I'm suggesting that if you wish to substitute something aftermarket that you made in place of a known HIGHLY engineered product that it would be beneficial to have some data to support doing so. Why do you think the EFR series is so incredible? Because Borg put ALOT of design work into every aspect of the turbo and its operation, and I can guarantee you that included thermal and fluid flow analysis used to sculpt the internal flowpath of that hot side housing to suit the turbine wheel shape (which was also designed for optimal efficiency using engineering analysis software).

Nowhere was I antagonistic in my question, I simply asked if it had been considered which is a valid question given that is how turbo manufactures design a turbo for performance and efficiency.

Skeese

Im having the same boost curve with the 8374 than with the stock S4 setup...ok porting might help here.
But Im not sure if you get it, when using a flap to close one runner to the turbine to reduce A/R to improve spooling, you also sacrifice the pulse charging effect, which imo is very strong on 13Bs.
So effects on spooling kinda cancel each other out. Mazda made this progress when going from S4 to S5 design, so why going back ?!

Quick spool valves do not shut off a complete runner, they reduce the size of the opening to focus the airflow path which, I could be wrong, but is supposed to create more velocity to spook up the turbo faster.

The pulses from the two rotors should still exist.

I didn't take skeese's question as antagonistic. While I think this new manifold with built in valve could be awesome, it'd be interesting to see if it's really better than existing quickspool valves and compare that to the price difference to buying a manifold and a valve vs this new design.

Quickspool valves aren't that common on rx7s (I've never seen a build with one), but a lot of supras sure have them. It'd be interesting to see how it really benefits a rotary. Are we talking 100 rpm shift or 500, or 1000.

My apologies then, perhaps I'm too used to seeing people pick at eachother on forums and just assume.

The reality is to assess stuff like this even a $15K/ year Solidworks flow sim licence is useless if you wish to assess anything other than steady state flow into the housing, even then it's unlikely you will be able to calculate an accurate energy imparted to the turbine (for a nominal smooth flow ~16 cylinder on a long/large volume exhaust manifold etc etc). You are probably looking at 100k in software and an engineer who has multiple years of sim work to get data which reflects differences near the accuracy of a well instrumented engine/chassis on an engine or hub dyno with consistent anbient conditions. I went down a path exploring something slightly different requiring high levels of accuracy for work and concluded it would need to be a Phd for me and my employer forking out for an entirely new software package that could do iterative solutions for combined flow/fea simulation to get close to a couple of weeks of comparative bench testing (with controlled clearance operation on one) on piston cylinder unit pressure gauges. The more you learn the bigger you know the holes/assumptions/fudge factors are. Sometimes moderately good real world data which can't answer all your questions is much more effective than months of fancy simulation.

Reason quick spool valves aren't common on the rotary are because the improvement in spool with a divided exhaust housing versus the log manifold required by a quick spool valve are well documented by Mazda (S4 vs S5 FCs) and in the aftermarket where both log and divided cast and tubular manifolds were present.

The Borg Warner and Turblown designs of putting the variable geometry swing gate in the turbo exhaust housing the allows for using existing divided exhaust manifolds and the associated advantages.

Only thing better would be to have a variable geometry and variable volume exhaust manifold. This would be consist of four smaller diameter runners (two front and two rear) so each turbo volute would be fed off both the front and rear rotor- then with the quick spool valve closed the runner would have small diameter runners to only one turbo runner for highest velocity.

Would be complex to weld up out of tubes, but fairly easy to investment cast such an exhaust manifold.

The popular Supra quickspool valve by Sound Performance is for divided manifolds.

LOL think! :biggrin:

The quick spool valve is for a divided flange on a log style manifold.

Are you going to get better spool running your two rotor with one exhaust port blocked off?:dunno:

This is not a log style manifold.

High end single turbo supras don't use old log manifolds like what came on the FC.

They have six runners, into a divided t4/t6 flange, and then they add this on to focus flow into only have of the twin scroll. I agree it's an abrupt flow blockage, but they do have proven results. The velocity at low rpm to spool faster more than offsets the less than smooth blockage. Then when it opens up its just like normal.

Not really much different than the borg Warner design, just a bit further from the turbo.

The one being designed and fully integrated into a cast manifold in this thread could be better due to improved flow. But the existing option of sound performance QSP can be had for $500. If the new cast manifold with built in QSP is not more than $500 more than a standard manifold and flows better than the QSP, then great.

