Howard Coleman

iTrader: (6)

since AI is a newish topic for many board members and since AI is going to be the single most adopted new mod for turbo rotaries i thought it might be helpful to have a "resource" thread re AI delivery sizing.

after all, Scott is currently running an AQ (that's Aquamist) .8mm nozzle and Brian is running a 15.1 GPM (gallon per minute) nozzle, BDC is running an M10 nozzle, JRat is running an 7 pound nozzle.....

throw all that in w water or alcohol or a mix...

WHAT???????????

as usual, not everyone is together on denoting delivery/size as to nozzles. we of course see this in our primary fuel system components. i run 850/1600 cc/minute injectors which some would call 80/150 GPH (gallons per hour) or 13.5/25.4 Pounds per Minute. of course Rus Collins at RC says they are 888/1658 but that's another story.

before we get into AI delivery needs let's get straight on how much base fuel we need to make power on pump. please note that the rotary engine needs significantly more air and fuel to make power than the piston engine so don't use piston metrics.

a combination of Max Cooper's Rotary Fuel/HP calculator and RC Engineering make our job easy. thanks Max and Rus.

links:

http://www.maxcooper.com/rx7/how-to/...tem/calcs.html

http://www.rceng.com/technical.htm#WORKSHEET


let's quickly work thru an example using Max and Rus's sites:

a single turbo setup w 55 lbs/minute of air delivery running the usual 850/1600 injectors.

what can we make hp-wise...

first... how much fuel is needed to make our horsepower goal. fuel requirements relate to engine efficiency and that differs between piston and rotary and whether excess fuel is needed to cool a turbo’d charge. the metric is referred to as Brake Specific which is expressed as the amount of pounds of fuel required to make one horsepower for one hour.

depending on state of tune/mods the turbo rotary requires between .64 pounds to .60 pounds for highly modded. so let's pick .61.

next is fuel pressure. generally we run about 40 pounds static, that is at no vacuum or boost. raising static fuel pressure is a simple way of increasing delivery and many of us run 43 psi of a bit higher. do note that the increased delivery varies with the square root of the % change so going from 40 to 44 doesn't add 10% more fuel, more like 3%. let's use 42 psi for our calculation.

fuel injector sizing is next, 850/1600 is entered and converts to 81 and 152 pounds per hour.

injector duty cycle concludes the given for this calculation. RC suggests we use no more than 80% tops as when exceeding that number the injectors tend to go static and are therefore “untunable.” but what do they know about fuel injectors. (they wrote the book)... most of us routinely see 87% on our logs.. and more. more cycle does deliver more fuel BTW. so let's compromise here.... we are going for max hp. let's use 85%.

push the calculate button and...

632 flywheel rotary hp
542 rear wheel rotary hp.

more than enough fuel for a 55 pound per minute turbo that generally is capable of 425 rw rotary hp.

of course we all know you can't make 542 RWRHP on PUMP. Raising the boost to make that much airflow creates too much heat in the combustion chamber and tends to break things mechanical.

that's where AI works it's magic.

instead of driving around on engine-friendly racegas, which due to it's higher octane delivers a smooth burn at high boost levels but costs ridiculous, you can have your octane when you need it.

both water and/or alcohol, of course, are the potential injectants and both deliver.

let's now move on to nozzle sizing.

alcohol and water differ as to volumes needed to effect a proper result, as more alcohol volume is needed than water.

it should be no surprise, therefore, that alcohol nozzles generally are bigger.

so let's get into AI delivery volumes...

caveat: various experienced AI players come up w somewhat different numbers. we have an inexact science due to setup differences etc so what i give you is of course an estimate offered to get you started.


WATER
EngineRunnerup: 1-2 gallons per hour per (piston) flywheel 100 hp. start low end and work up.

1 gallon per hour = 63 CC minute. so for our typical 400 rw rotary hp (470 fwhp) engine that would be.... 4.7 X 63 CC minute or 296 CC X 1.1 (rotary inefficiency V piston) or 325 CC/Min. probably a good place to start. i believe Scott and Rice have mentioned they are running in the 300-400 cc neighborhood.

