AI 2022 refresh
 

i started this section in 2006 after adopting AI as an important part of the turbo'd rotary system architecture. at that time the TurboBuick forum had a huge AI section. i have been running AI since 2003.

of course any road traveled has it's share of bumps, turns and dead ends. just like the TurboBuick forum those bumps and turns are reflected on our board.

time for a fresh coat of paint, a 2022 update.

AI is a necessary today as it was 20 years ago since the laws of physics are unrelenting.

it is my belief that the community (generally) has a highly optimistic view as to IATs. primarily because of a belief that the thermistor is able to generate timely temp data.

whether the thermistor is "fast" or not, it moves glacially while IATs move instantaneously.

many of us use Thermcouples to monitor EGTs. Thermocouples, unlike thermistors, generate at least four accurate readings per second. "accurate" as in plus or minus four tenths of one percent.

that would be four tenths plus or minus of ONE DEGREE F at 100 F.

let's take a look at the actual data from 22 of my 3rd gear runs at 20 psi, 6000 rpm.

average boost: 19.7 psi
average temp out of the turbo before the intercooler 329 F (EFR 9180)
average IAT as measured in the stock location by a thermocouple: 162 F

welcome to the real world of IATs.

a few qualifiers
data generates at 2250 ft altitude... 8% less oxygen so turbo has to work a bit harder than at sea level so slightly higher temps out of the compressor
around 1320 cc of methanol as AI

this data was generated using four different intercooler cores. the best was my oldest, a Spearco 13 X 17 X 3.5. stock location, similar to the Pettit.

i have a wonderful Spearco 1995 technical intercooler booklet.

the technical data on my specific core (C #2-180) correlates with my data.

to calculate core efficiency:

air T before IC minus air T out of IC divided by air T before IC minus ambiant T

329 minus 162/ 329 minus105 (actual air T next to air filter not outside T)

IC efficiency 74.5%.

Spearco, however, calculates efficiency for this core at my flow rate as 50%.

IAT would have been 217 F if not for the cooling from the meth.

as i mentioned i had 3 other ICs built last year. two from C&R and one from Bell. i looked at Mishimoto and concluded they weren't worth the effort. all three had larger pressure drops and cooled less than the Spearco. George Spears sold Spearco to Turbonetics and i don't have a current read on whether the current offering resembles what i have. Pettit used to use Spearco cores but now they are made in Tiawan. (don't know if that is a negative or not.) Garrett sells intercooler cores and they might be next on my list.

if you look at ALL of the IC offerings... they are too small for our plus 30% flow to piston engine needs. oh well.

conclusion today same as conclusion in 2006... we need all the help we can get to lower IATs.

AI is essential.

here's an item that should probably be added to your system. i confess i am late to the party and ordered a couple today...

Snow Performance is currently the major element in AI. Snow was bought by Nitrous Express so you need to go th NX to find Snow, or Wisconsin.

this looks like a winner:

https://imagizer.imageshack.com/v2/1...923/V0FfX9.png

here's the description from Snow:

The Snow Performance Hyper-Sonic water-methanol injection nozzle is the worlds best selling and most advanced water-meth nozzle. Constructed out of 100% methanol resistant nickel plating, the patented Hyper-Sonic design spins the fluid to supersonic levels before forcing it through two venturi shaped orifices to create the Hyper-Sonic micron droplet mist that has become the industry leader in water-methanol technology. Complete with a removable 80 micron filter, the Hyper-Sonic nozzle can be cleaned and re-used on any application. To unlock power hidden within your water-methanol injection system replace your nozzles with Hyper-Sonic nozzles today! (1/8" NPT Threads)

we all can benefit from better atomisation, especially water as it doesn't flash like methanol. i will report in after receiving them.

what’s the pressure requirement?

