Slevin_FD

iTrader: (26)

Did some research today, and BNR's stage 3's are .63 a/r T25 and rated at 36 Lb/min . So there is more room to go, but the .63 a/r is choking them up somewhat. Does anyone have any idea where to find or I guess make a .85 that is internally wastegated? I found one but it's v-band and so that won't work, although I suppose i could weld something together.

Turbo8

iTrader: (17)

If you had a system that sequentially could tip past 450rwhp I think you wouldn't be able to build them quick enough. Factory twins response and 450+rwhp would be a lethal combination.

Slevin_FD

iTrader: (26)

In know right
Anyone with any ideas on how to do sequential without internal wastegates?? The best I can think of right now is some sort of double-"T" between the two turbos. But I have a feeling that idea wouldn't pan out. I'm doing some stuff on solidworks right now so I'll post something ( probably crappy ) in a bit.

thewird

iTrader: (8)

Honestly where there is a will, there is a way. Its just time and effort to put together. It CAN be done. The only real question is, do you really want that jerk at 4-5k. The bigger the turbo's, the bigger its going to be when that 2nd turbo come on. I think the compound turbo idea would be better just because that 2nd jerk isn't going to be as bad as with a sequential system.

Of course I'm speculating (as are all of us) so I could be way off :P. If I had unlimited resources, I'd try every combination and find the best one but unfortunately that is not that case hehe.

Slevin_FD, I think the only way to do this would be internal wastegates cause if you do external your gonna have to use 2 which is going to be a huge fit to fit all that on that one side.

thewird

Slevin_FD

iTrader: (26)

well if you could run .85 a/r turbine housings on the BNR twins you could break 500 non sequential pretty easily, and I'd imagine 400 + would be within reach if you stayed sequential without too much fuss. I used to have a single that was smooth as glass until it boosted. Then she became a monster. Much like my ex wife actually . But back to the point, lets remember this came out back in the day when the ZR1 corvette was only pushing 375, so it was a serious contender at 300HP lighter and with better handling. The stock twins are still respectable, the problem is that now their are 300HP 4-cylinders and 700HP V8's running around. And it's difficult to compete with that now that the playing field isn't even. Not to mention chassis fatigue and the reliability limits of the engine when its used for daily driving. As I see it there are only a few ways to solve this. 20B, Big single , NoS , V8 swap or modifying the twins. I gave some serious thought to doing a compound system but it takes alot of pipe, and space is already at a premium under the hood. Really I think a parallel twin manifold could work and have better boost response than a big single all while making the same or more power. right now I'm looking at some t3's rated to about 400HP each I'm going to print out some compressor graphs and do some plotting tomorrow, and I'll let you know if the math works out.

Et3rNaL

iTrader: (17)

This is Howards set-up that I purchased... hope it helps

Slevin_FD

iTrader: (26)

Any shots of the manifold?? This is exactly what I want.

thewird

iTrader: (8)

A parallel twin turbo system would be fairly easy to make but honestly for less trouble you can make a fully divided manifold with twin wastegates and a divided turbine housing would result in the same thing if not better and be much simpler as well as lighter. Or even Howards idea using a single gate but having it fully divided until the gate itself using a divider.

thewird

Slevin_FD

iTrader: (26)

I'm almost 100% sure these are internally wastegated. As for weight Large singles are heavy some as much as 50lbs. Given the option I'll take two 20lb turbos instead. Besides, spool would be much faster , or at least more tailoerd

Liborek

Based on what?
Smaller turbos are inherently less efficient than larger unit, both on compressor and turbine side. Fact that You would be using two inferior units against one more efficient just promotes this.