:lol:

I used the term "log manifold" because for a rotary that is usually the only reason one runs a pulse collected runner exhaust manifold on a 13B.

There is pulse collected runner manifolds for open volute turbos and there are pulse divided runner manifolds for divided turbos. Doesn't matter if cast or tubular or made from sections of pipe.

6-1 tubular manifold for a 2J is a collected runner manifold, a 6-3-2 runner manifold is a divided manifold.

On a 13B you have 2-1 for the collected or 2-2 for the divided.

The traditional quickspool valve has a divided flange and is to be used on a divided turbo exhaust housing, but must be used on a pulse collected exhaust manifold.

Please understand you cannot run a traditional quickspool valve on a divided turbo manifold on a 13B; when the quickspool valve is closed there would be no where for the exhaust gasses from that runner to go. Well, if it had a single external wastegate without divider the flow could back-up the wastegate runner and into the other wastegate runner and into the open exhaust runner- but that is super convoluted flow.

Moving the diverter valve into the turbo exhaust housing like BW and Turblown have done, a pulse divided manifold can flow both runners into one volute with the valve closed and each runner into each volute with the valve open.

https://www.rx7club.com/attachments/...s-photo125.jpg

That really appears to be a 6-3-2 pulse divided manifold and someone that doesn't understand that it won't work has placed a quickspool valve on it and taken a picture. :rofl:

From Supra store quick spool valve description-

https://www.suprastore.com/spquspva.html


Pictures of their Quick Silver manifold
https://sep.yimg.com/ca/I/supra_2271_60159464

I have personally seen more than one on divided manifolds. The manifolds were divided at the flange, they were not 2 sets of 3 cylinders bundled together

Ok, here is a home made, non SP version. True divided (2 sets of 3 cylinders kept separate all the way to the valve.

Dyno chart shows the benefit.

I'm not that passionate about this either way haha. I'm just hoping that if this manifold comes out it's not more than a standard manifold +QSV

http://mkiv.supras.org.nz/articles/qsv.htm

http://mkiv.supras.org.nz/images/articles/qsv4.jpg

Yeah, like I said all the exhaust flow from the front 3 cylinders is having to flow through the non divided wastegate passage around the divider and into the open turbo scroll at low rpm.

Luckily for that guy who has no understanding of physics his 6-3-2 pulse divided manifold isn't pulse divided because of the shared wastegate runner.

A real shit show.
If it were a proper 6-3-1 manifold (or just the divider cut out) the low rpm power would have been better.

Oh, wow. His design is even more wack than I thought.

So actually its kind of like a hybrid of a normal standard quick spool valve, the Borg Warner and Turblown valve and not having a quick spool valve.

At rest the valve does not close all the way so there is a 20mm gap for exhaust to flow from the front 3 cylinders into both turbo scrolls though the angle will direct it into the rear runner.

"closed"
http://mkiv.supras.org.nz/images/articles/qsv3.jpg

open
http://mkiv.supras.org.nz/images/articles/qsv1.jpg

Well, at least it helps keep it from being as restrictive to the front 3 cylinders, but it also isn't as effective as a standard quick spool valve.

I don't understand why he didn't use a standard quick spool valve that would totally seal off one turbo scroll and cut the divider out in the manifold for 6-3-1 collection... :scratch:

There's some even stranger ones out there. When googling yesterday I found a few examples of merging all 6 runners into half of the divided t4 flange and then using a wastegate where the runners first join together to open up another airflow path to the other half of the t4 flange.

I like Turblowns new exhaust housing idea. Granted, they even said BW prototyped it and never brought it to market.

It creates a restriction when it want's to, and completely removes that restriction when called for. So it isn't this huge valve directly in the way of exhaust gasses like the QSV above.

Just because BW never brought it to market, doesn't mean the underlying engineering or theory is bad. It could have simply cost too much money for BW to continue to develop it for mass market, or mass production could have been way too expensive, or it could be a million other things that add cost and complexity when a large company creates products.

So a small and agile company like Turblown could come in and tailor it for a specific application, in small batches, for a small market, could easily be successful.

Yeah I'm not too sure about the antagonism and criticism either. Whether or not you have a chip on your shoulder for whatever reason, Turblown are developing this off their own back and it's a cool idea. You have to respect that. I'm almost certain even the nay-sayers will be very interested in this space.

x2 Could you post this? Awesome little find there!

See now, my comment may have initially ruffled some feathers, but it directed the thread into a great technical discussion that has been awesome.

