Continuous Trimixing Question

[gxsr_sarge]

You may have noticed in another recent thread that I was somewhat succesful in building an Oxyhacker nitrox stick. Well now, of course, I need to push it and CB trimix. The issue is that I only have ONE Analox O2/He analyzer so I'd like to set up the trimix blending as follows:

- I don't plan on using a "premixer" to mix the O2 and He before it gets into the stick (as shown in the book when one DOES NOT have a He sensor). I plan on connecting a "Y" on the one input hose barb so that I can flow O2 and He into the "Y" on the top of the stick at the same time. That way, I don't have to drill a second barb into the pipe and cover it when I just want to blend O2. Both supply tanks (O2 and He) will have an regulator w/ an .032" orifice to get decent back pressure of about 30-40 PSI.
- I'll have my "El Cheapo" O2 sensor on the upstream side (on the sticks output) to test for O2%.
- I plan to connect the Analox on a downstream regulator and check the mix post compression as it's going into the bank. I plan to CB 21/35.

So the bottom line is that I'll have upstream and downstream O2 sensors but only one downstream He sensor.

Does anyone see a problem with this? What are your trial & error experiences in CB-ing Trimix.

Thanks!

 

[Scuba Duck]

Hi Sarge,

You may want a larger hole (orifice) for the helium side. Helium flow rates are typically larger than oxygen. For a 50% mix with a 5cfm compressor, you will need 2.5 cfm of helium, more than twice the oxygen flowrate you would need. (for 40% Nitrox you only need 1.25 cfm Oxygen) So your orifice will need to be bigger to accomodate this. Based on your desired helium flowrate, use the choked flow equation to determine the correct orifice size. (or use a needle valve)

http://en.wikipedia.org/wiki/Choked_flow

Note the ratio of specific heats (k) for Helium is 1.66.

I took the liberty of doing the math, you'll need a .065 inch diameter hole to get 2.5 cfm of Helium at 40 psi regulator pressure.

http://en.wikipedia.org/wiki/Orifice_plate

You can use the oxygen analyzer to roughly determine the helium flowrate. For ease of math, lets say your desired mix contains 50% HE. If you start mixing and you only add the HE, you will drive down the oxygen analyzer reading. If you initially calibrated the O2 analyzer to 21%, when it reads 10.5%, you have a 50% HE mix going into your Stik. Now add oxygen to raise the O2 to the desired level. Adding oxygen doesn't change the helium flowrate, but adding helium does change the amount of air that goes into the stik and subsequently the oxygen content. So you have to adjust the helium first. Once you start fiddling with the oxygen, you don't have a good way to measure the helium content pre-compression. It might also help to have a flowmeter on your helium side.

In your above example of 21/35, you would initially drive the oxygen analyzer down to 13.6% to establish your 35% HE flowrate, and then add the oxygen to get you back up to 21%.

You will be able to QA your mix on the ouput side with your trimix analyzer and make slight adjustments to get your mix right.

 

[rjack321]

The lag between the gas going in and the analyzer will be rather large. It will be hard to change your flow rates appropriately as the compressor output pressure changes. I would hook up a CO sensor to your downstream bleed off instead.

Then get yourself a small 2/L mini sampling pump like from these folks for <$50
Hargraves - Miniature diaphragm pumps and compressors for air and gas OEM applications.

Put in a barb just downstream of your stick and bleed off some gas pre-compression to send to your He analyzer. Use your existing on stick O2 sensor for that fraction.

 

[gxsr_sarge]

Derek, thank you very much for "opening my eyes" regarding trimixing. I would have just cranked up the O2 and then the HE and would have driven myself crazy trying to figure out why it wasn't working.

Also, thanks for doing that math! That orifice size is about the size of a 1/16" drill!

Hope you are well out in paradise!

 

[gxsr_sarge]

The lag between the gas going in and the analyzer will be rather large. It will be hard to change your flow rates appropriately as the compressor output pressure changes. I would hook up a CO sensor to your downstream bleed off instead.

Then get yourself a small 2/L mini sampling pump like from these folks for <$50
Hargraves - Miniature diaphragm pumps and compressors for air and gas OEM applications.

