Rotary Water Injection Set Ups

[atl93fd]

Greetings All,

Is anyone here running the aquamist 2d on their 93-95 RX7 (fd3s)? I'm currently running distilled water which kicks in @ 10 lbs. I'm using solely for anti detonation purposes. The rotary is extremely sensitive to detonation and WI can help me prevent a rebuild or new engine.

My question: Should I notice LOWER intake temps when water is injecting into the intake? Under normal conditions I don't notice anything.

When I circumvented the solenoid, Water Injection ran non-stop and I noticed my intake temps drop as soon as I accelerated from a stop light.

I'm wondering if I need to, turn up the pressure on the pump, have the WI kick in earlier then 10 lbs, or change the nozzle.

Any advice is helpful. If anyone can host them, I'd be happy to post pics.

[atl93fd]

I didn't get a lot biting on my first post. I'll try again. I have a 1993 RX-7, currently running around 14 lbs of boost. My water injection doesn't kick in until 10 lbs. I believe I'm making about 330-350 rwhp, running 950 cc injectors all around.

I don't notice my intake temps drop when injecting. I also can go 4-5 fill ups conservatively without a noticeable reduction in the amount of water in my resevoir.

I'm running the middle sizednozzle of the 3 coming with the aquamist 2d setup. I'm also running 6bar on the pump itself.

I do have an intercooler and am injecting after the IC, at the intake elbow. My pump is located in the rear hatch, next to the resevoir.

Any input on why I wouldn't notice my intake temps drop or notice the water levels reduce in my resevoir?

[robbilau]

I'm not familiar with the RX-7 but a few basics:

I assume the air temp sensor is after the water injection?
Have you done a test run just holding the jet out the window so you can confirm flow?
What duty cycle are you running in your injectors at max? Perhaps the duty is too small for significant injection?
The middle jet will be a 0.6 mm jet pushing 250 cc.min at 100% duty. If you're running say 65% duty on your injectors and the 2d matches that duty cycle you're getting a max of 162cc/min of water. That sounds like too little water to me. Richard will need to confirm the 2d behavior though.

I'd start the water below 10 lbs, say at 7lbs and upgrade to the biggest jet you have.

[Richard L]

My first thought is that you are not injecting enough water.

Need to know your duty cycle of your fuel injectors. The 2d follows the fuel injectors duty cycle so low fuel fuel equals low water flow.

It it possible to measure the voltage that come out of the purple wire ( fuel duty cycle monitor), please let me know what is the peak readings you get. It will be between 0-5V.

I await your reply.

[hotrod]

Once you confirm you have flow, you may want to log your boost if you can.
At least have a passanger ride along and watch a boost gauge.

You may spend so little time above the 10 psi turn on point that the system never gets a chance to spray a significant amount of water. As mentioned above, you might want to pull back your trigger point to a lower boost.

A 10 psi trigger on a system running 18 - 20 psi boost will be on for a noticiable period of time. In your case, especially if you spend most of you boost time in lower gears you may only be above 10 psi for a second or so in each of the lower gears.


By way of example on a drag strip pass, my WRX red lines low gear in about 1.7 seconds, and 2 second gear only lasts about 3 -4 seconds, 3rd gear is about 4+ seconds and I shift into 4th just a second before I cross the finish line. I probably spend 4 -5 seconds over 10 psi .

There is also the possibility that your boost gauge, boost control system and the boost trigger switch are not in agreement. If you boost control system is actually only giving you 12 psi boost and your trigger switch is a bit high you will have nearly a 0 duty cycle. The second issue is where are you tapping boost. Your sense point may never get above 10 psi even though the boost control system thinks the engine is at 14 psi.

Larry

[atl93fd]

Awesome replies guys, Thanks.