If you want responsive setup, use single which is just capable for your goals, and use boost - its well known fact, that target boost is one thing, that makes huge difference in spool and response...

thewird

iTrader: (8)

A 67mm turbo weighs about 25 pounds with a manifold that would weight less then anything twin based. Not sure what turbo weighs 50 pounds, maybe a truck turbo? With higher boost you can do it on an even smaller turbo.

thewird

Slevin_FD

iTrader: (26)

Wird, I stand corrected about the weight of the turbo It was late and I made an assumption instead of researching . My apologies. Having researched the GR35R is between 23-28 lbs depending what site you look at. As for the weight of a twin setup . Wouldn't this depend directly on the size of the turbos your running. ? Either way If two smaller turbos give me more power with a slight increase in weight and save me from buying two wastegates I'd at least consider it.

arghx

iTrader: (3)

BMW did a study of single turbo/single scroll vs single turbo twin scroll vs twin parallel turbo:



Now technically this study was done on an inline 6. Because of the firing order, actually a 13B isn't so different from an inline 6. It alternates between the front and rear rotor just like the an inline 6 alternates between firing from the front 3 and then the rear 3 cylinders. It's one of the reasons why the rotary and the inline 6 are both naturally balanced. The rotary has increased overlap and increased pulsing effects compared to the I6. Properly sized parallel twins are going to have a better balance of power and response than a single twin scroll turbo. The problem is the increased complexity of the setup.

You need to take sequential twins off the table. There is a 0% chance of being able to engineer a system in your garage that will be reliable and won't have a violent turbo transition. Do parallel twins with a custom manifold, and focus on the sizing of the runners and the turbine housing. If you've got the dollars go with the latest ball bearing GTX style turbos.

Slevin_FD

iTrader: (26)

I"ve been looking at the new Borg Warner turbos http://www.treadstoneperformance.com...7670+EFR+Turbo
Thanks for the post , and just breifly thinking about this. With smaller twins wouldn't it be possible to go above 500HP as a daily driver??? Obviously compressor maps need to be consulted, but possible?????

TwinCharged RX7

iTrader: (14)

I agree. I've been doing a ton of research on how to control this type of sequential system (in my case its related to twin charging, but many of the same concepts apply, except my controls are on the cold side, yours would need to be on the hot side).

In order to create a sequential system that operates smoothly, you'll need advanced control valves and advanced systems to control them. Its much easier for OEM's to manufacture these types of items (cast manifolds with built in valves), but doing it on your own will make it extremely difficult to get smooth transitions (most likely wild jerking/fluttering during transition). You might possibly be able to hack up the stock manifold and use the stock valving but that could be difficult as well.

I'd recommend compounding turbos, with the big turbo forcing air into the little one (like the video previously posted). This has been done many times and is rather simple and has smooth transitions. One view of that video and its easy to see how they route the piping and where the valving is needed.

Parallel turbos of the same size is another option that has been done, but doesn't seem as good for response as compounding a different size turbos.

And Parallel turbos of different sizes doesn't seem like the best option for response or high end power.

cptpain

iTrader: (1)

If he really wanted to custom build a sequential system, he could give Sound Performance a call for info on how they built a sequential system for the Supra's.

Look at the smoothness of the powerband during transition in the link,
http://www.supraforums.com/forum/sho...und-Up./page31

Slevin_FD

iTrader: (26)

You hit the nail on the head.
I've worked on lots of high HP imports , even a turbo vette once, my one complaint has always been lag. Yes it's all fun and games on the dyno when you find out you have 600 HP, but where are you getting it from in the RPM band?? Generally above 4K. that's not as bad as it once was, but when I'm crusing down the interstate and some hot shot honda guy in a turbo s2000 rolls up on me and manages to annoy me. I don't want that 3-5 second wait while I downshift and then build boost . I just want to push my foot down and feel the boost kick in or continue building to whatever PSI I have it currently set for. Ideally I would want a setup that could boost from 2500 -2800 through to redline , but that would be a tall order for any turbo. And outside of a compound setup I just don't see it happening. I think parallel twins is a solution that I and probably many others could live with. Plus it would look nice in the engine bay. EX. http://web.mac.com/markrolston/NewSi...and_Feeds.html

ptrhahn

iTrader: (9)

Agree this is virtually impossible to "engineer" from your garage, but I don't much see the point in basically two independent "single" turbos for most applications. Just get a single.

If you're going to deal with "two of everything", sequential is the way to go. It's going to take real engineering though... I can't view the supra thing.

