My Alky System Build

I've been having so much fun designing / assembling this alky system that I thought I would share it with you guys.

I wanted to design a system that incorporated as many safety precautionary measures as possible, but wanted it to be a simple DIY type setup as well - using relays, switches, lights...basic stuff. Unfortunately, without a flow switch in the alky injection line, there's no way you can guarantee that the meth is being delivered, so I plan to do some looking around to incorporate a flow switch/sensor into the system....but for now....a low level switch will let me know if the tank is getting too low via illumination of a red LED and pulsing, silence-able buzzer (mounted next to the radio opening). Also, I currently have the wiring configured to open the SPOUT plug connection, which will pull timing on the "race w/alky" tune, until the tank is either refilled, or the system is turned off and the tune changed. I also plan on a "pump status" LED that will indicate when the pump is running, in addition to a "system armed" LED to indicate that everything is ready.

I'm just planning to run a "check valve" (actually seems to be more of a one-way pressure "switch" since it is spring-loaded), so I thought it would be good to have the ability to purge the air out of the alky line (or prime the line), since that has seemed to be a concern for some users of the snow system that was group-purchased a few years ago.

I'm using a coolant reservoir out of an early 80's silverado (about 1.25 gallons), with a level switch from a Chrysler minivan, Devilsown alky pump, tubing, fittings, Snow boost-activated controller (until I work out a MAF unit). And...relays, LEDs, switches, terminals, wiring, etc that I had laying around.

Attached are a couple of pictures of some of the progress as of late: diagram and "control board" that will be mounted in the trunk with the pump and tank. I'm going to hold off on pictures of the trunk area until it's done..since it's a mess.

Look into fuel flow meters for boats.

Measuring flow would be the best way to go, but expensive. Measuring pressure in the line going to the nozzle would be cheaper.

Is there sufficient pressure in the line going to the nozzle to trip an oil pressure sender?

Many senders trip in the 3-8 psi range. Plumbing a sender in that line would be a pretty easy solution. Now if the nozzle gets clogged, and the pressure increases, have a sender that trips if the pressure gets too high. I have seen the crude that can form in alcohol systems.

Here is a list of some of the oil pressure senders.

http://www.wellsve.com/custom_searches/ds_oil.php?showall=yes

Matt,

Don't forget to include an anti-siphon valve located close to the nozzle(s).

David

Snow already makes a flow sensor...SafeInjection. Using it to pull timing via the SPOUT, as well as pulling timing via ACT's is definitely the way to go. Relying on warning lamps alone often ends in disaster, as by the time you react, it's too late.

David Neibert said:

Matt,

Don't forget to include an anti-siphon valve located close to the nozzle(s).

David

Click to expand...

Check!

I should be able to put it within 6 in of the nozzle.

racecougar said:

Snow already makes a flow sensor...SafeInjection. Using it to pull timing via the SPOUT, as well as pulling timing via ACT's is definitely the way to go. Relying on warning lamps alone often ends in disaster, as by the time you react, it's too late.

Click to expand...

Thanks for the suggestion - I am aware of the safe injection package .... It's pretty pricey though. I think I can build something nearly as effective for much less...however finding a flow sensor / switch that will work at a good price is seeming to be difficult.

I had thought about a pressure switch instead (Charlie, you and I think alike!). However, that would not detect a clogged nozzle, as stated, unless you had an upper and lower pressure limit.

BTW, I do have provisions to add some type of line flow or pressure sensing feedback to the setup, but for now, only a low level in the tank will pull the spout (in addition to lighting an LED and turning on a pulsing buzzer). I should know if the level is low on the tank nearly immediately....which is at least a little better than just an LED. Also, right now I'm thinking of keeping the tune on alky very conservative on the AFR. I still have a knock sensor, so it will provide some protection if the alky poops out with higher ignition advance being commanded.

Thanks Chris, David, Rod, and Charlie for the great suggestions so far....

Is your knock sensor pulling timing or are you seeing all the commanded timing? I know that I got to point that it had to be turned off as other things were affecting it, picked up .4s after turning it off without any ill effects.

rzimmerl said:

Is your knock sensor pulling timing or are you seeing all the commanded timing? I know that I got to point that it had to be turned off as other things were affecting it, picked up .4s after turning it off without any ill effects.

Click to expand...

