Twin Scroll Revisted

[GST with PSI]

the dsport test is flawed because they used a twin scroll manifold on a open scroll turbo so power on the open scroll was less than it should have been.

How so? The open scroll turbine housing isn't affected by the TS manifold. The turbine housing essentially just becomes the collector for the manifold. So, how exactly does that equate to less power?

 

[94awdcoupe]

How so? The open scroll turbine housing isn't affected by the TS manifold. The turbine housing essentially just becomes the collector for the manifold. So, how exactly does that equate to less power?

they dont show the manifold but those twin scroll housings have a cast inlet to match them. certainly the manifold would have been made with the cast collector. it would be a restriction right at the mouth of the single turbine entrance.

 

[bastarddsm]

Oh that garret piece? Yeah that thing is a turd.

 

[donniekak]

There is no free lunch, you trade spool for top end power no matter how you do it. The twin scroll may give you more power in an rpm range you'll never use at the drag strip for a better trade up top than just running a smaller housing. But, it's still a trade.

Go to the drag strip and watch the pro mods run. They run big open housings for the most part.

Twin scroll turns a laggy 800hp turbo into a better responding 700-750hp turbo. None of the tests that the twin scroll guys like to post are of totally maxed out turbos at choke flow.

If you want your car to fell more responsive on the street, twin scroll is an expensive way to get there.

 

[bastarddsm]

^ exactly. very well put donnie. A lot of the comparisons I've seen have been against sub par ss setups. Then the ones that do really good there isn't much information on them.

 

[ceedawg]

I've seen lots of disappointment on the evo front too. Lots of guys claiming things, but no one making the numbers or the slips.

How much power are you thinking yours is going to make? If it's not 7-800 you have way to much turbo, and a smaller ss turbo would have done you better.

If my car is not making over 450 whp at least I'll be shocked. I hate to speculate however I know that the link showed an airflow rate of 54lbs/min at 27psi but I would rather see what a Dyno says to eliminate any guessing.

 

[bastarddsm]

This is the point I'm getting at, if that's where your at you could have been on a bolt on 256 or 259 or something for a lot less hassle.

 

[ceedawg]

The twin scroll may give you more power in an rpm range you'll never use at the drag strip for a better trade up top than just running a smaller housing.

Twin scroll turns a laggy 800hp turbo into a better responding 700-750hp turbo. None of the tests that the twin scroll guys like to post are of totally maxed out turbos at choke flow.

If you want your car to fell more responsive on the street, twin scroll is an expensive way to get there.

Im sure an open scroll would register more hp and top end , I understand all of that the whole theory behind the size of the TS turbine having an effect on low and midrange power etc, but I also know from personally driving my own car that this thing is a monster mid and top and the response between shifts is on point in comparison to having an open scroll housing. I think any smaller than .91 ar would show some diminishing returns on a setup like mines but let's Dyno and see what I have. I'm impressed with how she runs and I bet the hp curve would look better with what I have vs an open scroll housing.

 

[ceedawg]

Stock 7 bolt 2g mani same turbo and TS T4 setup 658 whp 587 wtq

2gGSX, Jul 11, 2010 #1

You must be logged in to view this image or video.

<object width="640" height="426"><param name="movie" value="

You must be logged in to view this image or video.

"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="

You must be logged in to view this image or video.

" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="640" height="426"></embed></object>

Went to dyno the car yesterday expecting like 550 whp or something. I knew it was fast from the 70-90 and 80-100 times, but I had no idea what kind of numbers to expect. Needless to say, I was pretty blown away when they told me what I ended up making--658 whp and 587 ft-lbs

E85, V3 SD, 36 psi, stock 2g intake manifold, regular 2g head with some cams, and a freshly rebuilt stock trans. Could have ended a lot more disastrously!

Some people were asking me about the misfire and smoke show at the end--the misfire was from the dyno operator hitting the clutch before letting off the gas, and the smoke is a lot of blow by.. need more crankcase ventilation.


