Coheed

I want to hear your thoughts on this housing/turbo combo. I am currently using it in my car, and the response is very very nice, but the top end power is not where it should be. It seems like the .78 is way too small of a housing, but I was wondering if anyone here is running it with good results. Say... in a B20 vtec?

I am currently looking at other housings, I have a 3" 4 bolt .63 housing that is a bit larger that I could use. Just need to fab up a new downpipe. But I am almost positive the .82 housing would suit it best.

Here are my thoughts on the .78. Since it is divided, each pulse goes through only half the housing. So really, it is a .39 housing. Just two scrolls. It seems like the trade-off in backpressure offsets the gains in turbine efficiency that you may see (or VE increases)

JDogg

do you have a divided manifold?

Coheed

Yes it is a tubular divided manifold. twin V38 wastegates.

JDogg

what other divided housings are there?

whats the rest of the setup.

blinx9900

How much power are you making or wanting to make? Like where is the power now and what do you expect it to be? I have a .63 on mine and it makes great power but without numbers that's all subjective

91jdmhatchback

No offence but you really have no idea what your talking about. I will leave this to the shodan or someone else more knowledgeable then me to explain the divided turbine housings stuff.

But you only have a 2L and a turbo capable of around 500ish hp maxed out. that turbo is not big so why would it need such a big housing like the .82?

also why if you have a divided manifold would you even consider a non-divided housing?

i would rather have a power band over peak numbers any day.

HiProfile

Exactly correct. The main improvement is the lack of pulses causing reversion in other cylinders. At higher rpm's, you allow each cylinder to scavenge better because of it, as well improve flow due to overall larger flow capacity. At lower rpm's you can build pressure quicker, meaning faster spool.

The comparo's I've seen show a larger divided housing spool faster as well as make more power. There's a reason you can take the fast-spooling divided Evo turbos up to 400whp, and a good reason many other OEM's choose divided turbines for high-output applications.

blinx9900

A bit OT but you have a link to that? I have never seen a stock turbo evo clear 360awhp unless your referring to one of the FQ models?

AndyFloyd

actually...here is some info for you...might make you rethink that last statement.

I have a 2L Honda engine with a GT3076R .82 housing. Why? Because the car responds better to the .82 except for about 100-200rpm of spoolup that you will never feel anyways. The .82 housing is best suited to the compressor flow of the 57mm 3076R compressor.

If you make 500whp on the .63 housing its possible yes...but think about the exhaust reversion at high rpm on the vtec cam...yea it really pays to have the .82 housing. Because the 30R spools so well anyways the larger housing really isnt even noticeable. I have full boost from 4500-9100 rpm.


However...with his divided manifold...if it really is true twinscroll...he will want the .78 housing.

blinx9900

Please define full boost, i use the .63 housing on my 3076, i have 8.3 compression and a 1.6L engine (b16) and i see full boost (12psi) by 4000rpm, i don't know how much boost your running but it can be a lot more than 100-200rpm of spool.

AndyFloyd

full boost is 12psi ( and sometimes 19psi ). It is 100-200rpm of spoolup, its been proven before on other forums.

91jdmhatchback

You do know that there are and endless list of things that effect spool. It's not even worth listing.

and for a 2L 12 psi on a little gt30 at 4500, seems high.

blinx9900

Sorry bud your wrong, there is no way you reach 12psi and 19psi at the same rpm, your not gonna give all the info don't post, and so far it appears that my 8.3:1 1.6L reaches target boost 700rpm before you...

It is high, my 8.3:1 b16 makes 12psi by 3800rpm!

Coheed

Why don't you flow out a single scroll of the divided housing and post up the CFM flow rates to a undivided, then get back to me. I'm just asking a simple question for some opinions. Have you tried a divided setup yet?

Here is some info on my setup. The motor is an SR20VE, almost identical in flow to B18. Same Vtec setup with variable lift, except each cam is individually switched. The motor is 8.5:1 compression, stock cams (very similar to GSR).

But since the motor responds almost identically like the honda motors, I figured I would start here to see what you guys think. I am pretty sure I know what is going on, but I figured I would ask for some opinions. Why not? I tried a log manifold before this, and I had to downgrade cams to prevent reversion. The stock cams have way too much overlap (48*) for the log to work well, and you could feel a loss of power when the cams switched. So after a while I decided to go twin scroll to prevent reversion altogther. I put the stock cams back in, with the tubular manifold, .78 housing, and twin wastegates. Pretty sick setup. Spool is crazy. 15psi by 4000rpm.

But top end pull was still lacking, though you can't feel reversion anymore. The car has a nice kick when the exhaust cam switches at 3800rpm. You see the boost gauge peg out at 15psi instantly afterwards. Great spool.

I just can't seem to make the same power/psi of boost that I should. I know that 400whp should be attainable at 15psi.

I don't make any more power with the tubular manifold than what I made on the log manifold. With the log I made 442whp on 24psi. 360whp on 15psi. The new manifold seemed to make no difference in power.

So the .78 housing seems to be a large restriction in the system. It makes sense because the .78 doesn't seem to flow as well as the open housings. It spools like a .48.... and seems to make power like one as well.

