Well, you see, its very hard to preform a fair comparison. All of the "tests" I have read about include a complete manifold, turbo, down pipe swap in the test. It is hard to quantify the results being from the TS turbine housing or simply from a better header than the one removed.

When you look at the situation based on logic and science you can tell that a twin scroll can provide better spool up (just like a smaller open scroll). You can also see there would be benefits to running paired cylinders to band-aid a poorly engineered non equal length header or factory style cast manifold.

When you look at the situation based on rumors and "magic"..... Its the greatest thing since sliced bread. LOL

The facts are this: Twin scroll turbos were designed for the diesel industry where getting the turbo to spool up as quickly as possible was optimum. In these large displacement low RPM motors, off idle TQ is what is needed for towing or hauling heavy loads. This is as far from what we want in our small displacement high revving performance cars.

 

Is it really?

Don't both applications aim to spool a relatively large turbo using a relatively small air mass, with relatively long intervals between exhaust pulses? Whether it's a high displacement, low rpm engine trying to spool a turbo, or a small displacement, high rpm engine trying to spool a large turbo, the principle is the same.

The real question is (as far as I am concerned) not if it works (it definitely does), but as the turbine housing A/R increases, if the benefit of the TS design persist beyond the point where the TS housing equals the power of the open housing. FWIW, I've seen more testing to this point indicates that it does than not.

 

How come no one from AMS is responding to Ted B's posts? They seem to be making a lot of sense. I'm no expert but his argument seem logical. Especially the one above.

 

I'm not really expecting responses, simply because the answers haven't all been resolved. I'm just planting a few seeds for thought.

 

9sec240 is a Senior Technician at AMS. He is very knowledgable on the topic at hand.

Eric

 

A properly designed twin scroll setup vs and open volute (with the same vehicle and combination of internals,cams, intercooler etc. apples to apples) will ALWAYS yield more usable power (torque) in small or large displacement engines.
That is evident just be looking at AMS's dyno graph. The restriction (or should I say capacity of the exhaust system as a whole) comes to play when you are trying to process more air from a larger capacity turbocharger. In the end, your goal is to flow as much air as possible thru the engine. If you increase the size of the A/R on the TS unit you are just taking what you have gained down low to compensate for what should have been designed into the kit from the begining.
BUT you have to take into account what the customer wants and what the car will be used for. If he/she wants upper rpm power and a lot of it. An open configuration will yield better results. As far as peak numbers are concerned.
BUT, again it will depend on what the application calls for.
At this point I would tend to agree with AMS as I am basing my thoughts on engine and exhaust theory without much practical info as of yet.
When this debate first started from FullRace I found it interesting and decided to build a setup for R&D. I'll post those results when finished.

 

That seems to be the prevailing assumption for those who have not conducted proper testing, while those who have collectively report differently. And then again, it's another matter to actually test drive the subjects of the test, which exposes other interesting, profound differences.

 

Ted,
What is the "right" type of test that would prove either sides point or "opinion" on the matter?
Like I mention above I would be more than happy to post apples to apples results when I am done with the setup.

 

If I wanted to conduct a test in the simplest possible way, I would, changing only turbine housings, manifolds, and tuning, find the TS housing that matches the peak power potential of a popular open setup, then make my impressions driving both. To some extent, Drifto and I are and have been playing with this - with favorable results. In short, the open setups just feel like mud as compared to the TS.

So far, the only one that has conducted decent scientific testing (testing that eliminates the most prevalent sources of error) is Geoff.

 

I would have to agree with te B in his arguments and questioning...seems 9sec240 is only ever focuses on flat foot, already spooled, drag racing type powerbands; which is not what a TS is designed to benefit. It is designed to benefit turbo response in time (not RPM) as pointed out by Ted B and 9sec 240 in the diesel comparison.
A TS is a great design for people auto-xing and tracking (road course) their cars...it is what I look for as I've been spoiled tracking/auto-xing Formula SAE cars.
I really would like to see a more scientific test done...make it more like a lab report... show various graph's: both power in relation to time and rpm...also pressure in relation to time and rpm. These would show that a TS either does or doesn't truly benefit over a smaller a/r SS turbo and would show what is the better product.
Of course then there are variables that can't be identical such a exhaust manifold, but that is part of the trade off of TS vs SS and what you can get.

 

+100

 

+1

 

read on to my reply to his. then ul get it

 

I know you did not ask me specifically, but I figured I would chime in.

In my opinion, you would need to build two equal length manifolds that were identical to each other with the exception of the collector. One would be a 4 into 1 collector for the open scroll and the other would be a 4 into 2 for the divided scroll.

The reason for the same design manifolds would be to insure that the header is not attributed to any power gains. Obviously a well designed TS header will preform better than a crappy open scroll header.

The reason for the equal length would be to insure that the open scroll would benefit from separate exhaust pulses like the TS artificially creates.

You would use ONE turbo and change only the turbine housings.

You would need to test ALL the turbine housing AR ratios available for both open and divided.

Each test should be done at two boost levels. One level representing average street boost and one level representing "race" boost levels.

Obviously, the same car would need to be used for all testing with nothing but the header and turbine housing being changed.

Testing on the same day with weather conditions being the same all day would be preferred.

Obviously this would be a huge undertaking with a massive investment in money and time. I would PERSONALLY help in any way I can.

 

I really like the method you have proposed...it would be the same way I would test it. The other varaible to the manifold being equal radius for flow velocity loss. That is if you want to keep all things equal but scroll.
For the manifolds, I think packaging will be the limiting issue as to how close you can get the testing to be even.
what different A/R housing are there?
OS:
.63
.82
1.06
1.15 (I believe someone also produces a but may be mixing that up)
TS:
.78
.85
1.06