Report
#1 Bookmark Like + Quote Reply

 

[ceedawg]

400 whp and tq @ 5k And over 500wtq at 5.5k I'll take that for sure!

 

[bastarddsm]

It's 500rpm slower to get to 400ft*lbs, but look at that top end.....

 

[ceedawg]

I'm not disagreeing with you man in the sense that the open scroll with the same turbo would have more top end yes you are right however you cannot say that the two car examples I have shown you are not making descent top end with better drivabililty and more midrange power on a ts vs an open scroll.
Mr Peepers car and the few examples we have discussed show some good #s dyno wise and at the track. You would be more than welcome to drive my car if we lived closer, Alex(Boostdriven) tuned my car and he's a Holset guy. He himself told me that for the size of my turbo and my setup the car felt like it spooled up almost or just as quick as the smaller HX40 but with much more torque, now Alex could tell you better than me exactly what he noticed, but he thought the car felt very strong all around and in between shifts ,topend included. Hopefully I can get him to chime in here.
I spent a little more coin ,yes indeed but I know Im making more power all around vs having a S259 at lower boost pressure. My present turbo with an open scroll would be less responsive down low as a S259 but would eventually anhilate a S259 turbo in the higher rpms (common sense). The only thing an open scroll will give me is what I discussed in my opening post (lag time where I can count 1 2 3) but will pull somewhat harder upstairs,now why would I not be able to tell the difference between how my car felt with an open scroll setup vs it's present TS setup now?

 

[Canadian_CD9A]

Despite being a twin scroll fan, I'm going to have to agree that the test is flawed. The manifold has an unnecessary restriction with the opposing flange, that's why it doesn't make higher power. This test proves that a twin scroll manifold will spool a twin scroll turbo faster, but a 'proper' single scroll manifold to single scroll turbo would be more indicative.

 

[bastarddsm]

The report is worse than that. On the same A/R housing the difference should a much much greater difference. I don't think they even used real data. - unless the garrett housings have the a/r adjusted for the t/s, which they should have mentioned if it is true. They claim to be a "tech" magazine, and then don't report any tech. Just some voodoo magic.

 

[Canadian_CD9A]

Going by the numbers, I would think this has to be a GTX35 variant. Those vband housings are made in 1.01, .83 and .61 but the single scrolls are too. That said, I doubt they scaled the housing to suit (even then, there are no equivalents on that list), and there is a lot of info missing. Too bad..

 

[Canadian_CD9A]

For what it's worth, the car in question is WrenchMob's 240SX, and there is precious little info available other than Instagram photos.

 

[DSMReviver]

I'm not sure why this could be even in question. Manafactures like Mitsubishi, BMW, ford and others dump a lot of money in turbo technology and they all come up with twins scrolls for all their cars. It's best of both world for performance and responsiveness. There is an article in evom forum where English racing did a similar test and the top end lost if any was near to nothing. The overall usuable power band was increased tremendously going to twin scroll. I would not drive anything with single scroll from 35r to 42r size turbos. They will just produce stubid legs. If you change to smaller single scroll to fight turbo leg you end up loosing top end anyways. You cannot have both in single scroll format. You are going against physics and volumetric effiency.

 

[bastarddsm]

Hi, The phyiscs laws that control volumetric efficiency do not change when you go to a twin scroll setup.

A quicker spooling turbo just means that the turbine nozzle hit choke flow sooner in the rpm band. The point where nozzle exit velocity is sonic or there abouts. This means that the nozzle hit choke flow at a lower mass flow. From then on the only way to get more mass to flow is to increase the density, by increasing the pressure.

I'm not sure, doesn't this imply higher back pressure for the same mass flow, on a quicker spooling setup?

Now you'll go on to say that the divided housing more effectively collect and uses the kinetic energy of the exhaust gas...blah blah blah.... Do you even know what that means? The exhaust gas has very very little kinetic energy before the nozzle. The nozzle is what turns the internal energy of the exhaust gas into kinetic energy by speeding it up. The nozzle size and mass flow is what determines the kinetic energy of the gasses, not weather 2 or 4 cylinder feed it.