First off, I'm running the .8 mm nozzle. I belive the kit came with .7, .8, .9 (and I'm missing the .8)

Air temp Sensor is after the WI. I have done a test to confirm flow. At one point, I disconnected the ground and was injecting water 100% of the time for 5 or 6 miles. As stated, when I pulled away from a stop light, I immediately noticed temps dropping. At 7500 RPMS, I reach about 93% of my fuel injector duty cycle. Most of the time (around town and in traffic), I am < 50%. I probably hit 65% duty cycles around 5500 RPMS (again, rough guess but not too far off). Additionally, I notice I spend very little time boosting above 10 lbs. I spool up to 1 bar @ 7500 RPMS and reach 10 lbs of boost VERY quickly. So, under a hard pull, I probably have longer sustained boost then most (quicker spooling sequential systems and higher redline). Having said this, I'm not on the throttle boosting like that as much as I would like.

Richard, I'll get back on the voltage that comes out of the fuel duty cycle monitor. I'm sure I have this info in my PFC, somewhere.

Thanks for the help guys. I'm a sales bloke. I'm not a gearhead by any means yet. However, I'm certainly an enthusiast and willing to learn.

[rxstephen]

Hi,

Here's my experience with my FD3s RX-7 and using a mapped water injection.

I have the injection just after the intercooler, it is currently set to come on at about 6psi and is mapped to run about 30psi (water pressure) at 2000 rpm and 90psi at just over 5000 rpm. I can't remember what nozzel I'm running at the moment though.

I set these values based on winding it up untill I got miss fires and them wound it back a little (nice and scientific).

I have a 1 litre water tank, and go through it about once every 3 weeks with general driving. Typically three or four short WOT runs each day.

When on the race track, I go through 1 litre in approx 12 to 15 minutes full on racing.

Looking at manifold temp, there is little noticable improvement with short WOT runs, this is because the stock sensor is slow reacting and gets heat soak from the plenum chamber. It's pretty usless really and I'm considering moving it to the elbow. On the track if I don't have WI turned on I get heat soak after two laps and the power drops off. With WI turned on I get a constant (about 45 deg C) manifold temp, and repeatable power for every lap.

I'm running an ARC intercooler that is only slightly bigger than stock, plus I'm running stock turbos at 1 to 1.1 bar, so I'm asking a lot of the system and generating a lot of heat. The WI allows me to run at the higher temps while also leaning out (currently between 11.8 to 12:1 AFR, but I plan to lean out some more yet).

My last dyno was 334HP ATW, at 1.1 bar boost, I'm running stock fuel injectors peaking at about 93% duty. With some fine tuning I'll get a little more out of it, but I suspect not much more :)

Cheers
Stephen

[mx5]

Awesome replies guys, Thanks.

First off, I'm running the .8 mm nozzle. I belive the kit came with .7, .8, .9 (and I'm missing the .8)

Air temp Sensor is after the WI. I have done a test to confirm flow. At one point, I disconnected the ground and was injecting water 100% of the time for 5 or 6 miles. As stated, when I pulled away from a stop light, I immediately noticed temps dropping. At 7500 RPMS, I reach about 93% of my fuel injector duty cycle. Most of the time (around town and in traffic), I am < 50%. I probably hit 65% duty cycles around 5500 RPMS (again, rough guess but not too far off). Additionally, I notice I spend very little time boosting above 10 lbs. I spool up to 1 bar @ 7500 RPMS and reach 10 lbs of boost VERY quickly. So, under a hard pull, I probably have longer sustained boost then most (quicker spooling sequential systems and higher redline). Having said this, I'm not on the throttle boosting like that as much as I would like.

Richard, I'll get back on the voltage that comes out of the fuel duty cycle monitor. I'm sure I have this info in my PFC, somewhere.

Thanks for the help guys. I'm a sales bloke. I'm not a gearhead by any means yet. However, I'm certainly an enthusiast and willing to learn.

Which ground did you disconnect? Disconnecting the ground from the air pressure sensor will turn on the W/I all the time but it will still mirror what the fuel injector is doing. The only thing it does at low IDC is to fill your intake with water!!! It doesn't do any mist. So you're not achieving anything with this - what you will do is full the intake manifold with water at cruising and may destroy your engine.

