When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
These numbers fly in the face of everything you read on the rx7 forum about how much air a rotary consumes. They typically talk about 60lb/min being required for 450 crank. plug that into the calculator and it doesn't compute . We are missing something here .....
Oh the calculator doesn't work for HP on a rotary. Our BSFC is way off from piston engines. I just fiddled with the HP numbers to make the pressure ratio come out where I wanted it.
Here's a discussion: Typical BSFC for a rotary - RX7Club.com
Although the HP per amount of air is wildly different, the mechanics of air entering the engine are the same.
I agree about 60lbm/min and 450 crank as a ballpark, but you need a pressure ratio of at least 2.4 to get there. There are ways of increasing the turbo's pressure ratio to get it back on the efficiency island.
Oh the calculator doesn't work for HP on a rotary. Our BSFC is way off from piston engines. I just fiddled with the HP numbers to make the pressure ratio come out where I wanted it.
Here's a discussion: Typical BSFC for a rotary - RX7Club.com
Although the HP per amount of air is wildly different, the mechanics of air entering the engine are the same.
I agree about 60lbm/min and 450 crank as a ballpark, but you need a pressure ratio of at least 2.4 to get there. There are ways of increasing the turbo's pressure ratio to get it back on the efficiency island.
I used 0.62 as a BSFC per the RX7 club notes. Yes, I found that thread too ....
I think you misunderstood me slightly. I was simplifying a lot just to get the variables down, and go down to basics. I'm talking about the pressure feeding our normal intake. If you need 1.01Bar or 100Bar from turbo to get this pressure, does not matter for this simplified reasoning. I am no expert at all, and have large holes in my knowledge, but I'm usually usable when it comes down to basic understanding. In stead of just throwing an opinion.
From the numbers mentioned, 60lbs for 450 flywheel HP, it kind of sounds like the (very) simplified math is closer than I expected. We have 230+Hp at NA at 106% Ve. At 1 Bar overpressure, nothing else changed, we are at 460 -470FWHP.
When we dont get this, we can dig into why. My guess is that turbulent flows at intake and exhaust ports are contributing with quite some. On top of heat generated in compressor, pressure loss through IC. And remember that you are pumping in air with high density, where at the exhaust pressure has decreased, and volume thereby increase. And thats before you start adding effect of heat and other things from combustion.
Ok a couple of points. 106% VE is not at peak HP, peak HP at around 8k is closer to 103%VE and that's with a stock engine. I can accept a ported engine either has a higher VE at peak HP or a higher peak HP at a higher RPM both leading to a higher flow rate but we are talking about a stock engine.
Please ignore HP... The turbo doesn't care what power the engine is making, it only cares about flow and pressure.
Try these numbers on for size in Boost adviser:
650hp target with race gas (yes this is unrealistic, just forces the program to give the final pressure ratio.)
2.7l engine (up from 2.6l to correct the engines VE)
VE of 1.0
4000 mid rpm and 8000peak
No intercooler, but 100% efficient.
14.7psi and 72 deg
You get a pressure ratio of 2 and 57lbm/min. Close to what Brettus is claiming.
But lets go for (more) realistic numbers
Change the HP to 575 (again, just to make it put out the right pressure ratio.)
Put in an air/air intercooler with 70% eff and 1psi drop.
Here you have numbers closer to what I'm claiming, and that's with an optimistic intercooler and piping.
If the intercooler is any more restrictive it moves that number even more.
These numbers fly in the face of everything you read on the rx7 forum about how much air a rotary consumes. They typically talk about 60lb/min being required for 450 crank. plug that into the calculator and it doesn't compute . We are missing something here .....
A lot more than you realize that discussion already played out on this forum a long time ago. Kane even showed how/ran all the numbers for everyone back then ....
I ran all the calculator for the 7163 awhile back, including turbine size, inlet disharge temp at both the turbo and intercooler etc. You have to understand how to manipulate for the rotary engine accordingly. That said, I wouldn't calculate a BW EFR turbo using Garrett's calculator, or vice versa. A compressor map is only as meaningful as the flow points to which the turbine can actually spin it to ...