The other place we look at is revision due to pressure waves. We see this as an EVO event causing a pressure wave that increases the pressure at the port of a cylinder at EVC. Now most performance cams open the exhaust valve 30* bbdc, and close them 10* btdc. This is about 40* of crank rotation, with an open scroll system. At 6000 rpm the duration of 1 turn = 1/(rpm/60). So 1/(6000/60) =.01s. The time between evo#2 and evc#1 is (40/360)*.01 = .0111 Now in exhaust gas the speed of sound is somewhere around 600m/s. So in this duration the pressure wave is going to travel 600*.00111 = .66m, so further than the length of a standard header, still evc is on the very begining of the pressure pulse before it gets too high. and with a linger header it could completely miss this.

Lets look at the info for a twin scroll. The crank rotates about 220* between evc#3 and evo#1. (220/360)*.01*600 = 3.667m....looks like a t/s has just as much of an issue as the open scroll. #3 will be closing in the meat of the pressure pulse. The open scroll and twin scroll both have this issue.

I have an SAE paper from ford highlighting this exact issue. They go on to say it does help responsiveness, but it is primarily to quicken cat light off.

Also don't ever assume anything an OEM does is because it's better. Money and Emissions are the #1 drivers. #2 is acceptable performance, and way way down the list is good performance.

 

[ceedawg]

Bastarddsm You have great points. However I would say this and I'm not some big engineering guy or physics major. It seems to me that especially on my car I requested 1.5 inch runners on my exhaust manifold. Next I wanted at minimum the .91ar housing. I think that anything smaller than a .91 housing would just choke the system to death also creating a tremendous amount of back pressure and high egt's. I think the hks 272s I have with the ported cyclone manifold done by keltalon helped tremendously as well. The butterflies are inactivated left open. The biggest thing I would conclude with a ts system very importantly is to go big with the AR housing and exh to help the TS system flow.If one goes with Big cams , small TS AR housing, Big intake and you may be asking for problems. Seems like everything that I have is complimentary.

If I needed more topend with what I have I'd go with a bigger TS AR housing. So I think with the big turbos like I have,you definitely need to have the appropriate AR housing to not choke up the flow and inhibit whatever topend power you could gain with a TS setup.

If I am off base please respond!

 

[donniekak]

This is what a big name 4g63 fabricator name has to say on the topic.

http://www.yellowbullet.com/forum/showpost.php?p=38373914&postcount=13

It mirrors what I see, and hear from a lot of people. Twin scroll jacks up the back pressure to get faster spool. They don't lose quite as much power as a small single scroll because the cylinder grouping helps prevent reversion.

 

[GST with PSI]

There is no free lunch, you trade spool for top end power no matter how you do it. The twin scroll may give you more power in an rpm range you'll never use at the drag strip for a better trade up top than just running a smaller housing. But, it's still a trade.

Go to the drag strip and watch the pro mods run. They run big open housings for the most part.

Twin scroll turns a laggy 800hp turbo into a better responding 700-750hp turbo. None of the tests that the twin scroll guys like to post are of totally maxed out turbos at choke flow.

If you want your car to fell more responsive on the street, twin scroll is an expensive way to get there.

I think Donnie said it best. If you want it in a nutshell, there it is.

You guys can argue all day about specific setups, and how to mitigate lag while reducing top end loss between single and twin scroll setups. Like Donnie said, there is still going to be a trade-off. There's no sense in arguing the best way to balance the scale, since each user will have an optimal way to balance their setup depending on application. Being dead set on doing things either one way or another rather than picking the application that works best for you just means you'll have a setup that performs sub-par. Being able to consider ALL the options, and pick the best one for a particular application/goal will save a lot of time, money, and ass pain.

 

[ceedawg]

I think Donnie said it best. If you want it in a nutshell, there it is.