On pump gas (this summer), the tune was calling for 24 or 26 degrees by about 4500 I think...with this particular curve, timing wasn't hitting commanded in several areas consistently ( usually only by a degree or less), so I reasoned that due to warmer air (excluding ACT pulled timing) and too much timing being commanded, the knock sensor was pickIng up some activity. After adjusting commanded spark to 22 degrees max, I hit commanded throughout the pull (except when ACTS got high, then I could see the tune pull a little timing consistent with temp increase)

On race gas tune, I have commanded 28 degrees with minimal knock sensor activity.

I know it can be hard to detect by feel / ear, but I have not felt knock in my car due to any reasons - I assume because the knock sensor pulls timing before I have the chance.

Now, whether or not disabling the knock sensor would help me gain some power (because it is being a sissy), I don't know. I wouldn't be surprised, but I'm not comfortable finding out just yet. The feeling I have is that it is not picking up irrelevant noise, but true knock, since merely adjusting the overly aggressive timing (with pump gas), fixed the problem instantly. I could be wrong. Now, with the alky system, and cooler ACTs, I might be more inclined to play with disabling the knock sensor, after I have seen what results when X degrees of commanded timing w/ knock sensor enabled yields.

What did your actual timing logs look like before disabling the knock sensor, Ryan?

Want to borrow my zex 75hp kit for the shootout? Takes no time to install.......

Jacob_Royer said:

Want to borrow my zex 75hp kit for the shootout? Takes no time to install.......

Click to expand...

Tempting Jacob.....

I'm not the type to just throw something on the car though!

My 340k 4.6 is holding up to it fine! Won't take anything to setup has a built in wot switch

Got the alky system working yesterday! Had a couple of issues....first of which was solved by installing the nozzle in the proper direction! i swear i read the directions 5 times and was positive i knew which end of the nozzle went into the airstream.

Also, I kept having trouble with my blue "pump status" led burning up after the pump would kick off. I figured it must be getting a reverse voltage spike when the pump turned off or something. The other LEDs were fine though. after burning three blue LEDs (even with relatively large resistors), I decided to try an amber one instead....maybe the semiconductor material would be "tougher"...so far so good!

I did my first datalog of the system in action with a little bitty nozzle 66F ambient. One 60mph to 100mph pull...ACT maxed out at 124 without water injection. About 1 minute later, 60mph to 100mph pull.... ACT maxed out at 120F. Not a big difference, I know! This was mainly a test before really getting into it. I have a Devilsown D02 nozzle right now, but bought a D10 also, for permanent use. I might do some before/after water only testing with it too, before adding the alky....since water is cheap!!! Got some pictures of the install I will share a little later.

If you really want to keep the blue led you may try to ad a one way diode in line of the led.

r1dd1ck913 said:

If you really want to keep the blue led you may try to ad a one way diode in line of the led.

Click to expand...

I'm not exactly sure what is killing the LED's, but I suspect it is motor when turned off (either negative current or just a high voltage spike), so what I really should probably do is add a flyback diode across the motor leads. Since I have a boost contoller as well (that SHOULD already be protected), the flyback diode is probably a good idea. I would have liked to use blue, but the amber is okay. If/after I put in the flyback, maybe I'll try another blue led....I only have about 10 left though

Do any of the other alky kits out there recommend a diode across the pump? I haven't seen any mention....

So, after driving around all day yesterday with a IC pipe coupler that cracked loose, this morning I was commanding 15.1 WOT AFR! Apparently, the WOT Fuel Multiplier is more like "WOT AFR Multiplier" numbers less than one ADD fuel. Counter-intuitive, if you ask me. on top of that, I was trying to add 2% fuel, but added (in actuality, subtracted!) 20% instead....oops!

No harm done thanks to the Innovate gauge. plus it didn't't feel right....so even without, I'm sure I would've caught it.

Maybe on the way home today I can finally test the D10 nozzle!

Finally got my junk straightened out.

D10 w/ water only...nice...definitely cooler ACT's....

50/50 with denatured alcohol....AWESOME!!!!

Haven't found the limit of benefits yet....more to follow.

rzimmerl said:

Is your knock sensor pulling timing or are you seeing all the commanded timing? I know that I got to point that it had to be turned off as other things were affecting it, picked up .4s after turning it off without any ill effects.

Click to expand...

Ryan - I'm starting to think that my knock sensor is being a weenie too.

After reviewing several logs from the past several months, and even with the alky injection, perfect AFR, cool air, it still twitches sometimes at 4000, 4200, 4400...inconsistently. Only a degree or so, but still does.

Do you remember what your logs looked like before disabling the sensor?

Everything is working perfectly so far.

I decided to lengthen the low level sensor to read low with about 1/2" of fluid left, instead of half full!