Take it how ever you choose...

Michael Ferrara, I do agree with the conclusion on this as I have that first hand experience between TS and SS. I really don't see how anybody can argue against after experiencing it. It's night and day difference on time to torque at anything below 6000+ RPM. Doing a more elaborate test might change the magnitude of the results slightly, but I feel it's still going to be the TS as a clear winner for anything looking for overall powerband and response. I do think you'd see PEAK HP difference though.


However, I am also of the opinion that the difference in collector is significant IN THEORY. I feel there are a couple concerns:

1. The poor management of cross-sectional area changes when using the open scroll housing on divided scroll manifold. The change at the Vband acts as a boundary to flow and will cause a reflected wave on the SS setup that won't exist on the TS setup, since there is no abrupt area change. You will also get a reflected wave at the 2-1 runner junction. With a 4-1 header, you would only get one wave reflection (neglecting the reflections caused by the turbo) that will be stronger and can be targeted to a specific RPM range in combination with the cam shaft selection.

2. Area contraction then area expansion at the runner merge. This is a throttling process that is simply dumping exhaust energy into entropy. As a fellow mechanical engineer, I'm not too found of unnecessary energy waste, particularly when that energy is EXACTLY what you want hitting the turbine wheel.

3. Flow area restriction of the Vband divided flange. See the video above, but I always FELT this was a big deal. That video though has definitely thrown a wrench in my thought process. However, I think in the case of an N/A header with like 1-7/8" primaries making 550HP on a V8, it's no where near the point that a contraction like that potentially puts the bulk flow well into the compressible flow speed range. A turbo 4 cylinder making 600+ HP through a 1.62" ID primary on the other hand, such a restriction might push the velocities high enough you get some compressible flow losses that could have a negative impact?

Anyway, thanks for the article. Always happy to see tests and comparisons in magazines instead of lifestyle garbage and pretty pictures.

 

Convergent nozzles aren't all that sensitive to rate of cross section change. For example, a simple converging cone can have up to 30* of total angle with virtually the same losses as a 7* cone. Divergent on the other hand is a whole different animal. Anything over 7-8* and you start getting flow separation and turbulence.

The argument I would make is that the SS housing might ultimately end up at the same cross sectional area as the TS at the T-T junction, BUT the SS housing has a couple inches where it's not that small. Simple static flow losses WILL be lower because of this. You also have lower losses with the SS housing simply because you have less boundary region with no internal divider.


ALL of this stuff is all theory though. Sure, on paper it matters, but paper and real world aren't the same thing. Personally, I think this test answers many questions but also creates more, just as any good test should.

The HTA3582R should not be compared to the GTS3576R though. The HTA has 6mm more exducer diameter and with a well optimized wheel, the exducer (along with tip height at the exducer) is USUALLY the limiting factor. Compare it to the HTA3576R or the FP Red. I don't THINK I've seen an Red make 630WHP? It would be interesting to see how they compare in TS with each other though.

 

English Racing built and tuned a 2.3L (100mm stroke 4g63) Evo with a single scroll FP red that made 633whp on C16 with a magnus IM, bushmaster race head, and GSC R2 cams. Aaron said E85 would make more power.


Abacus's 2.2themoon build made 613whp on E85/C16 mix with a 2.2 (94mm stroke 4g63) on E85 with an MHI twin scroll FP red a skunk 2 IM, a pocket ported head, and GSC S2 cams. So "smaller" manifold, "smaller" head, and "smaller" cams, but with E85.


I haven't necessarily been arguing that twin scroll is better or not than single scroll. But using a TS mani on an SS turbo just doesn't seem like it was the greatest decision to be able to present an unflawed test, especially with the TS collector that was used.