Hi Howard,

Interesting. Have you thought of ditching the factory IC placement to make it easer to use a properly sized core? I would have to believe making sure the core size for all the brands you are testing are able to cool the amount of air you are flowing without AI would be a much more valid test. I would think adding AI should be the cherry on top for extra safety but not necessary if a properly sized core is used. Garrett cores are really good as you would expect and should be higher quality then what you have currently used. Would be interested to see your results with one that is properly sized.

Here is a log from a few years ago using a 66mm 9180SXE with fairly large v mounted Garrett core. Core part number 703522-6005 which can be found here. Peak of 29 psi falling off to 26 psi by redline. Ambient temps were mid to high 70's. Sensor back then was the normal "Fast acting" air temp sensor that is commonly used. Unfortunately no pre IC reading at that time. Max temp was 114F which is a little higher then the log image. I'm sure the sensor was a still playing catch up to what is registered in the image.
https://cimg0.ibsrv.net/gimg/www.rx7...038c22baf2.png

the nozzle flow is rated at 100 psi...

there are 11 sizes. there is a bit more info on the EU site...

"Boost Cooler Water Injection Nozzle - Size 10

Flow Rate 630ml/min (6.0GPH) @ 100psi

The SNOW Performance Hyper-Sonic Water-Methanol Injection Nozzle is the worlds best selling and most advanced water-meth nozzle. Constructed out of 100% methanol resistant stainless steel/zinc plating, the patented Hyper-Sonic design spins the fluid to supersonic levels before forcing it through two venturi shaped orifices to create the Hyper-Sonic micron droplet mist that has become the industry leader in water-methanol technology.

Complete with a 80 micron filter, our water injection nozzle can be cleaned and re-used on any application. To unlock power hidden within your water-methanol injection system replace your nozzles with Hyper-Sonic nozzles today. The Boost Cooler Hyper-Sonic-nozzles offer up to 20µ fine atomised mist. - It's the last water-methanol injection nozzle you will ever need to buy.

The key to water injection is how fine the fluid droplets are upon injection. The smaller the droplets, the more total surface area per volume injected which results in more heat reduction = more power. Because of this fact, more cooling can be done with less fluid. More of the intake charge can be filled with cold O2 - not water.


Product Features:

• 2-chamber nozzle, Venturi-Design
• Nominal Flow Rate: ~ 630ml/min / 10.0 GPH
• Material: Stainless steel with zinc plating
• 80µ (micron) fine mesh filter
• Outer Thread: 1/8"NPT27
• 100% Methanol resistant

Note: To use our nozzles, you need our Boost Cooler nozzle holder. The nominal flow data relate to 100 psi pump pressure."

i think but do not know that the nozzles will screw into most of our "holders" as they are just 1/8 NPT. i will advise upon receipt.

https://imagizer.imageshack.com/v2/1...923/jXxpYL.png
https://imagizer.imageshack.com/v2/1...922/gELYJF.png
https://imagizer.imageshack.com/v2/1...924/LTVIvP.png

Neutron,
thanks for the post. nice turbo choice and i totally believe your power number. your data is quite interesting.

assuming the upper numbers are your time log;

peak boost reached at low 16 seconds on the log.

run ended when you cut throttle mid 24 seconds so about an 8 second run.

IATs rise throughout the run and actually peak after.

temps at your cursor (111 F) are higher until mid 34 seconds.

10 seconds after you cut your throttle. actual IATs don't work that way. range bound sensors do.

from my logs

i am at 99.9 TPS and 25.8 psi at 2:54.266 seconds

at 2:54.415 seconds i am at .5 of one percent throttle, -2.4 psi and my IAT has dropped 58.9 degrees. all in .149 of one second.

compressor outlet temp pre-intercooler dropped 58 degrees during that time.

as to intercoolers, i think i may contact Garrett and have them make me a custom setup. i am getting close to them as i am in process of installing my G40-1150. i understand that for some apps a 4.5 inch core might work but for other apps it is a 50 rwhp discount for heat soak.

