Building a TRIMIX/NITROX stick

Please register or login

Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.

Benefits of registering include

  • Ability to post and comment on topics and discussions.
  • A Free photo gallery to share your dive photos with the world.
  • You can make this box go away

Joining is quick and easy. Log in or Register now!

pescador775

Contributor
Messages
2,652
Reaction score
11
I'm not a tech diver but a freediver who also has an MS in physics. I got into physics because I like to build stuff and to play with gas equations. I ended up doing electronics engineering which is a drag.

I read the Oxygen Hacker's Companion, fourth edition, and came up with several new ideas. I refer you to pages 46 and 110 in which various design ideas and techniques are offered for building and using trimix "sticks". I feel these are unnecessarily complicated and expensive so offer you a simple, foolproof device at lower cost. Only one stick need be built and only one O2 analyzer is necessary. Forget the He analyzer, the flowmeters, etc, not needed. Forget the compressibility coefficients, not needed.

Build your stick in the usual way except make it about 3 feet long and, instead of one, install two gas ports on the side near the top. Make the intake filter by drilling holes in a round PVC pipe cap of appropriate size. Insert an 1/4", slightly oversized dacron filter pad (used in air compressor filters), in the cap. Insert the filter pad so that it conforms to the round shape of the cap. The baffles are made by stacking several stainless wire pads on top of a foam filter. The foam filter is made by cutting a thick round section from some packing foam, the permeable type. These items rest against some sheet metal screws installed in the side of the PVC pipe. Install the O2 sensor as shown in the oxyhacker, in line at the bottom.

Hook the oxygen and helium lines to the fittings near the top of the pipe. Hook up the supply tanks with Harris oxygen regulators (EBay, 10 bucks) which have been modified by drilling the restricter ports with a #60 bit. Start up the compressor with filter drains open.

For purposes of discussion, assume that a 17/45/38 mix is needed. Start the helium and gradually increase the flow. Watch the oxygen analyzer while adjusting the helium. Initially, it should read about 21% or a bit less (check your calibration chart). When the O2 level drops to 11.5% lock the helium knob in place (or just keep everybody away from it). Remember, the helium fraction is 45% leaving 55% for air (or enriched air)-----thus, 0.55 X .21 X 100 = 11.5%. Turn on the oxygen and slowly increase flow until the analyzer (remember, only one is needed) reads 17%. Close the compressor drains and fill the SCUBA tank(s). Watch the O2 level and tweak as needed. When tanks are full, close the SCUBA valve, shut off oxygen, shut off helium, stop compressor, in that order. Check oxygen percent in the SCUBA. You are done.

The Harris regulators are the 50 psi type and are ideal for nitrox when modified as above. They are usually military surplus, chrome plated brass, single stage. Of course, the trimix stick can also be used for nitrox fills. I carry a shorter version on trips for nitrox fills from the back of my truck or on my boat. I like to SCUBA dive occasionally.



Pesky
 
Have you made one and tested it yet?
I dont think it will work with both gas supplys arainged this way.
 
My first test showed some drift in the oxygen level. Not bad, but enough to cause an investigation. The problem has to do with the compressor. All compressors produce blowby. Compressors like my Bauer Capitano recycle this gas to the first stage with a simple positive crankcase ventilation circuit. Helium has a high diffusion rate and this gas is shunting to the 1st stg in quantities enough to crowd out some of the O2 and to cause positive feedback to the stick. This upsets the balance in the stick since blowby reduces the compressor suction and forces O2 and He levels in the stick to rise as the output pressure increases. When the operator sees this he tends to crank down the O2 a little bit, enough to change the final percentage. My approach is not to worry about the 1-2% He variance but the O2 issue needed to be addressed. I've installed a pressure regulator on the compressor output. The regulator feeds O2 to a separate analyzer. So, for the time being I'm using two analyzers, the other on the stick. The regulator is a cheap 8 lpm medical yoke type. I removed the yoke and screwed the regulator onto a 1/4 inch pipe T. The T is screwed into the compressor final filter. It has to be done this way, avoiding long high press lines which would delay the analyzer response.

Some weeks ago, I wanted to know the equilibrium pressure when two different sized tanks are yoked together with a transfill whip. This comes in handy when filling an emergency oxygen bottle from a single tank or cascade. I looked around and could not find a formula. There is none in the oxyhacker. I checked a webplace called 'nitrox toolbox' where they gave the answer to a sample problem but did not explain how it was calculated. I developed an equation to do this and I'll provide if anyone needs it. I sent it to the 'hacker' but there was no response.
 
I guess that, by now, NAUI or somebody has realized that they can add an additional gas barb to their Nitrox stik and call it a "trimix" stick, upping the price by a $1000, of course, LOL.

The key to getting started in trimix filling is the initialization of conditions. After seeing some published advice about calculating and measuring flow rates I was on the verge of frustration. "How does one start this thing anyway?" Do you estimate the flow rates by measuring the compressor output and dividing by arithmetically derived fudge factors for each gas, and then tweak away on the knobs, chasing one number after the other until one's head spins? Well, the simple little formula in msg 1 solves this problem. That is the information that I am giving you, free. Initializing relies only on the information provided by an accurate analyzer and it requires only one simple calculation beforehand. This is true if one or two sticks are used and if He and O2 are premixed, or if O2 and air are to be premixed. Always start the He first regardless of which stick is handling that gas. Watch the O2 level drop to the predetermined level and then start adding oxygen to the mix until the level meets the requirement. Make whatever notes you deem necessary including the knob positions (gauge readings on the regulators). Take note of the He supply tank pressures before and after the fill cycle.
 