50/50 Alcohol/Water
Snowperformance a legitimate AI player w some interesting hardware (flowmeter)... recommends a 50/50 alcohol/water mix. as such they recommend higher volumes due to the alcohol component.

for instance... a 350 flywheel hp piston application receives a 375 cc/minute nozzle size suggestion. Converting that relationship to our 400 rw rotary hp… 470/350= 1.34 X 1.1= 1.47 X 375= 554 CC/Minute for a 50/50 mix at 400 rw rotary hp.

Therefore, at the 400 rw rotary hp level….

325 CC/Min water

554 CC/Min for a 50/50 mix.

Not carved in stone but a good place to start.

Running 100% alcohol as an injectant is a different proposition in that it just depends on what you want to accomplish hp-wise. Generally my vendor (Alkycontrol) recommends removing 15 to 20% of Pump and replacing it w alcohol.

IMO, it greatly depends on your objective… Stop/obviate knock? Maximum hp? Lower EGTs? Decrease your injector duty cycle? Decrease the stress on your fuel delivery system? Cool your air intake charge?

Max Alcohol delivery, like pump, greatly relates to the maximum air delivery from your turbo. If you are running OEM, or a mid range 55 pounds per minute single or an 85 pound per minute GT42 or the like your max will vary..

I run two TO4s that deliver around 84 pounds per minute. My vendor recommended two M10 nozzles (600-650 hp)that make approximately 1200 CC/min. I have currently reduced my base fuel (pump) map approximately 30%. I see no knock and 1200 degree F egts preturbo. Current indications are I am making approximately 600 rwhp. Since I am still rich (AFR 10.0) I will be taking more pump out of my map as I tune toward my target. Even though I have taken a lot of fuel out my injector duty cycle w 850/1600 at 7800 RPM is at 80%. That should drop as I lean it out… probably to 70%.

BTW, one M15 nozzle is recommended for 500-550…

Others might have reducing knock or egts as a goal and consequently run far less alcohol. It will be exciting to zero in on the various tune modes going forward.

The bottom line is you can have it all w AI, and AI refers to Auxiliary Injection which includes water, not just alcohol……

You can run the max boost your system will deliver. Knock, high egts will cease to be an issue. Your turbo air delivery will ultimately limit your hp. You can run max air delivery…

On the street.

On pump.

WOW.

I will gather together the nozzle sizing and cross conversion on my next post.

howard coleman

BDC

Quick correction, howard -- I'm running an M15.

B

sdminus

Howard

Top man in starting this thread.

you are spot on with the objectives as well.

for public info i run AQ jets for 1 reason only. there atomisation. They do tend to clog a lot thought. I have run m rated jets with ineternal filters but i was moved over to the dark side LOL

I am a realist ie there is a down fall in all types. I am not going to push a product for no reason.

It seems you have worked out he delivery in a similar way to me. ie % of incylinder cooling and then use the water and meth factors.

Howard i can get chemical compo details form my supplier if needed.

Scott

ronbros3

Im also running an M-15, with an SMC system since 2002, no variable rate,50/50. am setting up another car FD, with a SNOW boost cooler system, and trying for 1st time the NRS grey single pc. apex seals, and a complete custom intake manifold, 4" exhaust, GT 4085, .95 turbine hsg. 70mm comp. inlet, hope it all works properly, V-mount rad/intercooler. THX Ron

13BT_RX3

I am considering doing AI a little differently. I want to set it up like a separate fuel system. I would use 1 or 2 injectors that fire from the signal to the secondary injectors. Total injector impedence would need to be within ECU spec. Initial sizing would need to be close and fine tuning could be done with a pressure regulator. Has anyone done something like this yet? Anyone see anything wrong with this?

Howard Coleman

iTrader: (6)

AI thru fuel injectors has been done. if you search a couple of the key forums...

