Those injectors, cleaned and in good working order, are good for about 260-270HP at about 80% duty cycle. They will be plenty big enough for most N/A applications. The 460cc injectors are good to 230HP at 80% duty cycle. The formula I used is on www.sdsefi.com. Even though I couldn't find it this morning. I pulled those numbers from memory but I think they are fairly close.

Knowing the injectors are good for that amount all you have to do is get an intake that will provide enough air and you should be good to go.

 

Pulsewidth is dependent on rpm. Pulsewidth is merely the ratio of open time open time to close time. The reason its important is because at about 85% pw, the injectors stop closing, and just stay open the whole time. At high rpm's, the time between opening and closing is a lot less than it is at lower rpm's. Bigger injectors allow you to inject the necessary fuel in less time.

RPM and hp are going to determine what size injectors you need, not hp alone. But I cant be of much help as far as exact numbers go. You might want to send an email to RC Engineering. And in reference to the SDS site, youll want to find out what rpm that duty cycle is at.

Bill, I know how you feel. I always get crazy ideas stuck in my head....and those ideas seem to find their way onto my cars. Ive actually really wanted to build a partial PP. That will probably be the next motor I build, hopefully it will coincide with the dunebuggy I want to build...

 

I meant Microtech LT8, not the LTX. Oops.

I guess Ill post a pic of the primary port, and my exhaust port.

 

 

BTW, that intake port isnt finished, just the rough porting is done.

 

that exhaust port is ridiculously insane. is heat going to be a problem with no exhaust sleeve?

 

On injector size, even Hitman sugests(why can't I remember how to spell that right now?) using 550cc on a 12A Mild port.

 

Thanks for sharing your port photos M7. I see you are opening the port a little earlier than a mild port with just enough metal for the corner seal to be supported. And the exhaust port . . I have to ask . . what type of lifespan would a bare aluminum port have at the temps of rotary exhaust? The reason I wasn't going to take out my sleeves was for longevity reasons. I will get rid of the diffusers, though.

Again, thanks for sharing.

 

Just a quick clarification on the injector sizing. As part of megasquirt development a lot of people have been doing tests on fuel injectors to get the right balance between turn-off time and heat dissipation in the driver circuit. And this is what has been found: for a low impedance injector it needs 0.8ms on time and 0.2ms off time.

So for safe operation you should allow 1ms between injection events. OR at 6000RPM you can run 90% Duty cycle, but at 8000 this has dropped to 87% (quick aside: the Megasquirt has a very neat staging feature, which if you have equal size primaries and secondaries allows you run 720° time cycles on each injector instead of 360, allowing you to stay above 90% DC all the way through.)

I just scanned throught the entire thread again and noted that there was no mention of the recent article by RB on a street ported 6-port with both DCOE, IDA and 4-bbl manifolds. Net result was that this setup CAN get 230HP with the right exhaust.

Bill

 

I ceramic coated the inside of of my exhaust ports. And I have NEVER had any cooling problems with this motor. Stock radiator, and an e-fan. My temp guage never moves (S5). Ive got 9K miles on on my motor so far, and no problems whatsoever. The powerband is a little high from the exhaust port, things dont really start get moving until 4500 rpm, and thats with the aux ports wired closed. Pulls hard all the way to redline.

 

Lots of thanks to Zach for making this Excel spreadsheet. The constants can be changed, and that will change the result. Its pretty self explanatory, so enjoy.

 

Thats a cool little spreadsheet. I had to stare at it for a few minutes though. The one thing that it doesn't take into account though is the acoustic affect going on inside the intake at the same time. Don't worry, no formulas take this into account. There is a very cool link I posted on another thread about exhaust design. It also gets into intakes a little bit but many of the theories are still the same between the two. I just thought you might enjoy checking these out.

http://www.cafefoundation.org/research.htm

All of the articles are very good. Yes they are tests on an airplane engine but a 4 cycle engine is still a 4 cycle engine regardless of what it is mounted in. Check out the one about "Aircraft Exhaust Systems IV". The entire site will probably make your head explode from all the info. Enjoy.

 

Ive read the CAFE site a few times. Some very good stuff there.

As far as the calc, the runner diameter is already optimized, but like you said, doesnt when the other constants are changed. Im not the one who made it, though, so Zach would be the one to talk to about workign that into it

 

Ooh, another good link. I'll get some good stuff out of that one too.
As far as runner diameter, its only a function of volume flow rate, trying maintain a certain intake air velocity. By the way, this is an average, so the final runner diameter should be slightly larger, but not by much since the intake will never close like a piston engine resulting in a large spike in velocity.

