I see your point about thermal loss through the turbine housing. It's a matter of materials and purpose. But I'm not really sure about the aerodynamic losses because that's not my engineering focus. I took a class on it long ago in Fluid mechanics and Thermal dynamics in turbine machinery, and I would have to re-read my textbook to really be able to have an educated conversation. But I will say this. there are people much smarter than myself that spend millions of dollars developing these turbine housings, and I would think that they have done everything possible to minimize any air gaps or other ineffecient areas in the design. Is it perfect I highly doubt it. Is it as close to perfect while maintaining marketability and keeping manufacturing costs reasonable ? without a doubt.

 

I will say it in other words. I highly doubt, that anyone could build parallel twin turbo setup on rotary, which would have any benefit in transient response and spool than single turbo capable of same power.

Studies posted by "arghx" are really interesting, but I think that most of its "response" is made by compact manifolding.

Didnīt You anytime think about why Howard ditched twin setup? Theoretically, it should be better than single, but I canīt remember one thing, that would make it superior or rather on pair with properly build single setup....

 

If I'm not mistaken howard changed his setup for his 200MPH run in Texas this year. Hopefully he can weigh in on this at some point.

 

I'm not trying to derail this thread but after readying it what I understand is the op want a faster spool ad that is why he chooses a twin turbo setup.
With that said if I'm not mistaken wich is very possible I am, many people use a twin wwastegate setup on singles for faster spool. I don't think anyone has mentioned this....
Also is it possible to create double twin wastegates (4). Would that be the ultimate setup? Or am I completely off here. I am not an engine builder so maybe someone can educate mr while helping other at the same time?

 

Your somewhat on the right path. I do prefer the faster spool of twins vs.a large single. This thread started out with me trying to find out if there was a way to further upgrade the stock twins beyone the BNR stage 3's. What I discovered is that the BNR's are capable of producing more HP , but are limited by the .65 AR of the turbine housings. There are in fact .85 A/R housings but they aren't internally wastegated and also son't support the 4 bolt flange of the stock twins.
From that the thread headed into the direction of somehow making my own sequential twins . In short this isn't really possible. Even as a shop owner , this isn't something I specialize in so I moved onto another idea I've toyed with from time to time. That idea being a parallel twin turbo system . There's been alot of debate on whether this woudl be a good idea or just another in a long line of failed experiments. So far we have discovered that on paper parallel twins would be able to provide the necessary airflow to produce in excess of 500RWHP while maintaining a powerband similar to the stock twins. By similar I mean they would not have excessive lag that you see in a large single. .
Ideally I would like to have a setup that can produce a wide poweband and boost between 3000 and 3500 through to redline. the issues that have been brought up against this is that small twins don't flow enough air to reach redline. as well as being ineffecient by nature of their size due to thermal loss and potential cavitation on the turbine side of things. As well as the necessity of having multiple wastegates. and more piping than a single turbo. At this point we found out the Howard Coleman had previously developed a parallel twin turbo system that did in fact produce in excess of 550 HP , although we don't as yet have a dyno sheet available to see if and where power dropped off.
The response to this was that BMW tested single turbos vs parallel twins and discovered that parallel twins were better for the production of HP than a large single be it twin scroll or single scroll. At this point we are waiting for Howard, or one of the larger shop owners to comment on the thread. WE have also breifly discussed compound turbo systems, and several other short lived ideas. But it seems the parallel turbos has some traction as it's been uses sucessfully on skylines supra's ans other exotic cars both past and present.

 

Yes, i almost forgot that the car had twins..... i also rode in my buddies BPU supra, and the twins were REALLY apparent, the transition to the secondary was night and day almost like our FD's transition.

 

seems like you are gonna do something custom but wouldnt hurt to check out http://www.a-spectuning.com/products.htm#twinturbokits

 

Sean @ A-spec tuning is currently building my manifold. It supports two ball bearing GT3076(.82 a/r)turbos from TiAL with stainless V-band housings and two Tial MV-R wastegates. Diameters on the two downpipes are, at least going to be 3". I'm looking at building a true dual exhaust setup with V-bands. I also plan on two TiAL 50mm bovs. He should be posting some pics of it in my build thread soon.
https://www.rx7club.com/nw-rx-7-forum-33/my-fd-build-636633/

 

Found some pictures guys. Apparently parallel turbos are somewhat common in Japan. Since they have been ahead of the curve with rotaries since..... well forever I think this lends more credence to the parallel twins idea. The last one is a compound turbo 20B. The middle is more along the lines of what I'm looking for in a parallel setup maybe slightly bigger..