Aquamist... though may be currently down is one of the best.
Turbobuick... 200 pages of threads
EFI101... amazing tech

i believe there is an injector degradation issue w straight methanol but can't say for sure.

as to running it off the secondary signal. i personally don't think that would work well. i run a Power FC Datalogit and looking at a graph of the primary secondary interaction it seems that the commands are all about overall duty cycle.

i do think the optimum setup would be to stage off duty cycle but no one seems to be able to grab hold of this calculation from the ecu and turn it into something workable.

i do think your general plan will be the type of AI system we will run in the near future. i considered something similar using some AQ components and finally decided that i had enough tech hurdles and i needed something that would just plug n play.

howard coleman

13BT_RX3

I suppose having the most important variable unadjustable is a little risky. Adding an auxiliary injector controller would give the ability to tune the system properly.

Howard Coleman

iTrader: (6)

Nozzle Nomenclature:

what's w M10? M15?

i think it would be great if we all settle on CC/Minute.

generally we talk in those measures re our fuel injectors.

so, re M10...

the 10 refers to gallons per hour. one gallon per hour is the same rate as 63 CC/Min..

M10 is 10X63= 630 CC/Min.
M15 is 15X63 or 945 CC/Min.

i run 2 M10s for a 1260 CC/Min rate.

this is all at 100 PSI. and for the detail people you need to net out the boost, so if you want a net 100 psi and you are running 20 psi boost your pressure at the nozzle must be 120 psi. just like what we do w fuel pressure.

howard coleman

J-Rat

iTrader: (2)

Correct, I am running an M7.

I think thats partly because back when I started using AI, no one was really using anything other then water on Rotary motors. So I kind of had to do some loose math and hope for the best.. Much thanks to my buddy Trevor for helping.. I think as more time passes, we may see some good info come out that might standardize the AI systems a bit.

At this point, I am considering going to the Snow Performance meth kit, and kicking up the delivery to dual M10s. Still on the fence about this though, as my Alchy kit (SMC Enterprises) has been the model of reliability and efficiency.

Also, I am running 720 Primaries and 1680 secondaries, as a reference.

Lastly, just wondering what everyone is using for pump lube. I am using KLOTZ UPLON.

BDC

Although this current 5gal jug of methanol has 2oz. of Redline Alcohol Fuel Lubricant in it, I will not be running any lubricant afterwards per Julio Don's suggestion.

I am currently running a single M15 nozzle and it seems to be doing alright. The last time I ran the car I was sitting at 24psi of boost on my 60-1HIFI, with a mix of 70% 93 octane and 30% methanol to achieve low 11's:1 AFR's under load. Sitting at 50% duty cycle at ~7000rpm on 720/1680's.

B

J-Rat

iTrader: (2)

What pump are you using that you dont need Lube? Sureflow?

BDC

Yep it's a Shurflow. Julio modifies them internally. I am not sure what all he does to them but he was fairly insisting on me running just straight methanol.

B

coolingmist

shurflo pumps need no lubrication or additives. The shurflo pumps you see at Northern tool have Viton seals that you can only run a 50/50 mix methanol water. Coolingmist shurflo pumps use EPDM seals so you can run 100% methanol. There are other kits on the market that change out the seals from Viton to either BUNA or EPDM. If you plan on running Methanol, you want either Buna or EPDM, we have ours manufactured with EPDM seals because BUNA does not have as high heat tolerance.

If you just want to run 50/50 standard viton seal pumps will be more than sufficient.

David

ronbros3

I use KLOTZ UPLON lube in the SMC , but the SNOWs has a SHURFLOW pump, also I have heard that ETHANOL is less corrosive than Methanol, any truth to it! but slightly less BTUs, I do know out at the Daytona speedway back a couple of weeks Indy cars where testing , some used Meth, some used ETH, talking to people didnt reveal much tho. Ron

ronbros3

you know thinking about it, A fuel lube would condition the entire fuel system sounds like a good idea ,keep corrosion down, also lube the seals inside the engine. but possably lower octane and possible increase detonation . by the way can OMP lower octane?? ron

J-Rat

iTrader: (2)

The OMP system itself? I would imagine it would lower octane, but maybe by like .5 or something insignifigant.

Asleep

Hey Howard,

Glad to see a dedicated section up. Thanks for kicking in some starting fodder for discussion and the feedback on your (and BDC's) progress on injection. Hey I remember our frist phone conversation on the phone 3 years ago when you were on iteration 1 of the twins and we walked through the CFM calculations of twins versus single.