You guys can argue all day about specific setups, and how to mitigate lag while reducing top end loss between single and twin scroll setups. Like Donnie said, there is still going to be a trade-off. There's no sense in arguing the best way to balance the scale, since each user will have an optimal way to balance their setup depending on application. Being dead set on doing things either one way or another rather than picking the application that works best for you just means you'll have a setup that performs sub-par. Being able to consider ALL the options, and pick the best one for a particular application/goal will save a lot of time, money, and ass pain.

I hear that.
Also saying that I'm satisfied with my setup! Will dyno soon just to be curious knowing that it does not ell the whole story. However your point has been heard. I hope BMW(M) , Mercedes(AMG), and all the others who have gone TS on their turbocharged 400+500 + hp cars are listening.

 

[94awdcoupe]

they dont show the manifold but those twin scroll housings have a cast inlet to match them. certainly the manifold would have been made with the cast collector. it would be a restriction right at the mouth of the single turbine entrance.

http://cimg1.ibsrv.net/gimg/www.evo..._55c1943f65c7c105ae98e6703cd64127b6585656.jpg

You must be logged in to view this image or video.

 

[GST with PSI]

they dont show the manifold but those twin scroll housings have a cast inlet to match them. certainly the manifold would have been made with the cast collector. it would be a restriction right at the mouth of the single turbine entrance.

Wow. Yeah, I didn't realize the manifold used a cast inlet like that. That's a horrid design...

Good reading on the topic 94awdcoupe mentions: http://www.evolutionm.net/forums/ev...-twin-scroll-vs-single-scroll-turbo-test.html

 

[Canadian_CD9A]

Someone finally did a proper comparison between SS and TS on a 4G63!!!

http://www.4gtuner.com/topic/27035-single-scroll-vs-twin-scroll-turbine-housing-dyno-comparison/


I thought I'd put up my latest dyno chart comparing the power and torque curves of my old single scroll turbine housing setup vs my new twin scroll turbine housing setup with some comments around the actual perceived differences in drive ability of the car.

The turbo is a Garrett GTX3076R. The single scroll turbine housing was a 0.82 A/R size and the twin scroll is a 0.83 A/R. The motor is a 2.3 stroked 4G63 on E85 with a 3" exhaust.

In both instances, fuel, timing etc have not been changed. The only changes were the turbine housings and the exhaust manifold. The same dump pipe was kept but modified to accommodate the new geometry of the twin scroll manifold. The boost map was altered slightly to suit the new characteristics of the setup but in both instances max boost was set at 26psi.

The twin scroll exhaust manifold was completely brand new - I did not have the exhaust manifold from the single scroll setup modified to suit the twin scroll. The new twin scroll manifold was designed and built so that the cylinders would fire into the merge collector in paired pulses and the runner pairs are completely separated all the way to the turbo. The manifold uses two external 44mm waste-gates (1x each runner pair) that are also completely separate and each have their own screamer pipe. The single scroll manifold used one 44m external waste-gate with screamer pipe.

It's quite evident from the dyno chart that the power and torque response has improved by upto 600rpm, with a new peak torque of 570Nm at 3647rpm vs the previous peak torque of 550Nm at 4422rpm. Power output is the same but this was expected.

The change in torque delivery has noticeably transformed how the car performs on the track. Where I used to keep the car in 3rd gear around the tighter corners and it would exhibit slight turbo lag upon exit, it now "slingshots" out of the same corners in the same gear with absolutely no lag. Throttle response on each gear change has also noticeably improved and the overall performance of the car feels much better.

It was not a cheap exercise to do the change from a single to twin scroll. The cost of the new turbine housing + the cost of purchasing the additional waste-gate needed to be considered. The time and cost behind the design and fabrication of the new exhaust manifold was also substantial, however very necessary as a poorly designed and made setup would not have enabled the full potential of the change and resulted in it being a waste of time and $$$$.

However for my application and setup I feel this modification was very worthwhile - whilst the single scroll setup performed well, the twin scroll has made the performance characteristics of the car more suitable for what I want to do with it.