TurboTim

iTrader: (9)

I will buy a water injection kit, but first I have a question about the Devil's Own Progressive Controller. To the best of my knowledge (very limited) the kit does not include a solenoid to control flow, so it must control flow by varying voltage to the pump. After reading their website, it seems like they vary the duty cycle of the pump (quickly cycling 12volts on&off), starting at 15% and ramping up to 100%. If this is the case, then it may be possible to control the standard kit using the emanage or another standalone in a similar manner, using an injector output. I'd be nice to know how fast the progressive controller cycles the duty cycle (Hertz).

If this doesn't work, then can you use a fuel solenoid off a nitrous kit to control the water? They are relatively inexpensive for smaller ones (~$60). I know MSD's water injection (sea water=corrosive!) for jet skis uses what looks like a nitrous solenoid, around $90 for that particular solenoid.

I'd hate to spend another $100, find a place to mount another black box, and have more ***** to play with if my standalone or emanage can do it already.

Thoughts?

magnamx-5

iTrader: (7)

maybe pump is pretty hefty on the voltage though something like 4 amps i believe. seloniod would work but then you have the question of selonoid reaction speed etc, in relation to Duty cycle. if you could use the selonoid to control line pressure and have the system in constant spray while in boost past 3+ psi then you would essentialy vary the spray out put with pump pressure to me this is a more viable solution. I dunno but maybe if the selonoid was set for a boost progressive opening rate it could regulate such a flow and pressure. does any of that make sense? i can see where it might work but i've never heard or seen it tried.

VanMSM

Couldn't you have an ECU turn a relay on/off to control the pump? That way the ECU wouldn't have to handle the large current draw of the pump.

Ben

iTrader: (33)

a relay won't switch on and off quickly enough

Ben

iTrader: (33)

You could pressurize the water delivery system with a pump, and hold it at a constant pressure. Then use a small fuel injector to squirt, controlled by a map on the emanage. Should be easy and effective.

TurboTim

iTrader: (9)

Can fuel injectors handle water without locking up at some point? That'd be bad.

I have a 20amp solid-state relay that should switch fast enough to run the pump but probably just at idle. As RPMs increase do does the hertz the relay would have to trigger so this whole theory of using an injector output probably wouldn't work for this application.

This again goes back to the original question of how the progressive controller controls the water flow. If it does it by switching the power to the pump on and off, then it must have a relay that is quick enough. I'd also need to know how "quick" is quick (or slow) enough.

If I wanted, I could use my boost contol output (Pulse-Width-Modulated) to control a relay that pulses voltage to the pump. It has a user definable hertz that doesn't change as RPMs increase.

cjernigan

iTrader: (13)

If you look up Fast Acting Solenoids a few companies offer them.
Aquamist has one for 250 and FJO Racing sells one for around $130. They are what you would need to use to activate your pump. Most fuel injectors won't handle water, they will rust internally and like you said lock.

jhoexp

Same idea...in these old posts:

https://www.miataturbo.net/forums/showthread.php?t=6768

https://www.miataturbo.net/forums/showthread.php?t=6279

TurboTim

iTrader: (9)

Thanks.

Niether of those threads explain how the DO progressive controller actually controls flow. Dont' make me buy it and stick it on my oscilloscope damnit.

Joe Perez

iTrader: (8)

I'll try to remember to bring home a storage scope at some point this week so I can log the output, but you've basically got it. We're dealing with a DC motor, and using PWM (probably a FET switch) to vary the duty cycle.

Joe Perez

iTrader: (8)

Ok, we have our answer. Really, guys- this was an easy one. I mean, anybody who happens to have an $8,000 digital oscilloscope lying around the office should have been able to figure this out. :gay:

The two attachments each cover 0.4 seconds of operation. They were one-shot captures, triggered on the first pulse as I gently tipped into boost. Interestingly, even though the "Full" LED has never come on, I found that the controller seems to be going to 100% duty cycle much sooner than I'd expected: Somewhere around 8 PSI perhaps. (hard to see the 'scope while driving.)