What you want with the 2D is to ground the brown wire (without disconnecting it from the HSV). Then the W/I will activate at full power. This is what you need for your experiment. Also probably it is not a good idea to turn on the W/I at full power when the throttle is closed - so just mount a switch somewhere on the dashboard.

[atl93fd]

I was doing some looking around and inadvertently pulled the wire out from the female connector (the wrong ground you were referring to).

I called my engine builder (who also installed the aquamist), and he told me it wouldn't hurt to run to home depot and back. Went to home depot, spliced the wire, crimped on the new female connector, and was good as new. So it was somewhat "unintentional".

[atl93fd]

This thread may be dead, but I thought I'd throw it out there as I'm noticing at least 3 rotorheads here now.

My update: I hit the dyno last week, put down 325 rwhp @ 7200 RPM. The torque curve was nice and flat and I reached my near peak torque very early on. Fuel injector duty cycles were around 93%, I have 4 850 cc injectors. Water Injection came on just after 10 lbs of boost, my peak boost level was .91 or about 13.5 psi. Air Fuel was pig rich all the way through redline. At peak rwhp on my 4th run, A/F was 10.5/1 NO higher than 10.8 to 1. I did have a boost leak at the elbow and the IC. I'm using the stock solenoid and my engine management to control boost.

I still didn't notice any air intake temps drop. I THINK this is because:

I'm running sequential turbos with a 12-8-13.5 pattern, transition occurs around 4200 RPM. Water comes on, goes off, then comes on through redline.
Also I am using the stock sensor which is SLOW and the manifold probably heatsoaks.

Should I notice a dramatic decrease in intake temps while running WI?
A second question, how lean can I SAFELY go without detonation?

[Richard L]

atl93fd,

Reading through your posts again and I may have a probable idea as to why the WI is not reducing your inlet temperature.

Almost 99% of the high flow fuel injectors has a coil impedance between 2.5ohm (flowrate up yo 560cc/min) and 1.2ohm (550cc+). I believe you 850cc/min injector may have the 1.2ohm coil.

The 2d can only reliabily detect coil impedance bwteen 16ohm down to 2.5ohm. I suspect your 2d is not reading the duty cycle up to 93%.

The only way I think you can go forward is to drive the 2d directly from one of the spare outputs from your controller. I assumed that you are using a third party ECU?

please update me.

[atl93fd]

Richard,

My understanding is these OEM injectors are high impedance, around 14 ohms. The top feed injectors, which I do not have, are around 1.5-4 ohms. I think the WI is acting correctly. I'm uncertain as to the flow rate, but can tell you with the .9mm nozzle, I had water squirting straight up through my boost leak where the elbow met the throttle body.

A tank of water can last me 4-5 weeks thought which brings about suspicion.

It sounds like i SHOULD be noticing a drop in intake temps while injecting. However, I don't notice a difference between stock and go traffic and heavy boost in my temps. This could be because my stock sensor is located underneath the UIM or just plain sucks.

There are times, inconsistently (I haven't been able to repeat it) where my intake temps all of a sudden drop 10-15 degrees C and then almost immediately rise. Most of the time, my intake temps are ~ 40 degrees C..

I am using a third party ECU. Apexi Power FC, a popular unit for the RX-7
Does this sound as if things are working correctly?

[Richard L]

The simplest way to know if the 2d is detecting the IDC or not is by connecting a digital volt meter between the purple wire and ground.

The volt meter range shold be set to volts. 100% iDC should read ~5V and 0% IDC =0V.

Please check the peak readings and let me know.

[hamish]

Hi atl93FD,

Another rotorhead here.

You are right in that the FD RX7 injectors are high impedence. I assume you've got another pair of secondary injectors in the primary position which is a fairly common mod.

The stock FD air temp sensor is pretty useless. I'm not sure if it's the age or the design, but they don't seem to move much. I install Haltech E6K's on FD's and the difference in response is huge. As soon as you come on boost the air temp goes up (that is until the water injection kicks in). The design of the sensor is different with the stock one being a solid core whereas the Haltech ones have the thermocouple "open" inside a protector with holes in it. I think this might help.

