400WHP Renesis on pump gas
How?
W/M injection with custom two stage rpm based setup (Snow Performance kit)
Brettspeed manifold
G30-660 0.83 Turbo
Built engine with low(ish) 9.7:1 CR rotors .
Engine built by Brent Curran (Curran Brothers racing). Brent has a lot of experience building and tuning high horsepower Rotaries. He is also providing some much needed support and advice. Was his suggestion to switch to W/M vs ethanol.

That's the plan at this stage .......

Failsafe strategies :
1/ AEM failsafe gauge set to cut boost if AFR goes over presets.
2/ Pressure switch ... either just turns on green light OR prevents EBC solenoid operation if no pressure in line
3/ Boost cut at preset rpm (7500 ish)
For reference: AI Safeguards - RX7Club.com - Mazda RX7 Forum

 

Saved for whatever

 

Great plan. I have a plan too. (just putting it out there)
I plan to win the biggest lottery ever in history.
Getting a body transplant. Hey,I can afford the research.
I'm thinking a body like the "Govenater" but with out the beard he has today.
(and without the heart implant.)That's a start for now. Does anyone else out there
have any "plan", if so let's hear them..

Oh Yay, The car runs just fine as it sits .............

 

Sounds legit

 

best wishes for a great result!
.

 

Who are you and how did you get access to Team's account?
Seriously though.... thanks for that.

 

Sounds like an arduino to handle wmi+failsafes would be handy. The flexibility in control logic it offers is unbeatable. You want it to sing "take my breath away" when boost needs to be cut... you got it.

 

Just spent a couple of days bench testing the W/M system (again) to get familiar with it's operation etc.
Some things I've learned...
*Don't trust that the nozzles are 'ready to go'. Noticed they come with filters not screwed tight which upsets output. Also one came with corrosion evident at the nozzle tip , obviously it had had water through it and not been cleaned out.
*Don't trust the output on the nozzles as the actual output ...test it.
*When testing with a car battery ... have it on charge constantly. As voltage drops so does output.
*Check valves will prevent siphoning and minimise 'overspray'. Important for me as I'll be putting both nozzles post throttle where they will see full engine vacuum.
*Solenoid valve will also prevent siphoning but if mounted away from the nozzle ...a check valve is also worthwhile to minimise 'overspray' and keep some pressure in the line in readiness for next spray.
*They don't tell you this, but the pumps (sureflo) have an adjustment mechanism which on it's lowest setting produces flows close to what the nozzles are specified at. Adjusted to its max. setting and flow is nearly 1.7x the min setting. This means you can get some leeway on nozzle sizing.
*IMO ...... go for smaller nozzles that you can run at higher pressure so that you get the best atomisation.
*The controller that comes with the Snow Performance kit supposedly adjusts flow according to boost but this isn't how you might think.
See below two graphs of output vs boost pressure. In the top one I set start at 2 psi and max flow at 30PSI . You would think youd get progressive linear flow all the way to 30psi ... not so. Note how you get close to max flow at around 14psi .
The bottom chart I set start at 2 psi and max flow at 14psi . You get close to actual max flow at around 7 psi !
This is valuable information as this is a street car and I don't want to be drowning the engine at lower than max boost settings.

Edit: After many months of testing and street use I have made an important observation: With a fast spooling turbo (which mine is), having a controller that ramps up flow by psi can be problematic. Reason being, at rpms where turbo will spool quickly, the increase in flow can't keep up with how quickly the turbo is spooling. Result is not enough flow and a potential detonation event.
Running on the 2-30 setting shown above produced this effect so I now run at a 2-14 or less setting as this brings in more flow earlier.

 

few sources of some great data:
https://myaudis4.com/2015/06/14/boos...w-calibration/
https://www.coolingmist.co.uk/index....product_id=130

but great care has to be used with above, the frequency is to low and can results in one cycle receiving 100% of open time, and other 2 ones 0% or mix off.. (or any alternative)
can be significantly fought with high duty cycle on the solenoid valve when the variability between cycles is very low.

Plotted bellow is PWM (time of injection) in blue, VS in orange cycles (one than another oscillating)

2000rpm - 22% of total nozle capacity requested




6000rpm - 67% of total nozle capacity requested


8000rpm - 89% of total nozle capacity requested

 

@Brettus when the flow is adjusted by pressure it is very close to (P2/P1)^0.5 (or even less like 0.4) .. ie for double the flow you need at least 4 times the pressure.

I would say you need water as a function of MAF not MAP (as map is same betwen 3000 to 8000 rpm, while MAF is >> double) .. but it is impossible to match it with PWM.

 

This is why I chose to make it two stage ... all the controller will be doing is allowing a very low boost switch point without drowning the engine - nothing more.

 

What is the PWM frequency of the snow controller ? You could maybe just add the high speed valve and controll it by snow controller, and have it maf dependent but switched on by boost

 

Was going to use the VDI wire (rpm adjustable) combined with a boost switch to trigger the second stage.

Edit : removed the boost switch as it was just an unnecessary complication. The pump is already triggered by boost.