having logged both pressure drops and efficiency between 4 well built intercoolers i will gladly take the 50+ degree drop in IATs from my AI system. after the best IC you can find if you really know your actual IATs you will not consider it a cherry on the top that you can just as well go without.

this is sort of turning into a nozzle thread... also, to many who have more recently researched AI it is probably obvious that i am in a catch up mode especially w re to nozzles.

it appears that swirl and having a recess that allows a bit of spread from the orfice are the two keys to modern design. since the charge air is moving in the 400 mph area (as per AEM) there is a HUGE amount of instant shear hitting anything coming out of the nozzle.

while re-looking at the threads in the section i noticed "ProMeth"... glad i looked a bit further as it appears at initial glance they have THE nozzle. while AEM seems to have the same fundamental design elements, ProMeth is done in stainless steel rather than plastic. i am going to talk with them today. i do want to say, w re to AEM, that they have greatly upgraded their product (V3) and totally love their video w the dynamic intake flow bench, heated flow no less. bravo AEM. note, btw, the sensors being used to monitor the dynamic data. nothing but thermocouples.

a very interesting dynamic noted by AEM is not only how much faster a better spray pattern effects IATs but how Significantly Lower the flow requirement is... way better MPG:). way better.

here are some eye opening pics of the ProMeth nozzles:

https://imagizer.imageshack.com/v2/1...923/H771Bt.png
https://imagizer.imageshack.com/v2/1...922/8eHxnM.png
https://imagizer.imageshack.com/v2/1...922/fj8v6y.png

I was looking to upgrade from V2 to V3 nozzles is why I had asked. Thanks for the info.

I don't have much to add here except I was less than impressed with the Snow performance nozzles. The coating was flaking off around the nozzle orifice on two of them right out of the box. Looks like they may be coated and then drilled which leaves a nasty edge on the orifice. I am not thrilled of the idea that the coating could make its way into my engine. Even worse might be a flake inside the nozzle? May be fine but I am not sure I want to use them...

I also have a handful of the earlier ProMeth nozzles, which are virtually identical to the Snow nozzles (but no flaking). The ProMeth products I have purchased are fantastic, but the customer service is lacking. I recently emailed to order something with a custom configuration. Rodney was great to answer my questions but when it came to the cost it was not clear or concise. Did not seem to want to provide a formal quote, and this was ~$1000 of product that I was ready to order. I tried to call a couple times, no answer. I then got ahold of Snow performance and despite an 8hr time difference, they were helpful and accommodating and got me what I needed quickly!

Those new ProMeth nozzles look neat and I would like to try them, but I should probably get the car back on the road first.


***In terms of the nozzle holders, the ones from Snow and ProMeth also look identical. There should be no problem switching between these nozzles if your holder is 1/8 NPT based.

really appreciate your nozzle info 1badFB. i had a couple of lengthy convos w Rodney after i had ordered the Snow nozzles. according to Rodney, Snow copies lots of his stuff. apparently ProMeth had a prior design nozzle for a long time that Snow copied. recently PM came out w the current nozzle and the Snow is a copy, according to him. i did talk to Snow yesterday and was told that their current (new style) nozzle is "all stainless steel." just checked w a magnet and no magnetism. i also looked at them w my 10X and i don't see any evidence of plating.

i ordered a couple of mounts so they can attach through my silicone coupler just ahead of my elbow. it is nice that i can easily swap in the ProMeth nozzles if i wish.

Was tough to get a decent photo but here is one of the nozzles I purchased roughly 6 mo. ago from Snow Performance. Should be the most current style, looks like plated stainless.

Disappointed! This is a brand new nozzle and has not had anything run though it. The flakes coming off are large enough to block or obstruct the orifice.


https://cimg6.ibsrv.net/gimg/www.rx7...15d2c6f0ff.jpg

thanks for sharing.

looks like a Snow deal breaker for me. fortunately, i am not quite at the point of firing my new setup so it will be with ProMeth nozzles.