Put in a barb just downstream of your stick and bleed off some gas pre-compression to send to your He analyzer. Use your existing on stick O2 sensor for that fraction.

rjack321, thanks as well but I'm not sure I understand what you are suggesting.

-"I would hook up a CO sensor to your downstream bleed off instead"- I presume you meant an O2 sensor? Where would the downstream bleed off be? Post compression?

Are you suggesting that I put that pump (great idea by the way) at the end of the stick (pre-compression) in order to be able to connect a hose to my Analox and switch back/forth between pre and post compression using the same He analyzer.

- "Use your existing on stick O2 sensor for that fraction" - Again, sorry I'm a little thick-headed. Are you referring to Derek's fraction?

 

[Scuba Duck]

I think rjack321 is suggesting that you put together something like this...

Helium Station

 

[gxsr_sarge]

Thanks again Derek.

I don't see how that's different from what I was suggesting except that the He analyzer will be downstream (post compression) mostly to take advantage of the pressure for the Analox AND to get a more realistic mix% of what's going into the banks. Otherwise if the Analox is upstream as in your diagram, I'd have to use the little pump to get air to the sensors on my analox. That pump, though, is looking like a great idea so that I can swap the sensor tube back and forth.

I'll give it a shot over the weekend with my suggested set up (and your math!) and report back.

 

[Cave Bum]

One thing we found with continous blending of trimix was that the helium was 'leaking' out the top of the stick so for mixing the 10/50 in the shop I worked in we mounted the stick upside down so the lighter then air helium would flow up into the stick.

 

[Scuba Duck]

Sarge,

You would be able to swap the Analox Trimix Analyzer back and forth between the goes-in and goes-out side of your compressor. Its much easier to adjust the input side (oxygen & helium) with the analyzer right after the stik. There is no time lag associated with the compressor and filters, if you try to adjust the mix only using an analyzer at the output side, you will be constantly over-correcting. But you are right, the analyzer at the output side is more accurate and more representative of what is going into your tank. The analyzer you have requires pressure to drive flow through it, so you need some method to force gas at a controlled rate through the analyzer. Easy to do at the output side of your compressor.

Since you are now into continuous gas blending, I would encourage you to modify your fill panel with an additional Tee, isolation valve, and 1/4" NPT to Yoke (or DIN) adapter so that you can mount an old SCUBA first stage, and use that BC flow adapter to provide the controlled 1-3 lpm flowrate to your analyzer. The Tee should be upstream of your banks/fill whips so that you sample the gas before it goes anywhere else. That way you can continuously monitor what is comming out of your compressor and going into your tanks/banks. As you are in South Florida, the humidity there will contribute to a noticable difference between what goes into, and what comes out of your air compressor. Here's a couple pictures of a recent install on the Big Island...

http://www.rubberduckiedesigns.com/images/BID%20Short%20Hose%20Reg.JPG

http://www.rubberduckiedesigns.com/images/BID Analyzer at Fill Station Backlite2.JPG

 

[rjack321]

-"I would hook up a CO sensor to your downstream bleed off instead"- I presume you meant an O2 sensor? Where would the downstream bleed off be? Post compression?

No I meant carbon monoxide the worst possible contaminant you can get in your gas. Use the 2-3L/min bleed off downstream of the compressor to measure that, not the helium percentage which will be close enough if you are measuring it pre-compression.

Trying to measure He post-compression is like chasing GPS coordinates around, there's always a lag and you will continually overcorrect one way and then the other. The reality is that you don't need to measure it after the compressor anyway. Measure at the inlet and then the final mix in your tanks. A post compression bleed-off is better used to measure CO.

Are you suggesting that I put that pump (great idea by the way) at the end of the stick (pre-compression) in order to be able to connect a hose to my Analox and switch back/forth between pre and post compression using the same He analyzer.

You could switch it, although if you connect the little pump near the O2 sensor on the bottom of your stick what is going past that point is the same as you are pumping into your tanks. The differences in compressibility at most add up to 1-2% on the He fraction and don't affect the O2 fraction at all.

- "Use your existing on stick O2 sensor for that fraction" - Again, sorry I'm a little thick-headed. Are you referring to Derek's fraction?

You have an O2 sensor on the bottom of your stick right? Measure the O2 percentage there, not what gets pumped off to the He analyzer, nor any of this post-compression biz.