In saying that it sounds like your water consumption might be a bit low though. Usually on a stock FD you can't even touch the intake manifold it's so hot (after some "enthusiastic" driving). After installing water injection they run as cold as....well...something that's real cold. Nothing like seing a turbo/manifold glowing so hot that you can just about see through it and the intake is still cool to touch!! If your intake is still hot then you need more water.

Intake temperature is a handy way to tune the water injection. The idea (in my opinion) is to try and keep the temperature as smooth as possible. Not too hot, not too cold (too cold a mixture can have adverse affects on fuel mixing properly). This makes the car easier to tune as it's not jumping around the compensation maps so much, and makes the HP and A/F ratios very consistent.

Hope this helps

[Warren_from_PINZ]

Yeah, I agree with Hamish (and others). Sounds like you're not getting enough water (hot intake and not using water). I'd start by bringing it on at a lower boost level, say 5 psi or so. You should be able to get your intake temperature down to 40?C or lower. Then try bigger nozzles or something else that gives more water.

[atl93fd]

Thanks again for the replies. I don't think I will do anything, as I suspect my thermosensor is a piece of crapola.

Lately (more recently), I'm seeing it drop..from 40-30-20 to 17-15-13-0, then NEGATIVE numbers, -5-10-15, then back to ambient. I'm guessing the sensor is bad.

[DuMaurier 7]

another rotor guy here !!,
I agree with most everything said before , I use the Haltech E6K with the sensor in the stock location and it reacts to everything , from idle , stop and go traffic up to 15psi boost (so far he, he ) without any problems . I also think you aren't injecting enough , keep upping the amount till the car begins to complain then back off a bit. You can also safely lean out your ratios to 13: 1 with the wi , but remember the leaner you go the smaller the safety margein there is if something goes wrong !!. I personally don't go any leaner than 12:1 with good results . You can also get more power if you advance your ignition , but I didnt go too far ,mine is somewhere around 13deg , its always safer to be concervative wit a rotery , they still make the power wit less boost !!. Remember timing requirements aren't like on a piston engine , loads of advance isn't necessary , once yoy find the spot (correctly on a dyno) further advance will result in less HP gain and more risk of damage due to detonation .
I built a potential 600-650 bhp motor and , I am uppin to three nozzles !!!, one 0.9 (the clear one ) , one 0.7 (?the black one ) and a 0.5 (? the green one ) , the VE on a rotary is a lot higher than a piston motor so it would require more water .

[maximtaylor]

post removed - stumble not caused by wi - doh!

Hello btw :)

Max

[luv94rx7]

Just joined today, my 2D kit was installed last year by Steve Kan.

I have it kicking in at 5psi. I have a 5 gallon fuel cell in the back for the purified water. No problems detonating yet at high boost and 93 octane fuel. I'm using the .9 mm injector.

Ken, 60 years young
'94 white, pep, red leather,
mods: KDR street port & polished, 3mm Hurley racing seals, Hayes upgraded coolant seals,
Power FC ecu, Commander, Datalogit,
Tuned by Steve Kan,
XS T04e single turbo kit,
ASP(M2) Large Racing SMIC with 1630CFM Pull Fan,
Aquamist 2d water injection kit,
Racelogic Traction Control,
Profec B(15&25psi),
RP Racing fuel pump,
1600cc injectors,
MSD 6A(3),
Pettit ss resonated MP,
Pettit ss cat-back,
Pettit short shifter kit,
LaBreck's bushings,
Evans Coolant
http://nopistons.com/luv94rx7.html

[DuMaurier 7]

I am currently running a two staged system , the first one becomes active at 0.0 psi up and the second comes on at 16psi . I am using the 1.0mm nozzle in the first stage spraying into the turbos suction and the second 0.9mm one spraying downstream the IC . I am injecting a bit too much now , however I did notice my exhaust temp drop from 1400 deg F to 1000 deg F with the water injection so I am quite sure that the intake temperature falls too . I have a 5 liter reservoir and have to fill it like once every two weeks , the frequency is totally dependant on your driving style. I think you need to increase your delivery by increasing the water pressure since you already use a fairly large nozzle , increase till the motor "complains" then reduce it a bit .