 

I think the middle one is still for sale, just saw it posted not that long ago.

 

I dont see anything on the 20B setup that looks remotely close to a compound setup.....

Looks like another parallel setup to me.

The Blue FD pic, i've seen IRL before. Not much to say about it except that the cars never seen much of the street since its a show car, and IIRC its bridge-ported

 

I just went ahead and skipped all of the post after I saw 'Quick spooling' '500hp' 'Twins' all within the same page.

My suggestion is start spending money.

 

This is pretty much the exact setup I was going to recommend trying. The V-band turbine housings and new-generation vband wastegates help with the fitment constraints. I was going to suggest internal wastegates and smaller downpipes but if you can make them fit go for it. You may have to tweak the size of the turbine housing. Howard Coleman tried different turbine A/R's on his parallel twins and measured exhaust temperature and exhaust gas backpressure with the different setups.

 

i will review this thread and have some comments this evening.

hc

 

Thank you Howard. Your insight will be mug appreciated.

 

We have these dynosheets. Actually several of them. And IIRC there wasnīt single advantage in powerband against single of same power. And power dropped of around 7300 RPMs.

Study was about transient response and Iīm sure that most of the gains were made by compact manifolds with low volume. Its LONG I6 engine, use of parallel twins is proved in this aplication. Not in rotary.

 

i can understand the OPs interest in having some power when just driving around... the 3000 neighborhood.

apparently Mazda had the same idea and therefore the manifolding that we have on the stock FD. for 3000 rpm stuff it can't be beaten. improved ala BNR, but not beaten. like everything it has it's drawbacks, most of which we are aware.

while i did run from 2004-2009 a twin setup it had different objectives, namely top end.

FD factory turbo setup:

cold side total average area of the 2 turbos 5.972 sq inches
hot side total is 5.25

GT3582r:

cold side 6.386
hot side 5.171

stock turbos driven just to a point before surge make around 385 SAE rwhp.

385 X 1.92 = 739 CFM/ 14.471 = 51 pounds per minute.

adding the cold and hotside together and dividing by two:

FD 5.61

GT35 5.77

the GT35 makes 67 pounds per minute V the FD at 51.

advantage GT35:

3% more turbo area and 31% more air.

the culprit is the OE manifolding, not the turbo.

the advantage, of course, is lower end Tq. that's exactly why Mazda chose to move away from the Turbo 2 setup.

20 years later Tech marches on w major upticks in turbo options. you can pretty much have it all compared to the early 90s when the FD was on the drawing boards.

there is no need to bolt 20 pounds of heat soaking cast iron to your heat receiving aluminum rotor housings. especially when the rotary produces 300 degrees F more exhaust temperature than a piston engine AND requires 30% more exhaust to make the same piston hp.

if you do want 3000 rpm power do stick w the stock setup as it does just what Mazda wanted. upgrade the turbos (BNR) if you wish for a bit more.

i built my twin setup w quite different objectives.

due to the wide FD gear spacing when you shift at 7800:

2nd gear 4512
3rd gear 5384
4th gear 5607
5th gear 5607

(and the above numbers don't consider a few hundred RPM drop off from drag between the gears)

question:

when was the last time that you looked at a posted dyno sheet at 5000 RPM?
chances are you looked at top tick and maybe where peak torque was.

if you want to win races you look at "hp under the curve."

take the hp starting at 5000 and each point up in 500 increments to 7500. add them together.

top total wins the race.

hp at 5000 (and up) is V important.

and it isn't just about which turbo.

