Peter Haack made these nice DIY venturis by heating plexi and PVC-tubing:
http://www.haack.se/venturi.html

This one is very stylish =)


I'm anxious to see how they will perform..

Should the CO2 be introduced at the throat, just before the throat, in the straight section before the throat, or does it matter? For that matter, I still don't see the theory behind why a venturi results in microbubbles of CO2. But, this is certainly a simple, elegant way to make a venturi, and a venturi that should not result in much of a head loss.

Here's how a venturi works:

Venturi effect - Wikipedia, the free encyclopedia

Regards,
Tom Barr

A proper venturi is not as simple as a tapered pipe. There is a disproportionate difference in the angle of the taper IE: 90* input taper to 47* output to create a pressure differential to fascilitate the vacuum . This is merely an example as the taper is dependent on volume, and pressure.

Like wise there will also be a difference in the gas intake for Low pressure systems (Powerheads). IE: a quarter inch air intake might benefit from a .036" orifice on the leading edge of the air intake to compensate for lower static head pressure.

I'll provide a more specific example as time permits as I'm presently experimenting between proto-types designed specifically for use with powerheads, and Co2 induction. HTH. Prof M

For the record: The Mazzei Injectors are definitely worth the price, but are designed for high pressure sytems. Given their efficiency they still work splendidly for powerheads. Just not as well. A simple modification to the gas intake throat makes quite a difference, Though the machining is somewhat tedious.

The smaller the sizing of the orifice entering the pressure differential will help.

This is due to volume of air vs the volume of CO2 we are adding.
For our applicastions, we are not adding air and lots of it, just a small amount of CO2.

We need a smaller pipe and higher pressure than most venturis. you can modify the orifice effectively using those rigid 3/16" OD tubing they sell at the LFS's.

Using heat/lighter etc, you can taper the tip so that only a tiny opening is present.

This will release smaller bubbles.

The other idea is to place a mister in the venturi lead in and have a loop that feeds water through the venturi section.

Not sure how to describe it yet.
Better make a drawing and post that.

Regards,
Tom Barr

Yes the induction of Co2 as opposed to air is an entirely different matter as well. 1/4 " line is fine all the way up to the last orifice into the venturi. It allows you to offset the .036 opening to the leading edge. This increases the velocity of the gas flow when using power heads at a 150 to 180 gph flow rate. A very small augmentation to the gas flow produces a much finer mist, and allows you to maintain what precious little flow you have ( The flow rate is perfect for the reactor, but has a low static pressure as it relates to venturi performance).

I use this intake in a recombinant fashion to purge the reactor. Your original design was more than adequate for most systems. The short comings were always in the powerheads themselves. A longer reactor tube can handle the increased flow rate, but the escaping mist is actually a perk in my book. Now if you modify an internal duct within the reactor tube (Mixing Condenser) you can compress the stream for a more efficient exchange of gas, but it's really a mute point as the reactor is quite efficient already. Not many people push more than 180 bpm.

Venturi injection for Co2 is not really tricky, but it's considerably different than o2 or o3, and freshwater doesn't enjoy the same luxury of density as saltwater. Prof M

Your method of tapering the 3/16" pipette is much simpler I just have to figure out how to orient it within the venturi orifice ? A proud edge to the orifice would increase the vacumm energy !

From the Wikipedia reference: "A venturi can also be used to mix a fluid with air. If a pump forces the fluid through a tube connected to a system consisting of a venturi to increase the water speed (the diameter decreases), a short piece of tube with a small hole in it, and last a venturi that decreases speed (so the pipe gets wider again), air will be sucked in through the small hole because of changes in pressure. At the end of the system, a mixture of fluid and air will appear."
Or, to paraphrase: A venturi works to mix fluid with air by working to mix fluid with air. That isn't an answer. My question is why does a venturi produce microbubbles of CO2 which is introduced where in the venturi? I know very well how a venturi is designed and how it works as a venturi, having designed several many years ago to measure flow rates of both air and water, but I don' t know why it works to make microbubbles of CO2, and not knowing why it works, means I have no idea how to make it work better or best or well enough, etc. In principle, the lower pressure at the venturi throat should result in big bubbles at that point, which become smaller bubbles when the pressure goes back up in the downstream part of the venturi. But, that wouldn't by itself produce micro bubbles. In fact if you introduced milimeter size bubbles in the approach pipiing to the venturi, the bubble size in the outlet tube from the venturi should be almost exactly the same, because the static pressure will be almost the same.

If you are using a venturi to measure a flow rate, and callibration is very difficult, then and only then is it critical to optimize the inlet and outlet forms for the venturi. That lets you use theory to determine the flow rate versus the pressure change from inlet to throat of the venturi. But, for any other use, a simple smooth transitiion from the inlet diameter to the throat diameter is just fine, and the exit transition from the throat to the outlet diameter only needs to be gradual enough to avoid flow separation, which would result in a big pressure drop across the whole venturi. So, for our purposes - to produce CO2 mist - all we need is a simple venturi with a smooth transition at both ends to and from the throat to avoid pressure losses.

But, how that produces CO2 mist remains a mystery to me.

The Transition of the tapers are in fact critical to the performance. (Compressing, Drafting, Swirling and Re-expanding ) Either will provide a venturi Affect, but one will provide better (Sustained) performance. Due to the difference in the density of Co2 a smaller gas orifice is desirable, and this smaller mist is swirled into the impeller where it in turn cavitates to corrupt even smaller bubbles.

I'd be happy to provide a working model, and you can study it, and reduce your impression to an intellectual prose that suits your own taste. Never the less it does work rather impressively. You designed flow models. I designed injectors. Did you want to calculate the flow or inject a gas ?

Here's a link to a splendid design.

Mazzei Injector Corp. - How a Mazzei Injector Works

The only augmentation necessary for our purposes is to decrease and offset the orifice to adapt it to Co2. Basically machining or fabicating a drop in port to increase the intake velocity of heavy gas.

Venturi injection is mainly attractive for maintenance purposes, increased flow, and faster saturation.

Mi Dos Centavos, Prof M

As the gas is sucked into the vacuum of pressure differential, it is torn into smaller gas "apherons", their size are determined namely by the volume of gas being fed into the venturi through a small orifice and then the flow rips it further and prevents coalescing.

Think about tube of toothpaste and you cutting off pieces of the paste at a fast or slow rate.

The faster the rate it comes out, the larger the pieces.
The faster you cut, the smaller the pieces.

The slower the rate the gas comes out, the better the chopping the venturi effect will have.

Unlike the disc and other means to diffuse gas, this has positive pressure.
Addign gas to the suction side of a pump will also chop the larger bubbles veryb effectively, see the needle wheel skimmer from RedSea, they sell the pump and impeller as well, 15$ for the impeller, 70$ or something to powerhead.impeller together.

I'd like to see if I cannot use the impeller in other brands of power head.
That would chop the gas up well, then, recirculate it through a venturi.

An in line design can be done effectively.
I think I already have a veryu effective sump and in line Reactor design, but I'd like to focus on a venturi design that does not use a reactor tube at all.

Regards,
Tom Barr