[Scuttlerx]

Hi all,

Im a fd owner and newbie to the board.

Im about to step into the world of WI after hearing some great things about it, seems well suited to rotarys.

Hoping to go for a setup like yours DuMaurier, injecting pre-compressor and post intercooler, having been toying with the idea of ditching intercooler completely but think i'll see how it goes first.

Are you running an aquamist system? Im thinking of going for the 2s, but do you think the pump will be able to support 2 jets or will I need another pump? im reading conflicting things so not sure as to the pumps capabilities

After doing the math its look like i'll be needing to inject a fair amount of water, between 252cc and 378cc/min. When injecting pre-compressor does a certain amount of the water get, for lack of better term, 'used up' allowing more water to be injected from the 2nd jet without the engine bogging down?

Sorry for all the questions, but like i said, im a newbie :D

so far on the rex im running a power fc, downpipe de-cat and cat back, induction kit (heat shield needed :smile: ), and soon to have some 1200cc secondaries in there, with the boost set to 1 bar my IDC is hitting 96% so i need to get that down.

cheers

Andy

[DuMaurier 7]

Dont think you should get rid of the intercooler , let it do its job , you can only get a cooler charge with it . You can get almost all your questions answered in the Aquamist section , If you need a pair of 1300 injectors , let me know . :!:

[Richard L]

Hi all,

Im a fd owner and newbie to the board.

Im about to step into the world of WI after hearing some great things about it, seems well suited to rotarys.

Hoping to go for a setup like yours DuMaurier, injecting pre-compressor and post intercooler, having been toying with the idea of ditching intercooler completely but think i'll see how it goes first.

Are you running an aquamist system? Im thinking of going for the 2s, but do you think the pump will be able to support 2 jets or will I need another pump? im reading conflicting things so not sure as to the pumps capabilities

After doing the math its look like i'll be needing to inject a fair amount of water, between 252cc and 378cc/min. When injecting pre-compressor does a certain amount of the water get, for lack of better term, 'used up' allowing more water to be injected from the 2nd jet without the engine bogging down?

Sorry for all the questions, but like i said, im a newbie :D

so far on the rex im running a power fc, downpipe de-cat and cat back, induction kit (heat shield needed :smile: ), and soon to have some 1200cc secondaries in there, with the boost set to 1 bar my IDC is hitting 96% so i need to get that down.

cheers

Andy

I missed this post. If you need quick answers, post it to the aquamist section first I always check it section first.

If you require a flow rate between 252 - 378cc/min you have two choices, either use the Aquamist pump with a priming pump - it will get you to 580cc/min.

Shurflo pump is the other choice, it wil not be easy to control as it flow so much water even the pump run very slowly. Ideally if you can run the Shurflo pump continuously- bypass the excess water into the tank via a pressure regulator - some Shurflo pump can internal by-pass in built. If you go that route, make sure that the Shurflo pump do bnot exceed 65C as the in-built thermal switch will switch the pump off. I have run the pump continuously and after an hour ot so under the sun, the pump switched off, need to disconnect the power and re-apply to get it restarted.

I have tried many time to intergrate the two systems together (aquamist and shurflo) but was unable to reliably detect a block nozzle condition - if this is not too important to you, I would go for the Shurflo and some Aquamist MF2 controller pulsing a HSV valve.

[DuMaurier 7]

Hi all,

Im a fd owner and newbie to the board.

Im about to step into the world of WI after hearing some great things about it, seems well suited to rotarys.

Hoping to go for a setup like yours DuMaurier, injecting pre-compressor and post intercooler, having been toying with the idea of ditching intercooler completely but think i'll see how it goes first.