Is the 7163 the most efficient choice? No. It will transfer potential power into heat, but a good, well placed intercooler can account for it within reason. At 8 psig the turbine flow has not even flat-lined at high rpm i.e. wastegate is not stressed. The original reference for bringing 7163 up was regarding response and it will have response. It all just depends what your goal targets are. There's give/take in any selection.
The old SC claim about boost "now" is pretty much BS these days. 20+ years ago, maybe. You can't get 8 psi below 3000 rpm and carry it out to 8000+ with a centrifugal SC like you potentially could with a turbo. The ability to achieve it with a turbo requires more than lines on a compressor map. Then there is the SC efficiency drag draining total output at the flywheel down to a much lower number as well.
The only realistic SC choice is a roots style and although there are at least numerous successful examples here with the Pettit kit, the engine is still stressed much higher for a given output at the wheels and I crossed off those ideas a long time ago. Most of that success imo is due to Pettit putting a decent kit and tune together. There really is no decent turbo kit. Even the Greddy has to be modified quite a bit It's mostly various people hacking something together. So the potential for error is much higher for the less technically inclined owner. Moreso if they attempt their own tuning too.
.
Just for clarification, the issue with the centrifugal supercharger is that it's rpm is tied to engine rpm. For example if 8k engine rpm corresponds to 100k supercharger rpm, then 4k corresponds to 50k supercharger rpm. You can't get both worlds with a centrifugal SC. But you could use it for top end only. I agree roots style is much better for our RPM and power band, but that doesn't mean it's useless for all applications.
A lot more than you realize that discussion already played out on this forum a long time ago. Kane even showed how/ran all the numbers for everyone back then ....
I ran all the calculator for the 7163 awhile back, including turbine size, inlet disharge temp at both the turbo and intercooler etc. You have to understand how to manipulate for the rotary engine accordingly. That said, I wouldn't calculate a BW EFR turbo using Garrett's calculator, or vice versa. A compressor map is only as meaningful as the flow points to which the turbine can actually spin it to ...
Is the 7163 the most efficient choice? No. It will transfer potential power into heat, but a good, well placed intercooler can account for it within reason. At 8 psig the turbine flow has not even flat-lined at high rpm i.e. wastegate is not stressed. The original reference for bringing 7163 up was regarding response and it will have response. It all just depends what your goal targets are. There's give/take in any selection.
The old SC claim about boost "now" is pretty much BS these days. 20+ years ago, maybe. You can't get 8 psi below 3000 rpm and carry it out to 8000+ with a centrifugal SC like you potentially could with a turbo. The ability to achieve it with a turbo requires more than lines on a compressor map. Then there is the SC efficiency drag draining total output at the flywheel down to a much lower number as well.
The only realistic SC choice is a roots style and although there are at least numerous successful examples here with the Pettit kit, the engine is still stressed much higher for a given output at the wheels and I crossed off those ideas a long time ago. Most of that success imo is due to Pettit putting a decent kit and tune together. There really is no decent turbo kit. Even the Greddy has to be modified quite a bit It's mostly various people hacking something together. So the potential for error is much higher for the less technically inclined owner. Moreso if they attempt their own tuning too.
.
Ok a couple of points. 106% VE is not at peak HP, peak HP at around 8k is closer to 103%VE and that's with a stock engine. I can accept a ported engine either has a higher VE at peak HP or a higher peak HP at a higher RPM both leading to a higher flow rate but we are talking about a stock engine.
Please ignore HP... The turbo doesn't care what power the engine is making, it only cares about flow and pressure.
Try these numbers on for size in Boost adviser:
650hp target with race gas (yes this is unrealistic, just forces the program to give the final pressure ratio.)
2.7l engine (up from 2.6l to correct the engines VE)
VE of 1.0
4000 mid rpm and 8000peak
No intercooler, but 100% efficient.
14.7psi and 72 deg
You get a pressure ratio of 2 and 57lbm/min. Close to what Brettus is claiming.
But lets go for (more) realistic numbers
Change the HP to 575 (again, just to make it put out the right pressure ratio.)
