ZX-Tex

iTrader: (15)

Agreed completely on both statements above. I think you have it nailed. Very cool. In fact I could implement your injector test scheme on what I have now in the M3. Obviously I cannot use the wastegate failsafe.

Threadjack. I need to go to your other thread.

C. Ludwig

AEM and Labonte both have flow meters. Just curious if either of those fall into the above categories.

ZX-Tex

iTrader: (15)

Part of the problem is that if you use a HSV the flow rate is not constant. So the flow sensor and your PWM map to the injector have to be mapped together for good sensitivity. At least we talked about this before in a past WI thread. So the AEM or the Labonte have to somehow talk to the MS map, or, vice versa, or you need correlated maps in each, or you let the AEM or Labonte do all the control based on inputs from the MS, or....

Hornetball's leak-down test, although not real-time, is simple and seems like it would be pretty effective. Essentially that is what it is, a leakdown rate test.

hornetball

iTrader: (4)

In my mind, yes. The complexity of a sensor is directly related to it's price. Of all the basic sensors we use (pressure, temperature, flow), a flow sensor is by far the most complex and expensive. They come in basically two flavors (for liquids -- I'm excluding the air flow sensors that have become commonplace): (1) an impeller mechanism; or (2) an orifice with a measured pressure drop mechanism. The impeller type is quite delicate, while the orifice type is a flow restriction. Both types are sensitive to calibration and only work well over a range. Plus, what's the alarm point if you've implemented some type of variable flow system (like the High Speed Valve example given by ZX-TX)?

Anyway, I think an easily actuated and observed manual leakdown test is pretty much perfect for detecting a clogged nozzle.

Now, if I were to add a HSV to my system, how would I test for proper HSV operation? Hmmmm . . . .

Man, we've really hijacked this thread.

Faeflora

iTrader: (2)

You could test for proper HSV operation by pulling a nozzle and looking at the spray as you increase/decrease the pulsewidth.

BTW I think the point of the flow meter is to detect extremes. You have your normal operating range then extreme high if you bust a hose/connection, and then extreme low if you're clogged.

ZX-Tex

iTrader: (15)

You know if we had an electrical engineer around here that was into WI **cough** Joe Perez **cough** you could even go as far as to build a simple timer circuit. It would precisely measure the time between when you release the test button and when your pressure threshold switch changes state. Hell I'll bet Digikey or someone like that has a pre-built device with a built-in display. They certainly have the components.

Is this a work in progress or do you know about how long the leakdown interval is? Based on what I have seen, unless you add an accumulator, it is going to happen pretty fast I think, like in a few seconds or even much less. Even silly things like where a diaphragm pump is in its pulse cycle when you release the button could introduce a lot of error. That varies your pressure at the start of your test.

hornetball

iTrader: (4)

Ideally, it would either be automatic or something foolproof where you toggle a button and see a light.

A flow meter would have to be setup to flag on extremes with a variable flow system. My question would be how do I know I'm getting the knock protection that I've tuned for at the various RPMs and boost. Tough problem.

hornetball

iTrader: (4)

It's a work in progress . . . but the progress is quite advanced. I hope to be able to answer this question by the end of the weekend!! I'm really looking forward to it.

ZX-Tex

iTrader: (15)

Very cool. I am interested in your results.

Here is a general example of a timer. It counts the interval between the change in state of two inputs, and either input can be set for rising or falling edge. Something like this would work, except it needs to be much smaller, less expensive, and DC powered.
http://www.micronmeters.com/store/gl...stopwatch.html

tasty danish

iTrader: (9)

Figured most of you would be interested in this:

I'm not going to quote benefits or anything like that as I have not run these myself, and don't want to be a fake e-thug however, pre-turbo water injection is a big thing on rx7club.com and if you check out their auxiliary injection forum, you'll find a wealth of knowledge on the subject.

Of note, there is one member who designed his own pre-turbo, completely MECHANICAL (no pump) injection system, using a boost pressurized reservoir that is considered the hot ticket.

I can say, from my research that both my turbo rx7 (when I get it back together) and my turbo miata (when I stop whacking my pud and get around to it) will use mechanical pre-turbine injection per the rx7club model. Interesting fact, users that have done it already report that their turbo behaves a touch larger (efficiency map shifts right) while spraying, which the article posted seems to get at.

Techsalvager

I plan to make the same setup tasty danish, I already know of a miata owner doing the same thing as that with his turbo 1.6

hornetball

iTrader: (4)

OK. Here's the answer. My leakdown takes about half a second after I release my test switch. It is a very repeatable time period. My components are:

1. New Cooling Mist 150 PSI Sureflo pump.
2. New Cooling Mist check valve.
3. Honeywell/Hobbs 35 PSI pressure switch -- purchased from Snow Performance.
4. New Cooling Mist nozzle. I "think" it is 90cc/minute. They sized it based upon expected HP. I told them 200HP.

Anyway, the test sequence goes:

1. Car off, ignition switch on.
2. Confirm low water indicator is OFF (note that if low water indicator is ON, it is possible to check the lock out of the water pump using the test switch).
3. Press test switch.
4. Watch for illumination of water pressure indicator. You should also be able to hear the pump operate.
5. After water pressure indicator illuminates, release test switch. After a brief time period, water will seep through nozzle and lower the line pressure below the pressure switch setpoint. This will cause the water pressure indicator to extinguish. In my case, this takes about 0.5 seconds after I release the test switch. Your results will depend upon your components.
6. If it takes significantly longer for the water pressure indicator to extinguish, then there is a clogged nozzle.

Works like a charm. You can read about all my final installation travails as well as first blood under boost over in my build thread. Exciting weekend (i.e., ups and downs).

Dust

http://www.dsmtuners.com/forums/nitr...t-results.html

pre-compressor IAT numbers for those interested.

Godless Commie

iTrader: (2)

I employ a very simple solution to ensure the WI nozzles are in working order:

I bought an ultrasonic cleaner from ebay for $30.
I remove the nozzles once a moth and give them a thorough cleaning in a 50/50 water vinegar solution for about 10 minutes.
Takes care of clogging issues..

Dust

I go to the glasses store and stick it in the ultrasonic cleaner they have outside.

hornetball

iTrader: (4)

I do a "press to test" each morning. No clogs so far after ~1000 miles of turbo use.

johnmatt

iTrader: (8)

I'm not coming in here as any kind of expert on anything but I did some reading into Pre turbo injection and from a very lengthy thread on dsmtuners I got the impression that meth is better pre turbo and more of a water mix works better pre TB

tasty danish

iTrader: (9)

How so? Pre turbo makes up for one of the problems associate with water: atomization, and also helps with the other: IAT

johnmatt

iTrader: (8)

I don't remember all the particulars of it since it was a very long thread with alot of testing done and read it several months ago but I'll try and find it again and post some of the results

johnmatt

iTrader: (8)

Quote taken from this thread http://www.dsmtuners.com/forums/nitr...-h2o-meth.html
"Water cools better than methanol but it doesnt vaporize as methanol does. Increasing spool is done by injecting with a 2gph or less nozzle pre turbo. Methanol is preferred pre compressor due to it's low vapor point to avoid damaging the compressor blades."

Another good read if you have the time is here http://www.dsmtuners.com/forums/nitr...will-do-3.html