Great input, data, logic and points made from all involved.

Skeese

Not my picture. It's from a site for trucks.

http://i2.photobucket.com/albums/y29/dohcsky/WGQSV.jpg

Found the thread on that design above. The guy was getting 5-8 Psi at around 2800rpm on a HX60 with a Ford Lightning.

He also said he was getting a 1:1 back pressure ratio. I figure that a low preload spring in the wastegate would keep backpressure down.

https://www.theturboforums.com/threa...296574/page-28



https://cimg8.ibsrv.net/gimg/www.rx7...2737b9260b.png

Really smart design, and with a teardropped entrance back into the main pipe, it could be even better.

Afaik NO!
If you combine 2 rotors or more than 4 cylinders the pulses will delete each other.
They have to be seperate for I think 240° of crank angle at least to remain.

https://cimg6.ibsrv.net/gimg/www.rx7...bc8c1f405f.jpg
I made for my 8374 a very short manifold which is only possible with IWG, so pulses are super strong


https://cimg7.ibsrv.net/gimg/www.rx7...46d04eddd7.jpg
And this is boost curve in 1st gear, boost builds from 2500rpm

If you were to make something like a large 3 port orbital spool valve (think of a 3 inch by 3 inch cylinder with an offset port cut the shape of a steam pipe bend through it, just need to ensure it is alway open from exhaust port to one or both sides as it transitions) you could have something that presented a pretty much standard pipe section in divided mode and could join the two rotors cleanly at low rpm while adding almost no additional manifold volume that sat in/on the manifold side/between manifold and any split pulse turbo rather than the nasty poppet valve of a conventional wastegate. I was considering drawing something up myself (if i can't find something suitable in an industrial process valve catalog) as a test just with a split pulse 35R before moving to a 9280 later on.

Has anyone ever tried one of these. I haven't gotten to the point of actually looking at a specific turbo and comparing to an EFR or spoil valve, but reading about how it works.

Mostly diesel truck applications. So maybe this is only sized to be beneficial for much higher than street FD boost levels (Typically less than 18ish psi).

VGT stuff is always interesting to talk about. You need to offset the inefficiencies of added mass and complexity (tuning, reliability, etc).

I'll stay out of it until I see something I totally disagree with.

Yes, the diesel VGTs have been used on a couple rotaries- you can find some on youtube.

Some issues.

1)
Spool on a rotary is largely defined by the compressor wheels surge line because there is so much exhaust energy available.
Example, EFR 8374 with standard fixed exhaust geometry can already surge the compressor from spooling up too fast.

So, there has been great response (how fast the turbo spins up and max boost per rpm available is achieved), but little improvement in spool (how much max boost/torque is available at any rpm).

2)
Variable geometry exhaust means variable exhaust manifold pressure for any set rpm and exhaust manifold pressure affects the high overlap rotary engine Volumetric Efficiency greatly.

That means, it is really impossible with a full VGT turbo to correctly map the ECU to what the engine needs for fuel and spark without some additional inputs.
Exhaust manifold pressure compensation could be used like in the native diesel application.
Using Air Flow Meter input in addition to MAP input could be used instead or in addition to EMP sensor.

However, with the on/off of the "quick spool valve" style VGT you are fine with traditional ECU inputs/mapping as long as you have a set rpm the on/off transition occurs at.

3)
These stock diesel VGT turbos are small internally and huge externally. They are only ~400rwhp turbos on a rotary and too big externally to be used as twins.

4)
They use VGT as the crutch to spool and so don't use other technology like Gamma Ti exhaust wheels or even ball bearing centers.
An EFR 7670 without VGT will put them to shame in a much more simple package.

Turblown-
What is the smallest exhaust wheel you are planning that these housings can be machined to?

Can you divulge what the single and combined A/Rs are you are shooting for on this exhaust housing?

Ironically, because of compressor surge limits and Gamma Ti exhaust wheel max RPMs I think this design will work best on the smaller EFR turbos or if designed for large single A/R (like 1.00) and huge combined A/R (like 2.0) to accentuate high rpm power instead of trying to accentuate low rpm power with regular 1.00 A/R combined and half that for the single scroll A/R.

Unless the goal is just transient response where the gate is closed down to one scroll in vacuum and opens as boost comes in. In that way it should work great on the bigger turbos to improve driveability even more.

Damn physics...