Speaking of hose barbs, I don't use them on the stiks that I've made. I take a length of straight 3/16 brake line, thread one end to 10-32 and cut off with a pipe cutter. I tap the stik with the same thread size and screw in the little steel tube. It needs one wrap of electical tape to form a good seal with the medical plastic tube (cone style female end). This makes a nice friction fit which seals but can be easily removed and replaced multiple times.
 
It has been my experience that doing continuous blending of trimix is not as straight forward as you make it out to be. The percentage of mix is changed when you add the third gas in the stick. The flow rate to the compressor does not increase to account for the increased input volume when you add the oxygen. In addition, the flow rate of a compressor will decrease as the output pressure increases. Hence you will need to make adjustments as you go along. This is irrespective of the blow back from the first stage of the compressor.

I have a “trimix” stick (see attached photo) that is much easier to put together than the hacker units and works great. What I do now is take a sample at the outlet of the compressor (just after the water separator) and use that for the gas analysis. I have found it to be much more reliable then the gas analysis prior to the compressor (on the stick as shown) and much quicker than waiting until after filtration.

omar
 
Hi Guys,
I'll do a pic soon as I can. It's not as impressive as Omar's being made like the oxyhacker instructions. I guess Omar didn't read all my stuff since I'm also doing O2 sampling at the compressor outlet, as is he. I haven't seen any diminution of compressor output that cannot be explained by blowby. Blowby can be compensated for by running the pump at full press while sampling. Electric compressors don't work like that, eg they don't mechanically slow down, although some gas engine powered compressors might slow down at max psi. In fact, because of the simple governor controls they almost always slow down. Omar, I don't understand your complaint about alteration of gas ratios which occurs when a third gas is added. Actually, that is exactly what is intended. Adding a third gas displaces some of the incoming air altering (lowering) the incoming nitrogen (and oxygen) content as per the equation that I presented. Once the helium level is locked in the correct position, adding oxygen up to the required level pretty much makes this technique foolproof. If the He and O2 are correct then there is just enough room, percentage wise, for the specified helium content. The reason this is so easy is because the (non atmospheric) oxygen and helium are being added under pressure. These added gases displace (alter) the incoming atmospheric gas (air) per the equation. The helium is not altered since it is added under pressure. This being so, and inasmuch as the final oxygen level has been determined to be correct, the nitrogen level must be correct within the limits of the variables of the compressor itself. These variables have to do with compressor speed (gas) and blowby (gas and electric). Compensation can be roughly arrived at by setting the levels of various gases while the compressor is under load.

jbisjim:
Intresting picture omar, our stick looks nothing like that. ill try to get picture of our stick too to compare.

Pesca do you have a picture of your stick? If its in the Oxygen Hacker's Companion guide I will have my copy in a few days.
 
Electric compressors don't work like that, eg they don't mechanically slow down,
I know that the electric compressors do not slow down, and I did not say that. I said the output (flow rate) of the constant speed compressor decreases as the outlet pressure increases. It is simply a reduction of volumetric efficiency with an increasing pressure ratio. To test this check the time it takes to fill a 50 cuft cylinder from 100 psig to 2200 psig versus the same volume from 1000 psig to 3000 psig. Any tank monkey can tell you that this occurs.

Omar, I don't understand your complaint about alteration of gas ratios which occurs when a third gas is added. Actually, that is exactly what is intended. Adding a third gas displaces some of the incoming air altering (lowering) the incoming nitrogen (and oxygen) content as per the equation that I presented.
AND
The helium is not altered since it is added under pressure.

Adding the third gas does indeed displace some gas but it is not selective and displaces BOTH the air and helium. Saying the helium is not altered because it is under pressure is not correct in practice. In fact that is why I have the check valve on the inlet (high) side of my blending stick. I also messed around with multiple inlets, different locations etc. This configuration has turned out to the best one I have found. And the check valve does help by keeping the blending gases in line.

Also you really need to verify the final mix contents with both types of analyzers before you say that this setup works otherwise you are just guessing. I am sure that it gets close but in my book close is not good enough when there is a readily available and easy way to do it. It has always amused me the lengths that people go to to avoid getting the proper gas analyzing tools.

Also the Harris single stage regulators do not supply a constant pressure. This is especialy true when they get down to the lower supply pressures that you can run with when continuous blending. They vary all over the place. I have a couple that I no longer use for blending because of this. Do youself a favor and get a good double stage regulator. They can be found on ebay cheap if you are patient.

omar
 
All unbalanced regs have varying pressure. However, monitoring the flow gauge is a good idea with any of them.
Yes, Omar, both incoming gases which compose the mix we call 'air' are altered as other gases are injected. However, when the Helium fraction is known through calculations and the oxygen is known by measurement that leaves 100-X for nitrogen, pretty simple.

The variation of compressor output is because of blowby. There is no mysterious factor which causes the first stage to change its input volume--unless there is resistance in the intake line, including the mixer, maybe you should check that.
 
https://www.shearwater.com/products/teric/
http://cavediveflorida.com/Rum_House.htm

Back
Top Bottom