Which ProMeth Nozzle are you thinking?

Well I think I've got most of the puzzle together, but now I need some advise on how to control things. I can either control the system completely with a Snow Performance stage 3 controller or via Haltech elite 2500.

I currently have the Snow UHO pump and two solenoids, but no idea of how to program the Haltech. PWM the pump through a solid state relay and trigger the solenoids based on MAP? Anyone have any idea of duty cycle vs pressure?


Also, are there any suggestions for failsafes beyond monitoring IATs?


Thanks,
Alex

"Which ProMeth Nozzle are you thinking?"

i planned, before talking to Rodney/ProMeth for an hour or two yesterday, to simply replace my two M10 (630 cc/min at 100 psi) with a similar PM size. as you may know they have 18 different jet sizes and delivery can also be adjusted by the number of atomizer slots.

i chose the .0362 with 4 slots. the nozzles are around $27 each. you can buy a nozzle and then adjust delivery by swapping in another atomizer.

i may be going in a different direction as to location so stay tuned, it might get quite interesting:) or not

https://imagizer.imageshack.com/v2/1...922/SOisz4.png

Pre-turbo? Don't change the formula on me now! haha

In all seriousness; I was hoping to compare pre-turbo, post IC, and semi-direct (manifold) injection.


Just for fun, I mocked this up a while back:

https://cimg0.ibsrv.net/gimg/www.rx7...3b09cd4a83.jpg

i like the clecos

With Summit Racing being local I ended up just buying the V3 conversion kit along with some other things needed. It doesn’t come with any literature on how to tell the nozzle inserts apart from each other. You have to go to the AEM site and pull up the documentation to find it. The little plastic swirl-generator flow nozzles are so small I almost need a magnifying glass. The threaded end of the holders are more different than I was expecting; much longer, but the 1/8” NPT thread size is the same.


https://cimg7.ibsrv.net/gimg/www.rx7...f601858dc.jpeg

any way just fyi, apology for the side diversion ..
​​​​​…

I'm thinking of using the prometh solenoid pulsed by elite 1500 and just powering up pump

right now I'm using a ssr and pulsing just the pump and having issues achieving full flow

"pulsing just the pump and having issues achieving flow flow"

your outcome is predictable as attempting to make a DC pump change output quickly is like turning the Queen Mary. you are on the right track. controlling a solenoid (which is essentially the same as a fuel injejctor) is the right way to go.

i also like ProMeth, both the proprietor and products. i may explore a small amount of pre-turbo meth controlled by my Link G4X Extreme and a ProMeth solenoid. i have always had a number of concerns re pre-turbo but ProMeth/Rodney opened my mind a bit. i am mostly looking to lower IATs as my new turbo has more capacity than i need for my app.while i will be powering just a bit more wheel, due to better efficiency around 80 pounds per minute, i will be dropping 20 F IAT.

spoke to Rodney today, he was full of help and very knowledgeable.
didn't push his products on me either, very genuine guy. I will def do business with them in future possibly starting with that nozzle

Was told recently that water/meth can actually eat out the seals in a rotary.
Anyone ever seen this happen?

Maybe Howard can comment in greater detail. I know of a lot of rotary owners that use some form of AI(Meth, Water/Meth water). Similarly people are using E30-85+. From what I have gathered, I think everyone in the community has agreed that adding 2 stroke oil to the gas helps to keep the seals adequately taken care of. There are varying opinion about how much and depends on if you still have the OMP, but Howard has an entire write-up on his website. I believe Howard has been running some form of AI mixture or even E30 for years.

AUXILIARY INJECTION
TUNING FUEL

i have run around 1000 CC/Min of 100% methanol as AI injectant since 2003. while i am aware that methanol does corrode aluminum my throttle body appears totally unmolested. as to the "seals"... i assume you mean the coolant seals? i see no evidence of any degradation.

on the positive side, the apex seals are far less likely to warp due to the reduction of combustion chamber heat.

