Check your nozzles!
 

Recently, I'd noticed that my rate of consumption has been down quite a good deal. IOW, I haven't had to fill the tank for a while. Did a pressure-test on the system and found that the pump was operating normally, so I pulled the nozzle out of the inlet pipe to inspect it.

Now, we all know what a normal, healthy nozzle looks like on the inlet side, right?

http://img30.imagefra.me/img/img30/2...7m_1534c19.jpg

Right, it's got a cute little filter screen, made of stainless steel, around the inlet to prevent foreign matter from clogging the delicate little hole in the end. Well, here's what mine looked like:

http://img31.imagefra.me/img/img31/2...0m_d07579e.jpg

WTF, right? Where did the screen go? After a brief search, I found most of it at the bottom of the nozzle assembly, clogging the hole that it's supposed to be protecting:

http://img30.imagefra.me/img/img30/2...1m_4cda2e1.jpg

Nice, eh?

This particular nozzle was purchased from one of the major kit vendors, and has been installed on the car for about a year and a half. I've been running about 1/3 methanol to distilled water, which is certainly not an excessive amount. And yet that little screen simply disintegrated. Can't imagine how long it would have lasted were I running 50% methanol, as is the case with the "boost fluid" sold by most of the WI kit companies as being compatible with their systems.

On the plus side, I hadn't re-mapped my ignition table since moving from CA to FL, and apparently that 93 octane gasoline is some pretty good stuff as I hadn't been getting any knock. On the minus side, this was the one failure more that my failsafe system couldn't detect, and of course, it's one that I never expected to happen.

Needless to say, I'm done with Meth. From now on, it's pure water. That, and apparently I need to advance my ignition a bit more. :D

???Methanol doesn't eat steel??? Did it rust or what?

Wow. Strange.

Methanol does have a corrosive effect upon regular steel. Stainless is supposed to be impervious, though just how impervious depends upon the grade used. One supposes perhaps that these nozzles are made with the cheap stuff.

Wow, dat iz teh gayz :greddy:

one more reason i dont use nozzles with screens good catch man im glad you saw it before bad stuff happened.

Wow.. gonna check mine tonight!

Joe, on a related question, I have been looking for a flow switch as a fail-safe guard for the WI. I think if a I ran a flow switch in series with a pressure switch (on the pump outlet) that it would guard against just about any type of WI system failure:

Line rupture or leak - flow, but low pressure
Clogged line or nozzle - pressure, but little or no flow
Low or empty tank - no pressure or flow

In your quest for developing the ultimate WI system I am wondering if you found any small, inexpensive flow switches that toggle at low flow rates, like 3-6 GPH. Not a meter, just a switch, like a paddle switch type. I have been looking, albeit somewhat casually, and have not found anything yet.

I want to redo my WI failsafe to use flow and pressure monitoring on the pump outlet, and have it trigger timing retard on the ECU. I know how to do it all, I just need the flow switch.

Thanks in advance.

why do you even run meth? meth is good for substituting for fuel but water is WAY better at cooling

Damn. I knew it would promote rust but didn't really think it would eat steel. Well, I learned something today.

I've looked around, but haven't found any simple (and affordable) flow switches that will be effective at the low flow rates we use. For a system which uses any kind of progressive control, you'd really want a flow meter anyway, as you need to be able to discriminate between a partially blocked nozzle and the normal reduced flow at lower injection rates.


One word: Bandwagon. Everybody else was, there was lots of hype, so I felt I needed to. Besides, we can't leave Tom with a monopoly on the use of plenum-injected fuel as a coolant, now can we? :D

In the time since, I've been doing a fair bit of reading which indicates that water, by itself, is a highly effective anti-detonant, and I simply never got around to purging the system. It's done now, and it's going to stay that way.

Anybody here know why all those smart fellows in white coats even bothered with methanol back in the 40's during the heyday of liquid-injection research? Antifreeze. They tested a whole bunch of different alcohols, and methanol just happened to be the cheapest and most widely available that they could find to stop the water from freezing up inside the tanks and lines on the aircraft when they were parked on the ground. (During flight, waste heat from the engines was used to warm the fluid.)


Corrosion + pressure differential + flow = the damn thing just fell apart. I found several semi-coherent but corroded bits of the screen in the bottom, and one little sliver of wire that was partially protruding out the hole. I'm guessing that during the initial stages, a lot of the smaller bits just shot straight through the nozzle and were inhaled by the engine.