 

Yep volume is a funtion of flow rate. It is important to know this though since we need to be able to calculate maximum intake runner air velocity and detemine if the runners are properly sized. Obviously on standard engines we are limited in size to how big the stock port runners are. Maximum intake runner velocity at max rpm should correspond to .6 mach (7920 in/sec) airspeed through the runners. For some reason air moving faster than this encounters too many boundary layer restrictions and despite its greater velocity loses the ability to make any more power. Even using a mathematical formula though to predict proper runner length it still does not take into account the acoustic ramcharging effects inherent in all engines regardless of port timing or overlap. VDI would theoretically work on really high overlap engines. Just not at the distance that the S5 manifold has it at. Its a cool little program none the less. It would be really fun to have a program that could calculate the differences between plenum sizes, runner lengths and diameters, and individual runner setups. Probably asking too much here though. Oh well. One can dream.

 

Who would've thought this post would prove useful?

 

I thought the whole accoustic ramcharging thing was what tunning the runner length was for?...

 

Yeah I guess it pretty much is. The intake runners are too short for pressure wave tuning to any big benefit, so we do rely on acoustic ramcharging. This is all that the VDI system does. Since intake and exhaust principles are so closely related I sometimes transpose one thing onto another when explaining things. Intake plenum volume though does effect rpm tuning. The runner length that works best for a certain plenum volume wouldn't be correct if it were for an individual runner throttlebody system. If the car is using forced induction then runner length really only plays an important role when not under boost. This is true only to an extent though. Just got off topic again.

 

I disagree. The VDI system is acoustically tuned, but in a very different way than what was discussed here. The VDI redirects the negative pulse from one rotor over to the other, and times it so that the high pressure area following the low pressure pulse arrives at the opposite intake port just before it closes, so there is a high pressure area at the port as it closes.

Another thing to think about is that you dont have to tune to the first octave. If you half the runner lengths you will have raised the tuning one octave, and instead of the pulse going away from the motor and back once, it will do so twice before the port closes. But that doesnt make as much power, and the torque curve isnt as smooth. Like everythign else, its all a compromise.

 

The problem with trying to model too accurately is that you end up with all kinds of variables that are difficult to measure and even more difficult to simulate (unless you have the right SW). So you end up either needing a serious CFD package and lots of processor time, OR a flowbench to provide a lot of your inputs (and even if you simulate you generally still need a flowbench to cal up the simulations).

Having said that, the equations for a system with multiple resonances are well understood and I have some interesting papers on the subject which one day I'll get around to analysing wrt rotaries.

And don't forget that, as soon as you start putting resonant volumes in, you introduce throttle lag, which being subjective in its affect on the driving experience is a real pita to put a quality factor on to.

If you stick to first order calculations though you shouldn't go far wrong.

Having said all that, I'm a total wimp. I have an IR setup on my 13B and had a block machined to go between TB and manifold which can easily be lengthened or shortened. My next inlet will have sliding trumpets to give more flexibility and speed.

 

The more I think about it the more I confuse myself

The best thing to do would be to design as close to your goal as you can by with whatever means you have at your disposal and then tweak it with good tuning and a trip to the dyno to see how close you are.

 

Don't worry. Your comments on lack of pressure wave tuning in the inlet had me thrown, so I went back to all my notes. Only the gen 1 with carbs s too short. About 20" tunes for a returning pressure wave at around 6000RPM. Add/subtract for port timing differences.

Unless we have a semantics difference.

Bill

 

This thread has me soo confused,..... what im getting from it is ask ms7,bill shurvinton and rotary god what lenght to make your custom intake manifold, averge all there answers together, and build it, i guess i need more school,..

 

Actually there is no right or wrong answer, just a set of compromises to be made and the limits of your fabrication skills.

The main compromises are packaging and power band. As you optimise for high end torque, then low end drops and vica versa. Unfortunately close ratio boxes for the RX7 are fairly limited so you can't get around a peaky design with gearing. Next year I will be going to a straight cut ford box with a 2.2 1st which will stack everything closer together (but I digress).

So decide where your peak torque wants to be and work out from there. For example the T2 inlet is tuned for about 4500RPM. There are PP aircraft engines producing 160 odd HP at 6000 RPM. Not huge numbers, but very efficient and with hugely long runners.

 

Ive seen 210 hp and 196 ft lbs for a PP tuned to 6000 rpm(aircraft engine).