Okay here is a SR20VE dyno plot of twin scroll vs single.
Setup is as follows:
SR20VE. Stock head, stock VET cams. 244* duration 11mm lift turbo cams.
Pauter rods.
8.5:1 JE pistons.
24psi of boost both runs at 4500ft elevation. Only thing that changed was the manifold, 2 wastegates, and housing.

I switched intake cam sooner at 5200rpm instead of 6500rpm, that is the why the powerband is a lot better. Exhaust cam was not switched.


Here is a 3rd gear vs 4th gear. 4th made more power! Great to see the clutch actually holds 4th now! :biggthumpup:


You can see the TS setup spools faster, creates more peak torque and mid-range, and top end is about the same. I attribute the top end holding up so well because of the tubular manifold, the TS just seemed to get better spool over the log manifold. I must say, the TS transient response is much much improved, whether the spool is practically the same or not. When you stab the throttle, the boost needle swings up very quickly. It responds incredibly fast.

I do NOT recommend this small .78ar housing on larger cams. The .78 does not like big cams, because it is a rather restrictive housing. The 1.06 T4 housing will flow decent enough for you to use GSR cams, but type R cams will likely be too large and won't make as much power.

This housing seems to love turbo grind cams.

The top-end torque drop is likely caused by too much pressure in the housing. But it could be the MSD box retarding timing slightly as it comes up to the rev cut of 7300rpm. Hitting fuel cut is not good, so we left the limit on the conservative side. the Rom for this computer has a fuel cut at 7500rpm.

 

What A/R's on the 2 turbo housings? Or did the turbo already have a TS Housing just sitting on a Single Manifold? Was the Single Manny a log or Tubular?

 

The ARs tested were the undivided .63, and the divided .78. The undivided manifold was a log-style. The .78 housing was on a twin scroll tubular manifold.

.63 with log, vs .78 with tubular TS.

 

This motor is basically a B20 vtec. Head flow is like a C5 head on a good one. I would imagine similar results on a B18. Maybe not the best for drag racing powerband, but for a very responsive road car it would work out well. The T4 1.06 would make 500+whp rather easily.

*this was a rom tune*

 

This test would be much better if the first manifold was also tubular. I would think the 2nd setup with the tubular manifold would have made much better top end power than the log.

 

Ya I agree 100%. But the cams are the restriction at this point. The small housing (and the log manifold) necessitate the use of the small duration cams. There is no doubt in my mind that a 268-272 duration cam stock cam would perform extremely well. At this boost it should make over 500whp with good flowing cams. But the .78 and big cams do NOT work really well together. The backpressure is simply too high. The 1.06 T4 housing will work tons tons better.

One thing I have learned from having a log manifold is that big cams have reversion issues. The stock VE cams have 48* duration, and these VET cams have only 13*. So a larger housing with the larger cams will go a long way to a much stronger powerband!

If I did the log manifold with the bigger cams, vs the tubular with bigger cams, the tubular would make a lot more power. At this point the cams are the restriction. Wimpy 244/248 duration cams to keep reversion at bay.

Now the .78 housing has less reversion, but it is more of a restriction than the log manifold/ .63 housing on the top end. I would have thought the tubular manifold would offset the restrictive housing... but I was wrong. I would also expect the TS to spool a bit faster as well.

If you ask me, the .78 is a lot like running a open .48 housing. The reduction in reversion offsets the loss you would see from a normal .48ar though, so it ends up performing just like a .63. A 1.06 will most likely flow like a .63 (and spool similarly) but will make power like a .82. Of course this is just speculation.

I do like to see that the longer runner tubular manifold still spooled faster than the log.

 

Who made the manifold and does it have EL runners?

 

Element racing in Utah made the manifold. It is unequal length, but the shorter runners have tighter bends, so it turned out pretty decent. I can't really go into flow dynamics with it, because it hasn't been tested. But it is a variant of their manifold on their own car. We just lengthened runner 3 to make it as close to eql as possible.

The runners with tighter bends ended up 2 inches shorter than the rest. I try not to get caught up in the EQL bandwagon, because it depends more on just length of the runners. The radius and tightness of bends should also be considered. A 90* bend will not flow as quickly as a straight, for example.

But the manifold and TS completely negated the loss of response generally caused by longer runners over a log manifold. It now spools even faster, and on the street makes 15psi by 3700rpm in 3rd gear. With a larger TS turbo it would perform nicely. For a T3 to see a an increase in spool over a log mani is still pretty good. Over another tubular .63 I bet it would be ~500rpm.

I wonder how well a larger open .82 housing would do if I put the stock 268* cams in.