 

Have finally made some progress so time for an update:
Firstly ...engine is back in/running and run in. Engine running well on one newish housing and another older one that was in serviceable condition. Have not compression tested but I expect there will be a difference between the two rotors. Not ideal for what I'm trying to achieve here but didn't feel like shelling out 2k for a new housing.
Had an oops moment on startup .... let's just say the gge floor is very well lubricated now! Note to self ..... always remove the plastic wheel liner when connecting the oil hose to make sure it's properly clipped in and seated!
While running in the engine I set about piping and wiring in the W/M two stage system. Let me just say I'm no automotive electrician and that job proved to be a major challenge for me. Def. wont even attempt to instruct anyone on how to do this as I could barely manage it myself.
However I do now have a fully functioning two stage W/M injection system which is ready to tune.
One of my failsafe ideas essentially ....failed. Was planning on having a pressure switch that prevented boost solenoid from operating if there was no water pressure. It did actually work but it also caused a major hiccup when going into boost. I thought that it would all activate quickly enough to avoid that ...but.............. no. So now the pressure switch just turns on a green LED which at least gives me some comfort in knowing that there is pressure. Considering having a flow switch doing the same on the second stage for additional comfort.
I still have the AEM failsafe gauge which is setup to dump boost should AFR or boost go outside of desired range. Am hoping the methanol content in the mix allows me to set the limits close enough to facilitate this.

Started to tune it yesterday but quickly realised I had a mysterious boost leak. Took me some time to locate it but was really glad I'd bought myself a stop valve for my leak tester setup so I can now shut off flow and listen for the leak without hearing incoming air from the compressor.
Turned out to be the welded seam on the plastic UIM closest to firewall. Second time I've seen one leak in this same location ...def. something to look out for!

 

It wasn't the seam at all .... look carefully where the pointer is !

 

So ..... it's tuned on 9psi with two stage W/M system. Learnings:
*Lost a little power down low but things are looking 'interesting' up top.
*My idea of using the failsafe gage as a safety is essentially ...a fail. There isn't enough of a difference with/without the W/M to get a reliable solid trigger (only seeing 0.3-0.4 AFR change)
*Boost is creeping after the APVs open to the point of being dangerous (13psi) and eliminating the effectiveness of any failsafe measures I might try to implement.
*Placing one of the w/m nozzles close to the APV branch has done something spectacular with the way those ports are flowing ... this is a problem with the boost control issue I have. But could be a benefit once I get that sorted.
*Other than those disasters ...the two stage system is working great . Feels a bit like one step forward and two backwards at the moment so am stepping away for a few days while I decide where to from here.

 

First run to 7500 highlights the boost creep issue. This was at 9 psi (reached at 3000rpm) climbing to 12.9 at 7500 and the APV ports restricted. The good thing is the W/M seemed to do it's job !

 

moar boosts …
.

 

If you want to see if W/M does anything... play with this table:

I'm tuning mine to the point where 95RON/E10 gas sees some knock retard(up to 5deg) but 100RON/E10 does not.

 

PMd you re this

 

Update : Managed to reduce the boost creep to a manageable level with a small mod to the manifold.
Now I'm trying to stop water siphoning in through the nozzle that only has a check valve on it (other nozzle has a solenoid).
In so doing I've possibly realised why W/M was not successful for me the last time I tried it. I always thought it was to do with the delay caused by the distance to the ports from the pre throttle nozzle. And further delay caused by condensation on the plastic walls of the intake.
Those things may be part of it but also; distribution within the manifold is severely compromised by the bend after the throttle. Heavier water laden air will tend to flow around the outside of the bend and therefore more of the mix will go into the rear rotor ...leaving the front rotor starved for w/m. Just a theory but I'm now more convinced that I'm heading down the right track by putting both nozzles post bend.

 

I’ll see and raise you ♥️♦️♠️♣️

https://usrallyteam.com/index.php?ma...x&cPath=68_283


royal flush here though

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


you should consider one at the compressor, size it right and you can get a lot of cooling potential without worrying about flooding the IC. The same website has options for it.
.
.

 

I wouldn't use a static flow system: where its needed most, it will be undersized (as you cant oversize on low as it will quench the combustion)

@Brettus I higly advice you run a solenoid valve (not a proportioning valve! ) to be able to get a large min vs max flow value and dial in the optimal flow vs engine load..

 

The two stage system seems to be working really well .... I'll give that a chance before considering adding complexity.
Also : quote from rice racing :
"Unless a solenoid can open as fast as a fuel injector it should not be used to "pulse" water injection events as flow is not directly related to "duty cycle"."

 

1st is, ofc they will advertise as to what they are selling.

I gave you a calculation of interference from slower frequency compared to fuel injectors. There is some oscillation but not impactful at all. and it can be much closer to optimum than 2 stage. Also the flow is surprisingly very linear as documented by many who have tested them. Those also make the pump run at lower pressure as they don't rely on the over-presurisation of the pump to increase flow.

Another is that there are two types of valves, one is open/close (like fuel injector) and the second it proportional "constant opening to set %" .. the second doesn't work as it only adjusts the pressure.

​​​​​​​