How much CO2...?

[Kizer]

I'm thinking of using a CO2 bottle, plumbed into the intake, in order to shut the engine down incase of runaway.
Any input on the minimum size of the bottle required to accomplish this goal?
Not that I would go with the minimum, of course.
Thank y'all in advance...I always get useful feedback here.

 

[rhkcommander]

I had a huge write up but my phone lost it... So i wont be redoing the math - sorry

In runaway the engine will go faster than governed speeds... I estimated a 7.3 N/A will need ~650CFM of air at 5000rpm.

So divide 650 by 60 to get CFS... Roughly 11CFS. A 5lb bottle of co2 has ~44 CF. Dont know how fast/slow it is to be released either. You dont have to match the CFS, just displace enough air to ***** the combustion.

One additional benefit some may not realize is that the co2 will be coming out cold and help cool the cylinders which should help calm a runaway too...

Alternatively you could get a co2 extinguisher, heard some success stories there and it doubles up as a fire extinguisher. Or a plank of wood over an intake pipe. Or 4th gear and brakes on a manual (for zf5 i think the 5th gear is weaker but could be wrong..)

 

[Kizer]

Thank you for the reply rhkcommander. Can you think of any harm that would come of "test firing" a few different deployment methods/volumes?

 

[rhkcommander]

None really that I could immediately think of for the truck.

Just don't do it in a garage... Too much CO2 will knock you out.

 

[Greg5OH]

i looked at those c02 eqtinguishers..my god theyre expensive!

was thinking just buy a paintball co2 canister.
ne wone is 25 bucks..
http://www.amazon.com/Empire-Paintb...1411065389&sr=8-2&keywords=paintball+co2+tank

plumb a line into the intake and aball valve valve in an easy to reach location, if oyu have a run away, dump the valve, 25oz of co2 at 1000 psi should surely kill the engine?

 

[Kizer]

Roger that...no garage tests!
Thought I'd start with a10lb tank, full open ball valve , through 3/8" hose.

 

[rhkcommander]

If you can borrow one, get a flow meter of some sort. Would give you an idea of how fast you can dump the co2. If it was fast enough the paintball canister *could* work, but I would feel more comfortable with a 5'er or more at least. Better to have too much than just barely hopefully enough.

Every pound of CO2 is 8.741 cubic feet so that paintball canister has roughly 13 cubic feet in it. My guess - two of these *could* do it with one acting as a backup just in case the first didn't kill it off - but I would feel a lot more comfortable with a 5+ tank .

Neat project, I bet you can get it working pretty easy

 

[OLDBULL8]

The only thing that would concern me is the fact you have a Banks Sidewinder turbo. If the oil return goes into the intake where the CPR was, that extra drain holes were not punched into it. The early intake gaskets will not allow the return oil to drain fast enough. The later gaskets for turbo's are made for return oil to drain out quickly.
To use a CO2 extinguisher, you would have to trigger it manually. A solenoid would not hold the pressure which exceeds around 250 PSI in the bottle, unless you have something like a nitrogen injection setup, even then the nitrogen solenoid has very small passages, which would not allow the volumn needed. Just my thoughts.

Pic is the new style. The hole at the far right is for the brass plug for any fluid to drain down the back side of the engine that might leak into the upper pan.

 

[laserjock]

Co2 is like 834 psi depending on temp as long as there is liquid in the bottle. If you want to really do it fast, turn the bottle upside down and inject the liquid. Or use a dip tube tank. Then you really get the added benefit of the cooling as the gas expands. The manifold will be really cold if you dump it in like at the intake hat. We had a thread on this a while back and one of the points that was made then was that you have to keep spraying until the engine spins down slow enough to not catch again.

 

[laserjock]

As another thought, you can get away with much smaller plumbing if you plumb the liquid instead of the gas.

 

[riotwarrior]

interesting thread....Hmmm...

 

[Kizer]

Co2 is like 834 psi depending on temp as long as there is liquid in the bottle. If you want to really do it fast, turn the bottle upside down and inject the liquid. Or use a dip tube tank. Then you really get the added benefit of the cooling as the gas expands. The manifold will be really cold if you dump it in like at the intake hat. We had a thread on this a while back and one of the points that was made then was that you have to keep spraying until the engine spins down slow enough to not catch again.

The idea that it may take upwards of 10-15 seconds of "wide open" CO2 to choke the engine down is why I think I'll start w/ a 10lb bottle.
Not a bad idea to have compressed gas onboard anyway...for a few reasons.

 

[jaluhn83]

Why do you want to use CO2? I'd say to use the good old flapper valve setup to close off the intake somewhere. I don't think the CO2 idea is going to work reliably....

Concerns I have off the top of my head:

1) You're going to be pressurizing the intake pretty good, and likely causing a pretty decent pressure shock wave when you dump the CO2. In order to displace enough oxygen to kill combustion you're going to have to basically replace all the air in the intake, which means putting a pretty big volume in fairly quickly. When you do that the air that was going in has to go somewhere.... Remember that the intake piping isn't designed for pressure so you could very well pop piping or blow the air filter up. If you dump in post turbo you'll wind up stalling the turbo which isn't good for it, and could even get reverse flow through it - not sure what exactly will happen but i can't imagine it will be good for anything. Could also pressurize the crankcase from the pressure spike in the intake (CDR closes on vacuum, nothing to prevent pressure from going into the crankcase).

2) When the CO2 expands it's cools - probably not enough to get dry ice, but certainly enough to freeze water vapor in the air. More concerning than ice would be thermal shock - all that nice hot metal is suddenly going to have subzero gas in contact with it, which could cause all kinds of fun. I'd be most concerned about the cylinder heads - thin sections with high thermal gradients and a brittle material to begin with. Could also get it cold enough to become even more brittle.

3) You'd need to displace pretty much all the air to have any real effect. The diesel will run even with a pretty low oxygen concentration, especially under low load. With a gasser you could lean it out enough to cause misfiring, but less it's much harder with the diesel since you're injecting the fuel - it'll find any oxygen in the chamber and burn eventually.You'd also need to be able to do this for long enough to get the engine completely stopped - if it's still turning over at pretty much any speed over 1-200 rpm and it starts getting air back in the cylinders (ie CO2 starts to run out) it'll take off again. Remember that everything is still hot, so it's not like starting from a standstill - as long as there's enough heat to ignite the fuel it'll run.

Can you do it? Probably eventually it'd work... but I don't think it'd be reliable or easy to get right, and to me there's a substantial risk of engine damage. Seems much simpler and easier to just cut off the airflow.

 

[79jasper]

+1 for positive air shutoff system

Sent from my SM-G900R4 using Tapatalk

 

[PwrSmoke]

It's your truck... BUT: Why is this even necessary or worth the expense and time? How many runaway IDIs have been seen? It's not like we are talking about Jimmy two-strokes here. Back in my Army Sailor days, we had runaway Jimmys on occasion (usually because the rack was set incorrectly). The air shut off device was built onto the engine or you grabbed something like a thick manual to shut off the air. The one time I saw CO2 used, it worked but wasted the engine (cracked pistons). You could build a system and make it work but IMO it would be a waste of time and money and would likely never be used because the likelihood of a runaway is in the single-digit percentages.