Slevin_FD

iTrader: (26)

I'm thinking about it this way. If I have two smaller turbos rated at 350 each and the compressor map works out that I can reach X boost on one rotor at the proper RPM then I in effect have a potentially 700HP engine. Maybe my theory is flawed, but I think it has merit. As for compounding I think you could use different sized turbos so the rear turbo for example has a slightly different compressor map than the front turbo. Perhaps this is just a matter of changing the A/R. but I do see your point about high RPM limitations being possible.

Slevin_FD

iTrader: (26)

I think with the parallel concept you might be over complicating the design. with smaller twins you could go with internal wastegates which simplifies the manifold design. dual downpipes wouldn't really be an issue because the turbine housing is smaller so it would be 2.5 in instead of 4 inch for example. So yes a total of 5 inches but routing it wouldn't be excessively hard. You would need a dual inlet intercooler, and plumbing the air intakes wouldn't be that hard either if you offset the turbos. Other than that it's the same as going single.

TwinCharged RX7

iTrader: (14)

Based on Arghx's post about the BMW analysis,

Looks like the inertia is the main benefit of 2 small turbos in parallel vs. 1 big single:
I think for two main reasons:
1. It takes less force to make them spool (quicker response).
2. Less back pressure on the engine because of the reduced parasitic loss required to make them spool (more power).


Back pressure comes into the equation again for another benefit. Two 3" downpipes from the turbos creates much less back pressure than even a single 4" downpipe. An area of 57 sqare inches on two 3" pipes vs. 50 square inches on a single 4" pipe.

But of course, this benefit would only be as good as the rest of your exhaust allows.

Liborek

I didnīt want to say that he wants to use inferior turbos. What I meant is, that smaller turbo, be it compressor side or turbine side or both, is inherently less efficient than larger unit.
Its matter of physics. You must take into account everything. Two smaller units will have more heat loss, higher aerodynamic losses through more air-gaps between turbine housings and turbines etc. Nothing what would promote spool.

You are right about Porsches. But its rather for packaging reasons of boxer layout and disadvantages are suppressed by better manifolding.

thewird

iTrader: (8)

Yes, that is a tall unrealistic order if you want true 500 rwhp. The stock twins get full boost by ~2500 RPM. Get a 3/4 rotor as that would better suit your goals. 2x twin-scroll 4 rotor turbo, now there is something that I'd like to see.

thewird

Slevin_FD

iTrader: (26)

You sir are insane. Much respect to your mad science.

arghx

iTrader: (3)

Yeah I attached the 22psi graph to this post. The dip in torque is pretty good for something a shop rigged up--better than I would have expected. It's nowhere near what Mazda and Toyota achieved with their sequential systems (in stock form when working properly). I suspect that Sound Performance did not use a sophisticated system for prespooling the secondary turbo. Have you ever driven a completely stock Mark IV Supra? You can barely tell its turbo. It drives like a DOHC V8 from that era (Lexus, Infiniti, etc).

According to Stoffels, "Analysis of Transient Operation of Turbo Charged Engines," 2010 (PM me if interested in this) the angular acceleration (spool) of the turbo is based on:

1) The torque generated by the turbine wheel
2) The torque consumed by the compressor wheel
3) Frictional losses
4) rotational inertia

When selecting a turbo the basic ways to improve spool are:

1) make the turbine more efficient through optimization of the turbine wheel, turbine housing, and exhaust manifold
2) use a more efficient compressor wheel that requires less torque from the turbine to generate the required airflow
3) reduce frictional losses due to better bearing design etc
4) reduce rotational inertia by decreasing size and weight of the rotating assembly

See attachment. Getting back to the BMW study, appropriately sized parallel turbos are still going to be superior to a single twin scroll for response. This is especially due to less rotational inertia. The BMW N54 twin turbo engine uses very nice exhaust manifolds and excellent internal wastegates, but the turbo rotating assemblies are plain old journal bearing TD03-10T Mitsubishi units. BMW recently switched from twin turbos to a single twin scroll unit. This was a cost cutting measure--they added their throttleless Valvetronic system so they had to go to a single turbo to offset that cost.