The traces are a bit noisy since I took these with an inductive current probe, rather than a direct voltage probe. Just pretend the rise times are instantaneous and the tops are square.

Oh, and for those who care, it looks like the average "on" current drawn by the Shurflo pump is about 3.5A.

thepope540

Tim, fuel injectors cannot handle water effectively. I looked into this before I started my solenoid project that you chimed in on. I knew that stock injectors would lock up with water so I called RC and one other company (don't remember who) that make and sell injectors and nobody makes anything compatible with H20.

The only way to spray water is to start on fuel, transition to water, and then switch back to fuel before shutting down. A similiar process is used on some drag engines running crazy fuels that the injectors cannot deal with long term.

devilsown

I guess i can chime in.

Our controller does use pwm to control the pump. The controller is capable of a lot more then whats physical possable then the heavy pump can. We will have systems out that will have a very high degree of control over output. Its going to be considerable more money, but this is high end system. Our controller as it is capable of 10a intermidate. Which the pumps are capable of pulling on full load of a 5v input.

Most injectors can't handle meth either.

Relays are very slow. Solenoids is what your going to want to do what your wanting it to do. We should have some out soon. But your going to want something thats 3 watt or less and can pull in .5ms

cjernigan

iTrader: (13)

Well make sure you keep us informed. I'll eventually want to look into controlling my DO basic kit with meqasquirt so one of those solenoids would be nice. Any idea on price yet? FJO has them for $127 and aquamist for like $230, you going to be any lower than that?

OT: What's the word on the new nozzles that screw in from the outside, haven't heard if you were getting closer to a source for these.

TurboTim

iTrader: (9)

Excellent responses guys! Thanks thepope540, I'm glad I don't need to call RC myself Thanks DO, I'm glad to see you here. I'm sure a lot of us are interested in your solenoid.

Thanks Joe for doing the scope work! It looks like the controller pulses around 38Hz (.026 seconds). I think I'm using 35Hz for my GM solenoid-boost controller. Hmmm. I doubt the PW driver in the AEM EMS can handle the current of the pump. I have a solid state relay lying around, I wonder what the response time is. 1ms? That could work. But to have fancy controlled water injection or electronic boost control...I should just spend the extra $100 on the progressive controller.

Or $127 on the FJO solenoid. Time to research that.

Joe Perez

iTrader: (8)

Make sure it's rated for DC load. Some solid state relays (primarily the big high current ones) are SCR-based, and this cannot switch DC.

If you're feeling ambitious, you could simply use a big N-channel MOSFET as your driver (assuming the PW output of the EMS is active-high). But yes, having the DO controller makes things easy, and besides, it looks very cool mounted below the radio.

TurboTim

iTrader: (9)

Starting to get over my head Joe, but thanks. My last circuits class was 5 yrs ago, oops. I do know the relay is for DC, and around 20 amps @ 12vdc.

brgracer

iTrader: (14)

I can't find the thread anymore, and it doesn't make sense since IIRC the cost was pretty prohibitive, but someone claimed that they found a fuel injector that could handle meth/water that you could have used the emanage/EMS/MS to control, but it cost ~$250 so kind of defeated the purpose if you were looking to save $. I'll try looking for the thread/injector again.

TurboTim

iTrader: (9)

thanks tom.

I went to FJO's website. Their solenoid looks jsut like a fuel nitrous solenoid (just 3x the price). FJO makes teh progressive nitrous controller for nitrous express. After searching Nitrous Express' site I believe this controller is a constant speed PWM controller at around 15 Hz. Soooo i'm thinking FJO's "water injection" solenoid (if it is indeed a modified fuel solenoid) would not work being driven by an injector output because it's response time is no where near fast enough. Something makes me think aquamist's is the same way.

Kelly

iTrader: (6)

Aquamists is not the same. Aquamist is owned by a much larger electronics company. Thats why all their gadgets and parts are all so bad ***. Their solenoid is the real deal and probably made in house by them.