blending with a stick?

[omar]

It is not a set of directional jets..... "static in-line mixer"

You can see the detail on the US patent office site.

Yes, it is expensive. Probably twice as much as it needs to be. But it has not had any real competition to drive the price down.

omar

 

[Genesis]

does it really do any better of a job, at the outlet, than any of the other means of producing a good "mix" of the gas in the stick?

And can that be quantified, with hard evidence?

If not then it falls into the "gee, that's nice" category - at least for me.

A precision orifice produces a LOT of turbulence in its immediate vicinity. Injected into an area of lower pressure (venturi?) and then having the gas given a good "spin" (perhaps a helix path through the stick down to the bottom, then back up and down the center bore to the exit?) I bet you'd get results indistinguishable from any patented gobbligook when it comes time to take a look at this from analyzing the output for its homogeneous quality (or lack thereof.)

 

[omar]

The static in-line mixer and the nitrox stik use a helix insert in the tube with multiple elements that reverse the direction of flow in addition to providing the "many turns".

Much simpler than running the gas up and down with injection nozzles to provide the turbulance.

I use mixing tubes with clear PVC, there are 12 elements over 24 inches. The mixing can be observed by running an input of smoke into the unit and just watching.

As I said it is a proven technology in industrial applications. The food industry uses them extensively to blend juice concentrate and water; color additives, blending syrup with carbonated water (pop) etc.

By contrast, just using baffles for mixing if not sized right and with too few of them the mixing will not be as complete.

omar

 

[Uncle Pug]

I was going to make a blender using PVC pipe and a machinist friend said, "Let me make you one out of stainless steel."

I said, "OK."

So he did and the gas (O2, Helium) is injected into a venturi and there are two delrin baffles in the main bore... with random holes drilled in them. So... I have two baffles over 6" and I have no way of knowing just how thoroughly mixed the gas is by the time it travels the 36" to the first stage of the compressor... but I suspect it is thoroughly mixed by the time it gets to the second stage.... but then I just suspect this to be so... no proof.

Now the post compression analysis seems to be very stable... and I am taking that to indicated that things must be getting mixed up fairly well upstream.

What do you think Omar... should I rework the blender to include more baffles/holes at angles/reverse flows ~ or maybe even a static inline mixer???

 

[Stone]

If it becomes apparent that I did not correctly baffle my Oxyhacker stik, I will probably spend the extra bucks to incorporate one of the in-line static mixers shown on this Cole-Parmer catelog page

 

[omar]

well UP

I have always wondered if SK's blender had enough baffles in it for complete mixing. In your case it doesn't matter. This is because of the RIX. For oil lubed compressors I would rather err on the side of caution. An unmixed stream of oxygen could have nasty consequences.

If you look at:
http://www.usplastic.com/catalog/pr...k+Valves&product_id=Polypropylene+Check+Valve

A nice little check valve to go between the filter and air intake will stop the blow back that we talked about some time ago.

omar

 

[ericfine50]

Omar,

Do you think SK's dosen't have enough baffels? Silly question, does the length of the "stick" have anything to do with it?

Eric

 

[FredT]

Absent that lots of turns and flow dividers work. The twisted static mixers are good products where space is critical, and a 24 turn mixer is about a good as it gets for liquids. OTOH space arond a compressor is rarely critical.

For gasses eddies and flow division is a good thing.

A VERY simple mixer can be made from 4" PVC pipe and ping pong balls. 3 or 4 feet of packed pipe will do wonders for mixing gases. Another option is to pack the pipe with loose pack cooling tower packing (VERY high surface to volume ratio with minimum flow restriction). I think the last time I priced a ball mixer for a 20CFM compressor my cost came out under $150, including the 16' of pipe scrapped cause I was just making one.

http://www.usplastic.com/catalog/pr...ory_name=&product_id=Plastic+Floating+Spheres

The best bet is to get balls that are about 1/3 to 1/4 the inside diameter of the pipe, and another "filler" set at 1/4 the diameter of the first ball. Final screen mesh should not be an even fraction of either size ball.

FT

 

[spaz]

Thanks for all the great responses.

 

[oxyhacker]

If you really want to get gas to mix, the trick is to fold the path back and forth over itself, and make it turn at sharp angles, for maximum turbulance and intermixing. This is what a static tube mixer does - each set of vanes is angled opposite from the last one, so the flow constantly is reversing against itself, and this is what the Oxy Hacker mixer does, much more crudely (and cheaply), with the little shelf under the holes. BTW, for those who don't like the baffles system in our mixer (or the material, or the color, or the......) we give info in the book on static mixers too, the builder is free to use whatever method he/she choose.

The problem (if you can call it a problem) with mixers using only a couple widely spaced baffles with holes drilled in them is that if you flow smoke through them, you'll often see (though this depends a lot on the speed) several discrete streams of smoke zigzagging their way from one baffle to the next but never intermixing, looking almost like someone threaded several pieces of rope though the plates.

However our feeling is it doesn't take a whole lot to mix O2 to an acceptable degree so I wouldn't worry about it if the downstream analysis is matching the upstream fairly closely.

We tested ours by making a sensor holder that could be rotated around the tube, and moved in and out, to sample at different spots in the flow path, so see if it made a difference in the reading. It didn't. Not a perfect method, since the sensor acts like an additional baffle, and doubtlessly further homogenizes the mix, but good enough for our purposes.

BTW, even with a perfect mixer and two perfect analzyers, the before and after/upstream and downstream analyzers won't necessarily agree - the moisture removed in the compression process can alter the reading by as much as a percent or two!

Uncle Pug once bubbled...
So he did and the gas (O2, Helium) is injected into a venturi and there are two delrin baffles in the main bore... with random holes drilled in them.