water injection and engine vacuum question

[lunarhighway]

i have a few questions regarding engine vacuum, as i'm trying to come up with a basic water injection system for a naturally aspired carbed engine.

i'm trying to design something similar to Ron Novaks design, but with the inlet above the butterfly valve so that there's most water coming in when it's needed.

my main question is: what is the pressure there? i assume most vacuum is created below the throttle... since i rarely use full throttle i'm affraight the pressure above the valve is not low enough to suck up enough water vapour?

on the other hand i thought perhaps it's a good idea to use a mini ram intake to pressurise the water reservoir... might help generate enough presure difference to get things going.

any tips or corrections are welcome.

[GasSavers_Erik]

Most vacuum will be below the throttle vale, this is called manifold vacuum and will be highest at idle or during deceleration.

There is also a type of vacuum on carbureted vehicles called venturi vacuum- this is the type that is slightly above the throttle plate. I noticed that your car has a carb so this will probably be the vacuum source that you want to use. Venturi vacuum is highest when the most air is flowing through the carb (heavy load on the engine) and lowest at idle. The best way to find the venturi vacuum lines is to have the car running at idle and pull the lines off the center area of the carb one at a time until you find one with very little vacuum.

Its my understanding that the water container is sealed at the top where the vacuum line comes in. The only way the water container can vent (so vacuum doesn't build up) is through the vent tube. On the end of the vent tube that is submerged in the water is a bubbler stone. The other end of the vent tube is open to the air. So outside air is rushing in to fill the vacuum through the open end of the vent tube and through the bubbler stone. The bubbling action agitates the water and throws up mist which is sucked in to the engine through the venturi vacuum line.

I've never tried the water injection, but I'm not sure that ram air could develop enough pressure through a bubbler stone to make enough mist. I suppose anything is possible with a big enough air ram, but that would create drag and lower FE

[lunarhighway]

thanks for the reply... i wasn't sure if there would be enough vacuum above the throttle, but is suppose there will be some. i think i have an idea of where to look for these vacuum lines... my airfilter is a big circular thing with a snork that bolts to the top of the carb and theres some lines running to it.

the idea behind the ram intake is that rather than have the "vent" outlet t sit in the engine bay, why not connect it so a small funnel that sits behind the grill...maybe even behind the radiator... i suppose the pressure there will still be higher than ambient... wich would mean more bubbles at high speed.

i'm now looking for a suitable container for water and a bubbler stone, or other means of bubbling ... i hope i can piece this togeter

[GasSavers_Erik]

The ram air effect might help create a few more bubbles, but most of the bubbling will likely be the result of the vacuum in the jar.

You will probably have to remove the big round air cleaner holder to get under it to the venturi vacuum lines that attach to the middle part of the carburetor. My air cleaner has a few vacuum lines running up to it, but these are manifold vacuum lines. On my car, these lines open and close a door on the snorkel that allows the carb to such in heated air from the exhaust manifold.

Of course, my car is a Honda, so it might be a very different system from yours. The best way to tell is unplug one at idle and see how much vacuum is present.

Here in the US, the bubbler stones can be bought at pet stores. For the water jar, you could probably find a hard plastic jug (stiff enough that won't crumple in when it experiences some vacuum) and drill two holes in the plastic cap, each a little bit smaller than the vacuum and vent hoses. You would have to squeeze and stuff the hoses through the smaller holes, but you might end up with an airtight seal this way.

If it was me, at first I would make the vacuum hose extra long and run it through the window and place the jar in the passenger compartment just so I could see that it was bubbling/working properly. You could then move it under the hood and use the ram air idea once you were sure it was adjusted correctly and the jar was sealing tightly.

Good luck!

[Jim Dunlop]

If you want to increase the vacuum upstream of the butterfly, create more restrictions upstream of it. My MAP shows anywhere from around 5 to 10 psia at any given time, and we know the incoming air (the end of the intake) is atmospheric pressure.

Therefore, the pressure drops across all restrictions in the flow path must add up to 5 to 10 psid. (15 - 5 = 10; 15 - 10 = 5). That is more than enough to overcome a water column only a couple feet (max) high, even if the butterfly accounts for half the pressure drop. 2.3 feet of water = 1 psi. Therefore, as long as you have 14 psia (-1 psig) upstream of the butterfly, you should be able to suck water into the intake. Place the vessel below the intake snorkel; otherwise the water will constantly drip in and cause issues.

I am assuming a setup where water leaves the bottom of the vessel as liquid directly, and not as some weird bubbly mist in the headspace. I was going to toy with this idea and buy a couple different small spray nozzles from McMaster-Carr to screw into a bulkhead which penetrates the snorkel tube.

[ZugyNA]

You are confusing a bubbler with water injection?

Air bubbled thru water picks up a slight amount of water vapor.

Pulling all water puts more in.

The advantage of using a port above the butterfly is that it won't pull water at idle, but will slightly above idle.

Since this port is so far up and your flow rate is so low...the water will likely mist out pretty well by the time it hits the chambers?

Main issue will be controlling the flow rate...I'd use MMoil in the gas as a lube when tuning this in though.

[Jim Dunlop]

Once you select the correct nozzle (and there are some tiny ones available), the flow should vary with the throttle. If it's still hard to control, you could put a needle valve in the line.

[ZugyNA]

I tried making my own using copper tubing and a tie wrap wire to form it over...using vice grips. This worked but you need a very small orifice...and a good filter to keep it from plugging up. A lawn mower gas filter works good.

I was aiming for a ~2% flow rate...where 2% = 1/50th of the gas typically used.

You are not going to see a fine mist at this flow rate?

I once hit ~ a 20% flow rate when not paying attention...ran OK, but missed under acceleration.

Good needle valves are hard to find and expensive.

[lunarhighway]

i've got a plastic nozzle from a throath medicin spray can. i forced water trough it with a plastic syringe and it made quite a mist at high pressure, but at low pressure it would just dribble so i'm not sure if it will work with vacuum.

right now i'm thinking about a two chamber system where a water reservoir is connected to a second reservoir with the nozzle, wherin the water's misted and gathered by a tube going to the carb... problem is this second chamber would eventually flood, so some sort of one directional drain valve would be needed... maybe i'm seeing things to complex here

[Jim Dunlop]

My vote is to keep it simple.

The low-pressure nozzles can be found on p. 2000 and 2001 of the online catalog at www.mcmaster.com. (I figure 5-10 psi is the pressure drop across the nozzle). Calculate the flow rate you need based on your desired Water/Fuel ratio at idle (GPH from the SG). The flow rate *should* increase as you open the throttle, maybe not in a linear fashion though.

Or if you choose to use a windshield washer fluid pump, you can choose from a wider range of nozzles because you'll be generating pressure. Try the Misting and Fogging nozzles on page 2003 of mcmaster.com.

Needle valves are $15-20 and can be found on p.415 of mcmaster.com. You will need one b/c even the tiny spray nozzles have flow rates which exceed your car's idle GPH!