Late to the discussion, but my Rice Racing Pre-Turbo setup does not lowers IATs. Perhaps I am not pushing the turbo hard enough, or my intercooler is too big for my needs, or my IAT sensor is not fast enough to see the changes...

Would love to see your datalogs once you have it configured/tested!

A little off topic, and I'm sure it's been discussed elsewhere, but what did you consider to be (or were at the time you installed) the advantages of pre turbo location vs. something downstream closer to ports? I'm curious as to the pros and cons of various injection locations.

what mixture were you running? i would assume it would prob cool the pre ic temps more as opposed to the post ic?

Have been running this setup for about 15 years, at the time, Aquamist was the leader on commercial injection, all others were in infant stages. At the time, pre-turbo injection was less complex, not a whole lot could go wrong with the system.


My IAT sensor is before TB, unfortunately, can't tell temps pre/post IC.

I bought a Prometh pump because they say it has an internal pressure relief, but it looks pretty much identical to the Snow Performance pump, so now I'm wondering. I guess I'll go ahead and install it and see what happens. Does anyone know for sure that it does have the pressure relief? I don't see why they'd lie about that.
by https://www.flickr.com/photos/159205146@N07/, on Flickr

I'm planning on getting a Prometh solenoid too since it's only .7amp max, and the Holley solenoid I currently have is 30amp, which explains why it fried that circuit in the ECU. This way I won't have to use an SSR. I'm planning on most likely getting a driver for both the pump and solenoid though, or at least the solenoid, but I figure it can't hurt having one for both lol. I'd like to get a nozzle to go with the solenoid but Alex told me they're not available at the moment. What's the consensus on the Snow nozzles? Are they still lower standard than the Prometh nozzles? It seems to be conflicting info in this thread regarding those.

I see differences on the outside. Might be different revisions of the same product. I doubt they would say there was a pressure relief valve and not have one.

As for pre-turbo, I'm not sure I see the point unless you have already maxed out the turbo and need slightly more compressor. The hotter the air entering the intercooler, the better chance the intercooler has to shed off a higher percentage of the heat. Adding the AI/Meth post-intercooler adds to the cooling effect instead of just taking load off the intercooler. If you run pre-turbo, you are limited to ambient temps after the intercooler, but post-intercooler spray means you could thoretically drop below ambient. Pre-turbo just seems useful for stretching the compressor map or a no-intercooler situation.

Yeah there are subtle differences, but I figured I'd see a larger cap on the front end or something like that. But yeah, I don't see why they'd lie about it, so it probably has one.

I can already see a few minds asploding … :suspect:


.

Cool video.

Anyone here adding a lubricant to their water/methanol?

that might be questionable with water in the mix, you should consider a top lube in the fuel tank instead imo in addition to the OMP or premix oil being used.
.

I'd use cutting machine coolant oil.

I don't think cutting machine oil is meant to be burned. That sounds like trouble.

If you really want to inject oil, maybe use another set of injectors? You could even incorporate it into the block! Use a small shaft to help pump and meter the extra oil? Hook it up to the ECU, and have it act on a pre-defined map just for this oil injector? This sounds so familiar...

it keeps the injectors from rusting. Volvo patented the system in 1983. they used something that looked like an EFI injection system, but it used water with machine cutting oil to keep the injectors from rusting.
they were making ~300hp from the 2.1 engine in 1983. 1.5 bar of boost. its a mess too, they kept the K Jetronic for fuel, and there is a separate computer for ignition, and then the 3rd for the water injection.

https://www.media.volvocars.com/glob...acks-of-europe
https://dailyturismo.com/dto-volvo-2...oup-evolution/

https://cimg9.ibsrv.net/gimg/www.rx7...140f1f66e7.jpg
https://cimg6.ibsrv.net/gimg/www.rx7...8912b7b1cf.jpg

just ignore that it was 40 years ago, nobody does that now in the age where stainless steel and exotic materials are abundant. Not even on the newest systems that actually have direct PWM injectors

https://usrallyteam.com/index.php?ma...x&cPath=68_347
.