Unless you are pushing methanol boost levels or running a big full PP drag car I don't think many people are likely to approach critical speed on the 80mm turbine if using the new 74mm inducer compressor (9280). I think that will be the best implementation of the quick spool, hopefully be similar or slightly better to build boost than a conventionally set up 9180 with significantly more headroom.

You are right about the max RPMs for the 80mm Gamma Ti exhaust wheel.

What I was thinking is we know the 80mm Gamma Ti exhaust wheels are too fragile to exceed BW recommended RPM by pairing with a smaller compressor.

This fragility may also show itself when subjected to the harmonics produced by too high a rate of acceleration up to the max recommended RPMs. I don't think the ball bearing cartridge is as good at damping the harmonics as the standard bearings in the usual VGT turbos.

Imagine the 9280 with this Turblown VGT exhaust housing sporting a 1.00 A/R primary position and 2.00 A/R secondary position- that would help a semi p-port rotary push the compressor closer to its limits!

I don't think rate of acceleration will have a significant impact if people have been using ignition cut rev limiters and antilag already in terms of impulse or imparting vibration to the totaring assembly. It's just radial/hoop stress due to inertia at speed, lower exhaust temp will have a marginal impact.

I would guess Elliot is aiming at something like 0.7 area to radius of each volute so it comes out close to the largest factory housing?

x2 That would be pretty nice. :)

I actually had a functioning (sort of) VGT setup on a 13B 5 years ago using a Holset HE351VE. They have a 60mm compressor inducer and 60mm turbine exducer, not exactly stellar for a rotary, but good for 350ish whp.


I was controlling the VGT rather crudely with a spring-preloaded air cylinder with open and close pneumatic ports, a solenoid, and a small pressure accumulator w/ some check valves. The solenoid was controlled with a digital output from the ECU, so the only tuning option was change the open/close setpoint or add more preload, set the minimum AR size with a bolt/jam nut on the lever arm, and adjust the preload on the air cylinder.

Granted, this was 5 years ago and I was using an archaic Megasquirt MS2, I was very hardware limited – no monitoring turbine speeds, eMap, ect to make tuning changes.



What it did do well – at the small A/R setting (default for low RPM/low boost) it was QUIET. With a 3” exhaust and nothing but a Borla XR-1 muffler, it was quieter than my full Racing Beat 3”. Because I had the accumulator, I could store pressure and actuate the air cylinder when the engine wasn’t in boost. Which was neat because I could open the VGT at idle if I wanted the car loud or quiet. Pretty pointless but a cool party trick I guess.



With the VGT cranked down hard to get max spool, It would surge the little 60mm compressor, but the eMap would actually open the VGT before the air cylinder. It definitely felt choked out at high RPM. It also made drivability really jerky because it would come it to boost fast and abruptly. Fuel mileage also sucked.


Ultimately, I went to an EFR 9180 and it had better drivability, better torque past 3500, and made about 200whp more.

I think having a medium sized primary A/R and a giant secondary A/R would be great for the larger EFR turbos like the 9280.

We've finally got pricing from BW on the new EFR units;

https://turbosource.com/collections/efr

We also spent some time today again on the our new variable A/R housing. I will get some video up on that for you guys when I have a moment.

Can't wait to see some videos on the 8474....

HPA is now working away on the install, dyno will be in their webinar series;

Any update on these? After my wedding, I plan on doing a refresh/upgrade to the FD and selling my old setup. I'm interested in the iwg and the variable housing option. This honestly sounds like the most ideal setup for me on paper (same power if not more than my 6766 with an even broader powerband), plus the simplicity and reliability of an iwg setup.

Our variable EFR housing is EWG only. Its took complicated to use with an IWG setup, and probably make boost control even harder under certain conditions.

Bump for any results on the 8474 at higher boost levels (20+psi)

Yes please! ...X 2** ;)

Same here. Bump for updates on the 8474 please!

Currently deciding between the 8374 and the 8474. My only reason for leaning towards the 8374 is the surge line is shifted to the left more than the 8474. But if the 8474 doesn't have a surge issue and it can perform at high boost, i know which one i am getting!

Hpacademy is probably going to be the first ones.. They are working on the rest of the pipe work at the moment, and fuel system;

Guys at HPAcademy are buttoning up the V mount at the moment, then its dyno time!

Im realllly excited to see these results. The new EFR turbo seems like it'd be exactly the evolution of the already amazing turbo the hardcore racer is looking for... a potential game changer for the modded 13B :)

I guess good things take time ........looking forward to some results.