

 

I was thinking to replace my broken ECV also for an engine brake valve, but putting this valve before the secondary turbo. I'm also thinking to re-plumb the primary waste to in between the new butterfly valve and the secondary turbo, since all gasses flowing through the wastegate are lost.

 

Anyway, this topic can be interesting, especially with the work of acres

[Beckers]

Would it just be easier to use a single turbo setup ??

[sbv]

The single turbo option may very well be easier, cheaper and (depending on how serious you get with the single conversion) yield more power...

 

...However, the twin setup (when functioning properly and driven properly) can be quite a lot of fun. The bottom end is great around town and you get a nice kick from the twins when they work in parallel.

 

Having owned and modified both single and twin I can completely understand someone wanting to go to this extent with the twins! The mechanics of the twin setup is pretty interesting.

 

There are pro's and con's for each option, but remember easier and cheaper are not the only factors.

[Beckers]

I agree with everything you have stated above Josh,

but in a LHD it's more work, that was never standard from the factory.

[sbv]

Yeah - of course!

 

Another 'con' for TT and 'pro' for ST