Are you running an aquamist system? Im thinking of going for the 2s, but do you think the pump will be able to support 2 jets or will I need another pump? im reading conflicting things so not sure as to the pumps capabilities

After doing the math its look like i'll be needing to inject a fair amount of water, between 252cc and 378cc/min. When injecting pre-compressor does a certain amount of the water get, for lack of better term, 'used up' allowing more water to be injected from the 2nd jet without the engine bogging down?

Sorry for all the questions, but like i said, im a newbie :D

so far on the rex im running a power fc, downpipe de-cat and cat back, induction kit (heat shield needed :smile: ), and soon to have some 1200cc secondaries in there, with the boost set to 1 bar my IDC is hitting 96% so i need to get that down.

cheers

Andy

I missed this post. If you need quick answers, post it to the aquamist section first I always check it section first.

If you require a flow rate between 252 - 378cc/min you have two choices, either use the Aquamist pump with a priming pump - it will get you to 580cc/min.

Shurflo pump is the other choice, it wil not be easy to control as it flow so much water even the pump run very slowly. Ideally if you can run the Shurflo pump continuously- bypass the excess water into the tank via a pressure regulator - some Shurflo pump can internal by-pass in built. If you go that route, make sure that the Shurflo pump do bnot exceed 65C as the in-built thermal switch will switch the pump off. I have run the pump continuously and after an hour ot so under the sun, the pump switched off, need to disconnect the power and re-apply to get it restarted.

I have tried many time to intergrate the two systems together (aquamist and shurflo) but was unable to reliably detect a block nozzle condition - if this is not too important to you, I would go for the Shurflo and some Aquamist MF2 controller pulsing a HSV valve.

This is what I did , yes the blocked jet feature will be lost but , you can include a DDS2 to restore some safety and security .

[Richard L]

The DDS2 may be a solution for a fix-rate injection system but will be very difficult to diagnose a fault on one jet with vary amount of water when fed from a PWM controller.

The triggering window can either be set wider or narrow but it is not easy to set the whole operating range , say between 200cc to 300cc/min is the window width, the PWM controller could easy vary the flow rate betweenm 50cc to 400cc/min. In this applciuation, how does it determine if the jet is block or just low duty cycle?

The DDS2 was desogned to work with system 1s (fixed rate) and all other systems with variable delivery rate has their own diagnostic circuitry, able to detect a blocked jet in a wide dymanic range.

Consider this block jet problem is not an easy one to overcome when changing to a aquamist/shurflo system.

[DuMaurier 7]

I think on a PWM system , the lower limit of the DDS2 's range would be more important , I think you have to tune your system to operate safely within the pre set range at the desired boost (not too much ignition advance and not too lean on the a/f ratio . I have two jets so its hardly likely that they would both plug at the same time.

[many_107060]

hey I'm new to WI planning to buy a new kit for my fd3s : single turbo no cat new engine port, PFC looking to run 12 to 18psi ( 400 to 450WHP) of boost as long as is safe anyone had any recommendation of kit and jet size I got 800pry and 1650 secondary injectors. Thanks

[cewrx7r1]

You need to run around 20% by weight water to fuel. Read the complete thread that Peter "Rice Racing" posted on his FD build.
http://www.aquamist.co.uk/vbulletin/showthread.php?t=1590

[indio84]

Let the wi come on earlier, let say 7psi. And put the biggest nozzle.

[drtyr3x]

I have a series 5 fc3s rx7 and have bought the hfs3 and summer kit. I have a lot of fuel supply including 1000 primary and 1700 secondary injectors and an running a 750hp in tank pump to a surge tank with a Bosch 044 pump. I've got a to4z turbo and am running a microtech computer. It is on just normal fuel and will be guaranteed 10s after I get it into the shop to Bridgeport it(without the aquamist kit). Any info would be much appreciated. Thanks Matt

[Richard L]

Matt,

You have a total fuel capacity of 5700cc/min.
For M50/W50, you should be injecting ~900cc/min. Install the 0.9 and 1mm jets, as close to the exit of the intercooler as possible.