I'm still running a filter screen on the new nozzle, I'm just not going to pump corrosive liquid through it any longer. We'll check back in another year and a half, but my feeling is that this one will probably fare much better.

Also, a note on pressure switches.

I had previously been using the switches that Devil's Own sells, the ones that look like this:

https://www.alcohol-injection.com/im...e_switch_s.jpg

I've had two of those switches fail on me. Both times, the plunger locked in the fully extended position, forcing the switch on all the time. Not the way you want a failsafe switch to fail... Figured it out when I noticed that changes made to my Spark1 map weren't having any effect- the ECU was being forced into Spark2 all the time.

I am now using a genuine Hobbs brand switch, which looks similar to this one:

http://img27.imagefra.me/img/img27/2...lm_e1a0d3b.jpg

I got it on eBay for about $15, so it's both cheaper and appears to be much more robustly constructed. There are dozens of different models and most are adjustable within a wide range by removing a rubber cap on the back and turning a screw with an allen wrench. I set mine on the bench (using compressed air and a regulator) to close at 35 PSI, giving me a setpoint of 20 PSI above system pressure (you have to account for MAP pressing against the outlet side of the nozzle) which should be a reasonable turn-on point.

Boost Pressure Switch - For Turbo Water Injection hobbs:eBay Motors (item 230364469681 end time Aug-12-09 15:11:12 PDT) i use these nvr had a problem to date.

Damn it, I have TWO of those. One to trigger the WI, one as the failsafe switch. Sticking plungers FTL. That might be a motivator (among other things) to go to Adaptronic controlled progressive WI. That or use the switch you pointed out. Thanks for that BTW. I have seen those before, good to know they are working well for you.

My water consumption rates have not been horrible, so I may just stay with the simple single stage on-off system for now. Really though, to do programmable progressive WI, all I need to do is go buy a less than $100 IGBT/SSR (packaged, with supporting circuitry and terminals) and wire it to the Adaptronic. Done. The Adaptronic has fully programmable PWM outputs, just need the IGBT/SSR to provide enough power to the pump.

Also, good point about the flow switch. That would not work for progressive WI. I had considered that. A flow meter would require some feedback monitoring to compare commanded versus actual flow to work as a fail safe, and that is beginning to get complex/expensive I think, unless you have any ideas.

Even for single-stage on-off WI, a flow switch would work best if it was matched to a particular nozzle, or at least within a small range of nozzle sizes.

Ultimately, there is no inexpensive and/or easy substitution for periodic nozzle inspection I suppose.

Just ordered one...exactly what I'm looking for.

joe is there enough flow velocity / pressure drop in there to cause cavitation in the screen? perhaps it simply aerated itself to death.

http://coolrulespronto.files.wordpre...izzathehut.jpg

Not super cheap but how much did you spend on your engine and is it important to you?

IFS-30C - Injection Fail Safe - Combo

Cavitation? I'm pretty sure that the water in those lines is moving at sub-sonic velocity. And there's really no question that the debris which I fished out was seriously degraded beyond what mechanical stress would cause. I'm sure that the force of the water moving through the screen was a contributing effect, but without corrosion weakening the screen, I doubt seriously that it would just distintegrate on its own.

I have to imagine that exposure to methanol may have played a role in the failure of the switches as well. I have always mounted them up high on a long line to prevent liquid from entering them, but there's not much you can do about vapor working its way through the line and entering the switch. I'm just hoping that the newer, more robust switch, along with an absence of methanol, will improve its chances. Hell, this system only has to hold together for maybe another 6 months and hopefully it'll be gone altogether.


Periodic inspection is a good idea. Not running corrosive chemicals in your system is another.


The gadget that Thymer posted is a good idea so long as you have a controller capable of reading and responding to it. I intend to implement flow monitoring of that nature in my next system, just wish I could figure out who the OEM supplier of the flow metering device inside that box is. I can't find a good source of inexpensive flowmeters right now.

This is not interfaced with any controller, that's the point, it's standalone. The box looks for flow. When there is flow it outputs 12V which triggers a solenoid. This solenoid then allows your boost controller, MBC, whatever to raise boost pressure above base.

what is the possibility you accidentally made a methanol fuel cell and it just ate the stainless?