80 years ago, it turns out WM-50 had oil in it too.
"MW 50 is something of a misnomer, as it is actually a mixture of three fluids: 50% methanol acting primarily to achieve optimum anti-detonant effect, secondarily as an anti-freeze; 49.5% water; and 0.5% Schutzöl 39, an oil-based anti-corrosion additive...."

https://en.wikipedia.org/wiki/MW_50#

Haha, my setup doesn't "require" a lubricant so I was more curious than anything.

With alcohol generally being harder on the internals, lots of talk about premix lately and the quantities required to keep things happy; cant help but think a lubricant would be valuable to those with high volume AI systems. I'd absolutely add a lubricant to my fluid if it would help, doesn't seem like a popular thing to do, though.

Will stick with premix and possibly a top lube for now, but if anyone knows a good soluble oil that might take up some slack, I'm in.

Think about where the AI is being injected. You'd be coating your intake in oil, likely your throttle body as well. You can open up the OMP and fix the flow rate, so you will have apex seal lubrication off-throttle, and then add premix to get to a good % on throttle. You seem to want to replace the OMP, if you want to inject oil, why not just use it?

Recently added WMI to my FD setup using generic boost switches for the time-being. Running two stages: W/M pre-turbo to specifically extend compressor efficiency/max air flow at higher boost (Stage 2); and, H2O post-IC ahead of throttle body to maintain cooler/cleaner engine internals (Stage 1).

Plan is to let the Syvecs ECU control pre-turbo compressor Stage 2 spraying to shift compressor map efficiency while stretching the EFR 8374 near-max limits as applicable. Many options for control available; PWM, linear/progressive, etc. Wondering if simple on/off control is all that is needed for this stage or do I need to match injection rate w rising boost or turbo speed sensor rpm or other available metric (EGT, perhaps?). In other words, what are any obvious indicators to let one know that the compressor efficiency map has shifted to a more favorable region as turbo air flow limits are stretched? Seems to this small brain that if just a set amount is needed to shift compressor map efficiency when operating at 20+psi, why go the PWM route or similarly more progressive strategy for Stage 2? I.e., Can I keep things simple here?

Note: Stage 1 control for post-IC spraying seems more practical for me since I have the instant feedback of IATs during injection. It's the pre-turbo stage that I'm mostly interested in the forum's feedback.

Appreciate your comments in advance!

Prometh question not covered.

From their nozzle charts you can have the same flow rate by doing this:
(1) larger orifice with smaller number of slots or
(2) smaller orifice with larger number of slots.

Which of the two would result in smaller droplets?

the smallest/most narrow x shallowest slots imo, but perhaps my ignorance is showing :dunno:

that might be the one with the most slots, but is not truly defined as such.
.

Shame no one engaged with you on this, unfortunately I don't have any advice for you either. However we seem to have the same plan. I have the AEM V3 setup with 250cc pre turbo and 1000cc pre throttle body running 50/50 water meth. I too plan to max out the 8374 with my bridgeport, and plan to play with the progressive controller on the dyno to see the effects on power and egts in the spring. There seems to be a wide range of advice on nozzle size, but the AEM chart shows 1300cc for 600hp at 25psi. Not sure how those numbers relate to a rotary but I figured it was a good place to start. What nozzle sizes are you running?

since the rotary is moving 1.3x more air per hp that’s likely a factor.

so that 600 piston hp is only equivalent to ~460 rotary hp

fuel type type will matter

generally would run the pre-turbo in the region where compressor map efficiency is falling off at the right and thus, the compressor discharge temp is going to be escalating.