Before this post gets too long winded, I'll explain the turbine talk. The .78 housing is divided. Both scrolls of the turbo make up .78 area ratio. So only 1/2 of the housing (or one scroll) will only flow half as much. Since each pulse coming out of the collector is only going through half the size of a .82ar undivided housing, it makes sense why it wouldn't flow as much. A 1.06 divided I think would do much much better...but you can't get one anywhere. The gains you get from a twin scroll setup are from the increase in turbine efficiency and prevention of reversion. But i think the backpressure from the housing is enough to negate pretty much any gain you could see from VE increases or from turbine efficiency increases.

I am pretty sure the .82 will kill the .78 in power production. But it is just a thought. I will post up pics of the different housings in a bit and you can see the restriction that the .78 presents.

Coheed

Here are some pics comparing the .78 to the .82 undivided housing. You can see the casting on the .78 is very rough. The divider in the volute doubles the air-surface tension (boundery layer)and the opening has some very rough bumps right on the inlet. Also, the bolts that hold the housing on extend into the exhaust stream. A lot. They extend into the exhaust stream cutting the nickel-sized port in half. So I cut them down a bit to help flow, and to get them out of the air flow.







The two slits you see where gas exits the volute and hits the inducer are also very small. Maybe 3/16" wide. Compared to the undivided there is a huge difference.

Of course the DET guys can make good power with this housing at insane boost levels. But they seem to respond better in scenarios with high backpressure. This VVL head, however, hates backpressure. It induces reversion, and kills power. Even though reversion with this housing is nearly non-existent, the backpressure is enough to hurt scavenging. These are my thoughts anyway.

What do you guys think?

AndyFloyd

lol, i never said from 12psi to 19psi in the same rpm, you cannot read, its ok. What I said was the difference in spool from a .63 and .82 housing is 100-200rpm of spoolup. What gear do you make 12psi @ 3800 in? If I do a 4th gear pull it hits 12psi @ 4100rpm. I guess you beat me there, where you wouldnt beat me is on the street I would roll you into a burrito dude.

AndyFloyd

coheed,

you need the larger housing ( .82 ). I run a F20C s2000 engine. It is very similar to yours and I also had reversion issues with a log, going tubular I have picked up about 40-50whp everywhere and hold that to redline on a 3076. I make 431whp @ 12psi now, stock CR though at 11:1. I lose in spoolup compared to you...I make 12psi @ 4100-4300 looking at logs today.

blinx9900

I can read, you didn't specify so i had to interpret it. Anyhow i see 12psi on the street at 3800rpm in 2nd gear. Now whats this about you rolling me into a burrito on the street? Are you trying to say your car is faster then mine? If so i couldn't care less, bench racing = restarted.

AFI Turbo

I run the same housing on our Time attack s2000. Although the housing is small, the powerband works very well for alot of the events we run. The car makes peak tq (371ftlbs) by 3700rpms and drops to 315ftlbs by redline.

Coheed

The s2000 has variable intake cam timing too, I bet that really helps out a lot. I know this housing can make good power, and spool is just flat out incredible, But I really believe more power could be had with an undivided .63 housing. The .82 is better suited for this setup though. It will be hitting 500whp in the near future, and the .63 might be a bit of a restriction above 7000rpm.

I am going to put in some slightly bigger cams, and open up that exhaust flow to take advantage of the head flow and VE increases. Hopefully I can finally get into crossover, then the power will come easily.

AndyFloyd

have a log to show me? Seems kinda hard to belive man, there isnt one dyno chart for a b16 that spools a 3076R in that range. Sorry, not buying that for one second.

AndyFloyd

On my s2k with the .82 housing and tubular top mount it takes until 4700rpm or so for full boost ( 17psi). Your housing + twinscroll spools it 1000rpm faster. Thats nutty.

Coheed

trust me when I say the spool is nice, but the tradeoff to top end power may not be worth it unless you run E85 and big boost.

I have a feeling that the .82 and a slightly bigger stock cam would make about 60-80whp more on the same boost. We'll have to see. But I am pretty sure I can get pretty close to full boost by 5K even with the .82.

I guess I should also mention I am at 4500ft elevation. If it were sea level the boost onset would be even better. So the .78 def spools fast.

HiProfile

Not really a link, my friend's old Evo8 did about 430whp, although to his credit is was with meth and a clipped turbine. Then went to a 600whp Buscher build, then parted out after a few months. Was sad to see it go...

And Coheed, elevation removes as much exhaust [back]pressure as it 'robs' from the compressor, so spool isn't affected as much as people think. It's mostly just reduced power from the lack of air & higher turbo P.R.'s.


All in all, I'd say that .78ar housing isn't as problematic as the OP thinks. 315tq at 8500 is over 500whp...not too shabby. Especially if it adds another 1000rpm to the powerband.

blinx9900

Sorry bud not here to bench race, i couldn't care less if you don't believe me, its my car, i built it for me, not to brag, not to pwn people on the streetz, etc... If its any consolation to you my low compression allows for a very aggressive tune even with 91 octane, i run about 22.5 degrees up top and 12.5 a/f's with ease, iirc on spool up i run around 30degrees of timing. I'm not a tuner so i could be wrong on the numbers but my tuner is pretty aggressive, likes to push stuff ot the limit but i don't mind as the last motor i put together was experimental to begin with.