I don't have meth on my Miata but so on my Stang. I don't use and saftey switches but I did move the air intake temp sensor in line with the meth so it directly cools the sensor and tune to a temperature and if its over that temp it pulls timing. Its fail proof because you are not depending on floats and such also of lines get clogged for what ever resin it does not cool sensor it pulls timing.

I'm running straight 100% meth. Put some meth on your finger and some water, tell me what feels cooler???

Hmmm. The nozzle was connected with plastic lines, through a plastic pump head, to a plastic bottle.

Then it's worthless. You will have the same amount of flow at redline at 16 PSI with a partially clogged nozzle as you will with a perfectly good nozzle at low load. The controller needs to be able to read actual flow so it can compare it to predicted flow.

brass + stainless + methanol + water = obscure forms of fuel cell?

you have two metals of somewhat different nobility.... i'm just sayin, maybe the effect is just plain funkadelic.

Direct methanol fuel cell - Wikipedia, the free encyclopedia

Interesting.

Still and all, the exact mechanism of action is less important to me than the fact that with pure water, I won't have to worry about this anymore.

Anybody in SW Florida want a nearly full 5 gallon jug of VP M1 Methanol?

so your saying the shit just came unglued on him and collapsed under the pressure sounds viable to me. Atleast more viable than his desintigration theory if that where the case then the part of the screen that was embebed into other matrial would still be intact imo.
And as for icant55 yes methanol and all its toxicity feels cooler on the skin and can drop relative temp faster but water has more capacity per liter to store heat energy and is therefore more effecient + stable. I dont know if you have noticed but unless methanol is stored in a seriusly air tight container it tends to evaporate rather quickly even if kept chilled. Mix it with some h20 and it stays put alot better though.

No part of the screen is embedded into any other material. The screen is a discrete component that is slipped over the end of the brass portion of the nozzle back and retained by a little ridge in the brass.

my mistake then your pic makes it look dif though.

No, the knurled section was just to give you something to hold onto while you thread that part into the base. On the new nozzle there's a hex section you use to screw the two parts together, but the old nozzle is a slightly different design which does not have the hexagonal section below the screen. The screen ended just below the raised and knurled section.

Also, the screen itself was identical to the new one- a fine square weave, not a coarse diagonal one.

Nope, the box is adjustable, you set the flow rate at which the output goes high. The box does read actual flow. There is an optional flow display as well.

So I checked my nozzle this morning. I have only run about two gallons of washer fluid (0 deg F) through this one so it does not have much use on it yet. It is a Devil's Own external mount.

http://lh5.ggpht.com/_jh2HiMm4nKQ/Sn...0/P8080139.JPG

Thymer, I think Joe's point is still valid. If the output on the Labonte box is a switched high based on a flow rate set point, then for a progressive system, where do you set it? The flow display would be useful since one would know for example that it should flow x amount at 100% DC, boost pressure aside (delta-p).

It says though that it also has a 0-5V output that is flow proportional. That would be nice to have. I sure would like to know what flow meter they have in there. I'm guessing it is a paddle wheel or a turbine wheel.

Well I suppose one could fab a comparator circuit inputting the flow meter and motor control signals. Seems a little bit overkill though. I use the flow display from the controller for a visual and the labonte as a backup. Check the injector every once in while and I think I'll be OK.

I have not see a filter come off a nozzle like that yet. I would just put it as a fluke or bad screen.

I am not a big fan of flow detection devices. IMO most all of them react slow. You would be best off using a wide band or installing an IAT "i don't believe you ahve one from the factory" Whens the last time you saw a flow sensor for the fuel system of your car to monitor when a injector goes bad?

People are always bugging me. "when you going to make a flow detection device like the others" DevilsOwn has no plans atm to do so. There is one flow sensor on the market that someone else sells i don't think i would be able to sleep at night if i sold it.

Please do share with us. Which one and why you feel this way.

Anyone else having problems viewing the pics?

nope works for me but rick said the gallery feature would bug some of the pics.

All of my pictures are hosted via external sites (imagefra.me and imageshack.us) rather than here on the forum.

Works now. The OP's pics weren't loading. Those look like a McMastr carr nozzle I got. THe screen was wrapped around, but came undone. It's been in the bag-o-fittings since.

I just took out my preturbo nozzle, which was mounted right next to the BOV recirculation point, and my nozzle was filled with gunk from the valve cover, and wasn't spraying at all. Gotta clean it up

Took mine apart after about a year & the nozzle, screen, passages & everything looked fine.

Run a polypropylene filter..