Just a quich glance, but on an EFR 8374 @20 psi is going to be falling below 70% around 60 lbs/min or ~450 whp, 65 lbs/min @ 25 psi or ~480 whp. You might consider sooner to get a jump on it, you just have to study the compressor map and decide at what point past the peak efficiency to kick it in.
.

Thanks team! Yeah the increased airflow need for hp equivalency to piston engines would probably apply directly to the nozzle size needed. I can easily increase the pre-turbo nozzle as it's the smallest one. I have thermocouples pre-turbo (pre-nozzle), pre-intercooler, post-intercooler (pre-nozzle), FD UIM stock location and one in each exhaust manifold runner. So I should be able to play with the controller and pre-turbo nozzle size and see how much I can affect temps all throughout the system. Plan is to keep egt's under 950c (efr limit but should be able to keep it in the 800's) and otherwise look at air temps going into the engine and keep increasing the flow until temps stop dropping or power does. Obviously I'll be fine with some power loss for safety if the temps are continuing to drop substantially and not just marginally. I'm at 2200ft so that doesn't help the compressor, but hopefully the denser air from the WM will at least make up for that. My car really doesn't need this turbo maxed out, but if it can still put the power down then maybe the turbo is too small, lol. A LOT of testing coming this spring!

@FFR818

"As a GENERAL rule of thumb when using a 50/50 WM mixture, use a factor of 100 ml/min (cc = ml) for every 100 hp achieved; approx. 1.5 gph nozzle capacity per 100 hp. This is primarily a starting point for the 20 psi or less crowd. If using water only, go with 1 gph nozzle capacity for every 100 hp produced.

Note: These are general 'starting points' considering the variable factors of differing mfr pump pressures & nozzle flow ratings offered; more prudent to start at the minimal end, and work your way up as applicable."

I posted the above guidelines in another thread where the OP had unwittingly been spraying far too much WI, which was causing predictable internal combustion issues; hence, limiting otherwise achievable power to the wheels.

In your case, I would cut the pre-TB amount of WMI to something closer to the 400-500 ml/min range; 6-8 gph nozzle size. I'm very confident in this amount being a good starting point for you.

For shifting compressor map efficiency, I'm far more iffy on what amount of AI it takes. For my own setup, I was initially just looking to avoid the 8374 IWG surge point that typically occurs around 21-23 psi from what I've been told (special thanks to Ian/Blue TII, et al). But now I'm mostly interested in shifting compressor efficiency to max out 8374 airflow. So whatever boost that turns out to be (upper 20s/30 psi perhaps?), then that will ultimately be my max boost available on petro & AI for the random occasions that I choose to fully stretch the turbo's legs.

I have a turbo speed sensor to assist, including a Syvecs ECU, so I'll let my tuner figure out the applicable settings for "reliably" allowing max boost when applicable. But the majority of my typical wknd/highway driving will be in the mid-teens or fewer psi range based on my own driving style.

Thanks for the contribution Topolino! But this somewhat proves my statement about there being some confusion over injector size. AEM recommends just over 1300cc with their system for 600hp and 25psi, but by your recommendation it should be half of that. On the other hand, Team makes a good point about the additional airflow required for a rotary to make 600hp vs a piston engine that AEM would’ve been basing their numbers on.

After a lot of searching I decided to err on the side of caution and go with the AEM recommendation since I’m running their injectors, pump, and controller. And then I’ll do some testing on street and dyno to see how it responds. I can definitely move more of my injection to the turbo inlet if needed, but this will depend on the turbo outlet temps as I increase the boost as detonation protection is more important to me initially as I test then compressor efficiency.

my feeling is that the AEM recommendation is perhaps very conservative for the obvious reasons. The rotary application being unique, I’d tend to rely more on the true experience of rotary engine owners and what Topolino stated matches my own experience.

In your situation though, maybe giving up some performance to be on the conservative/safe side is the way to go.
.