The quote below from your thread starter here hit me the wrong way...it is an assumption.

"of course we all know you can't make 542 RWRHP on PUMP. Raising the boost to make that much airflow creates too much heat in the combustion chamber and tends to break things mechanical."

Call me optimistic, but I just don't know if it can be concluded that it is not possible. Not probable based on the average car owner's ability to control tuning parameters, yes. You know I am not trying to "inflame" anything here.

Granted, we're talking about what 99% of the rotary community is not willing to or needs to prove. Espeically with the benefits of AI being tested out.

Sorry, had to get that off my chest. Otherwise, great thread start.

Tony

Howard Coleman

iTrader: (6)

given 93 octane (pump) gasoline's really low (444 F) autoignition level you'd want to feel really lucky to throw the 72 pounds per minute of air at the motor necessary to make 542.

methanol autoignites at over 850 degrees which dramatically changes the equation.

perhaps i should have said "shouldn't make" instead of "can't make."

and, of course, we all know that given enough knowledge, will and probably money you can make pigs fly.... like Penske/Donahue w the American Motors whatever it was in the 71 Transam.

point taken Tony. now get busy putting together an AI system.

howard coleman

Asleep

Ouch is within touche! Engine is on the cherry picker ready to back in today... as soon as I break the engine in I will order from our friend in Florida. There is no way I am not running AI this time. Just need to figure what turbo I am going to run next year.

Tony

RICE RACING

For a pure water injected car I generaly run a 3lt to 4lt water tank to suit a 60lt fuel tank... this size depending on the driver habits on the street easily covers a full tank of fuel.

300 to 400cc per minute on 500rwhp cars is in the ball park.

NissanConvert

iTrader: (1)

Holy thread resurrection batman!

So what exactly is the formulae for rwhp : nozzle size?

Howard Coleman

iTrader: (6)

thanks for your interest... please read post #1.

hc

C. Ludwig

iTrader: (27)

Found this chart on enginerunup.com Makes it easier to compare nozzles which are rated in different ways. Anyone care to comment on the validity of the numbers?

M1 - (.011") - 1.0 GPH @ 100 PSI, 1.2 GPH @ 150 PSI, 63 CC/min @ 100 PSI, 76 CC/min @ 150 PSI

M2 - (.015") - 2.0 GPH @ 100 PSI, 2.4 GPH @ 150 PSI, 126 CC/min @ 100 PSI, 151 CC/min @ 150 PSI

M3 - (.020") - 3.0 GPH @ 100 PSI, 3.6 GPH @ 150 PSI, 189 CC/min @ 100 PSI, 227 CC/min @ 150 PSI

M4 - (.024") - 4.0 GPH @ 100 PSI, 4.8 GPH @ 150 PSI, 252 CC/min @ 100 PSI, 303 CC/min @ 150 PSI

M5 - (.035") - 5.0 GPH @ 100 PSI, 6.0 GPH @ 150 PSI, 316 CC/min @ 100 PSI, 379 CC/min @ 150 PSI

M7 - (.045") - 7.0 GPH @ 100 PSI, 8.4 GPH @ 150 PSI, 442 CC/min @ 100 PSI, 530 CC/min @ 150 PSI

M11 - (.073") - 11 GPH @ 100 PSI, 13.2 GPH @ 150 PSI, 694 CC/min @ 100 PSI, 833 CC/min @ 150 PSI

MichaelB145

I have a GT35/40, 850cc/1600cc inj, twin RP fuel pumps, making 500hp on Race gas. I bought a setup that uses twin M3 nozels in the IC charge pipe with all hoses and fitting. All I need to buy is a pump, a relay, and maybe a level indicator. It sounds like the twin M3's will put me right under the 400cc per minute range right? I was planning on using Denatured Alcohol from Home Depot instead of Meth. Is there any reasons why I shouldn't go this route?

I also called Snow Performances tech support and he told me twin M3's would be way to small.....is that because they were thinking of using the Boost Juice 50/50 Water / Meth? Any help would be greaty appreciated as I'm sick of rolling around on the street with 14lbs of boost!

MichaelB145

Here is a picture of the charge pipe I have....