Put in an air/air intercooler with 70% eff and 1psi drop.
Here you have numbers closer to what I'm claiming, and that's with an optimistic intercooler and piping.
If the intercooler is any more restrictive it moves that number even more.
Tried messing with it as you suggested and the numbers do get more realistic . From my point of view
450 crank
2.7L engine
BSFC 0.65
Pr2.12
53.65 lbs/min
Which is still assuming a very nice intercooler and plumbing layout. It's trivial to drop it <50lbm/min in real life example. The Garrett calc also assumes number for turbo efficiency so it can give you numbers before you pick, but in real life your turbo is getting more inefficient the more you push it out of the envelope which is making flow to the engine lower at any given pressure.
My point is that you can't really get that far off the turbo map with this engine, and pressure coming off the turbo is higher than you think (and that's not necessarily a bad thing). There is no point in super-sizing the compressor side of the turbo beyond the max needed for your HP goal, the hot side is a completely different story but that's for another day.
Which is still assuming a very nice intercooler and plumbing layout. It's trivial to drop it <50lbm/min in real life example. The Garrett calc also assumes number for turbo efficiency so it can give you numbers before you pick, but in real life your turbo is getting more inefficient the more you push it out of the envelope which is making flow to the engine lower at any given pressure.
My point is that you can't really get that far off the turbo map with this engine, and pressure coming off the turbo is higher than you think (and that's not necessarily a bad thing). There is no point in super-sizing the compressor side of the turbo beyond the max needed for your HP goal, the hot side is a completely different story but that's for another day.
Thanks for the BSFC tip, as mentioned, I have holes in knowledge that is important. Explains why a rotary needs so much air pr HP. And as Harlan say, turbo cares about flow, not HP produced.
So, still thinking very simplified, Ve = 100% at any RPM etc
For a 2.5 engine(?) at the dyno Team posted, but as an NA and estimated 6000 max RPM, cylinders will suck air at every second round. So:
2.5 / 2 = 1,25l.
1,25l x 6000RPM = 7500l/min = 8.453kg = 18.6lb/min.
At PR of 2.5(just putting the finger in the air, I GUESS that they are are around this at the Subaru, estimated around 200FWHP as NA), we are around 46,5lb/min, something that explains why the turbocharger Team suggested, is adequate for 500HP on an Otto- engine. But we need a lot more air to produce the same kind of power due to a BSFC that is way lower.
Sounds like an valid theory to me, agreed? EDIT: And that the BSFC number mentioned around the net is highly questionable?
There's some interesting info/ideas being brought up, but the OP was about SC research. Not another thread about Turbo vs SC...
As far as options for SC, I've commented in a couple other SC threads, Rotrex seems to be the best shot.
The rotrex 'clutch' system and high rpm capabilities allow for a much wider flow capacity we would need to make decent power. Even with parasitic loss, Rotrex seems to be very efficient and could make up for that loss in steady power band low to hi-end whp.
in reality it will either slowly taper up to 8 psi from nothing across the full rpm range i.e. peak boost only at peak rpm, half of that at 50% lower rpm, etc.
or it will come on like a freight train and be difficult to control boost level
regardless, maybe you will be the FIRST person to either complete one to full fruition AND be successful at it ... bunch of youtube vids of an italian kit on the dyno, they never posted a single number and it disappeared entirely thereafter, the topic will surely come up again and again though
Just to point out; All the contributions so far is valuable in my opinion. As said earlier, I'm quite **** about understanding before I go into something.
It's a big undertaking with several potential issues, just so you don't go in unaware or under-funded. With the rotary engine's unique lower efficiency yet higher flow requirement spread over a wide rpm range as compared to a piston engine it's a bigger challenge overall.
Sry for necropost, but did anyone ever actually get a Rotrex build going? Most seem to be looking at the C30-94, but I think the Renesis could benefit from going slightly larger (C38-71).
04-07-2015, 10:49 AM
that discussion already played out on this forum a long time ago. Kane even showed how/ran all the numbers for everyone back then ....