HUTC is also greatly influenced by:

Porting
Manifolding
Tune

there is a small book that could be written on each subject as it relates to creating midrange power.

for example, it is common to tune using AFR. get the AFR right and you're done. not exactly. by logging pre-turbo EGTs you will find the your timing is probably nowhere where it needs to be given you need to see 1400+ F from zero to one bar. change the timing, gain power and all that time you will find no AFR change!

i mention these factors because if properly exploited they do move the lower end hp nearer the stock setup, and because they will significantly effect HUTC which i consider all important.

my entire setup, engine, manifolding, support systems and tune is primarily designed to make 400 rwhp at 5500. this can be done without compromising topend. i make peak torque at 6500.

as to turbo selection for twins... OP wants 500 max. that'd be 66-70 pounds per minute divided by 2. so you are looking for a couple of 35 pound turbos. it is way too easy to go too big. too big costs... spool primarily. i ran two TO4E 46 trims. around 44 pounds per minute. i was looking for 600 rw.

the setup made 507 SAE at 20 psi with what was determined later to be a significant boost leak around the secondaries. (oops). fixing the leak moved the boost curve to the left 500 rpm and added 1.5-2 psi. i never was able to max out my setup (27 psi) due to the wastegate spring but am confident it would have made 600.

while i learned alot from my twins and really like them the advances in turbos in the last year make the simple single setup increasingly appealing. i am very very excited to tune out my own design single manifolding especially w regard to midrange hp. we should be re-starting the process in a week or two.

howard

 

just curious what does a setup like that cost

 

Around $7K just for the 2 turbos, 2 gates and manifold w./twin downpiping. Don't forget the other supporting mods. Here's what I plan to spend/spent.
$3000 Heat shielding
$4000 Fuel system (fuel cell, pump, filter, fpr, fittings/lines, rails, GZ lim, injectors)
$3000 Engine managment(Microtech) Aux. injection(Aquamist)
$TBD Custom exhaust, V-mount setup, bov(s), piping, tuning and so on....

 

another reason to go with internal gates

You can rebuild a motor with good used parts for that. It's your money but I'm sure there are less expensive ways to get almost the same amount of benefit.

Way more money than you need to spend for almost the same final result.

That sounds right.

Just make sure you have enough money left to actually finish the car.

 

I agree with you 100% ^ Time is on my side though, money- comes and goes, you'll always find something to spend it on. I'm always motivated to build my dream car with the philosophy "you get-what you pay for"

 

I personally feel that if you're spending 7 grand on 2 turbos, 2 wastegates, and the associated piping, you're doing it wrong.
Break down the price and you probably shouldn't spend more than a tad over 5k. That budgets 1500 for EACH turbo, 1500 for the fabrication, and another 500 for the wastegates, I'll even give you the benefit of the doubt and give you 800 for 2 tial 44 wastegates, which on a twin setup is just overkill as you should be using smaller turbos (which WON'T cost 1500 a piece) and shouldn't need the 44s. The gtx3582r which is probably too big, just barely breaks 1500.

 

This could be done for as little as 3K with internal wastegates, but the issue would be transient response, and other than different sized turbos I don't see this really working well. That said I can completely seeing this work for a 600HP road warrior FD. without some crazy aggressive tune.

 

I should clarify that. I see larger turbo taking time to spool and smaller turbos lacking top end HP. So other than two different sized turbos ( which is basically a compound system Basically not a true compound ) it would be difficult to get low end torque. Any thoughts??

 

Well that's the fundamental problem of turbocharging. If you go with responsive parallel twins you don't want porting, exhaust manifolds, etc all designed for power above say 6500 rpm. Realistically you are looking at a setup that will be done by 7000rpm which is basically how the stock twins are. You will need compressor sides that can be wound out to 25-30 pounds, like a GT3071.

That sounds like an obscene amount of boost but it really isn't because you will need that kind of pressure ratio to get the airflow you want. That's how Evos do it. A mildly Evo X will run 25+ pounds of boost but the actual flow will be a lot less than what you would have on a big single Rx-7 running much less boost. It's a port injected engine running 9:1 compression pistons and variable cam timing for both the intake and exhaust cams.



The numbers are lower than your goals, but this is the kind of power and torque band I am talking about. This is an Evo X with basic bolt ons and a tune, on a Mustang dyno. This is running 93 octane with no meth or anything like that.That's a peak of 26psi in the mid range, steadily tapering off until redline. The powerband isn't too far from an FD on stock sequential twins--the stockers run out of breath by 6500-7000 at most. You can see there is 259 lb/ft of torque by around 3000rpm but a peak of 358 lb/ft by 3600.