Filling station

[sulaco]

Hey guys I'm looking to build my own filling station for nitrox, I'm planning to use a cascade system with 4 t-tanks and a whip with psi-tronix gauge, plus a Max Air compressor.

Can anyone suggest a better solution or is this fine, I'm not really bother about fill times, I just want to be able to do it myself.

Is the Max Air compress model 35 0k for filling ? or do I need the next higher model

Any help would be great appreciatied

 

[Uncle Pug]

Originally posted by sulaco
Can anyone suggest a better solution or is this fine, I'm not really bother about fill times, I just want to be able to do it myself.

Hi Sulaco,
I continuous blend nitrox (& trimix) as does Omar who will no doubt weight in here momentarily.

Rather than have a cascade of O2 cylinders I only have to have one on hand and I can run it down to 50psi while blending.

I have three 3500 psi K cylinders that I use for a storage bank of EAN32. When we come back from a dive with say 1600 psi left we can just top it off from the bank without having to start the compressor.

Another advantage of continuous blending is that tanks don't need to be as scrupulously clean as you are not introducing straight O2 into them.

 

[roakey]

Unless you have one heck of a lot of lot of cylinders to fill at one time, I'd only bother to get two T bottles of O2. That's what I have and I'm able to draw the low bottle down to about 50 psi, so I'm not wasting much of anything.

Even the shops around here usually run two bottles max, some even one.

Roak

Ps. If you don't already own the Oxygen Hacker's Companion, do not pass go, do not collect $200 and order it: http://www.airspeedpress.com/

 

[omar]

Partial pressure blending is a simple method of obtaining nitrox. However, the risk of oxygen coming into contact with hydrocarbons and contaminants found in oil based compressors makes it potentially hazardous. PP blending with oil lubricated compressors requires connecting the compressor into a proper filtration setup, such as Lawrence Factor’s Hyper Pure filtration system that removes excess hydrocarbons that might cause combustion when exposed to pure oxygen. It is also strongly recommended that cylinders receiving oxygen be cleaned for oxygen service.

The National Oceanic and Atmospheric Administration (NOAA) Nitrox Continuous Mixer/Blender system was pioneered by J. Morgan Wells, Director of NOAA’s Diving Program. Continuous blending combines oxygen and air into a blending chamber and then directly into the compressor. The nitrox mixes can be delivered right to scuba tank or into a storage bank. NOAA recommends that you use an oiless compressor like Pug’s RIX SA-6. Also, there are systems such as the Llewoc SIS Nitrox Stik, developed by Ross Cowell, that preblend the nitrox and run it through an oil lubricated compressor with appropriate filtration. The only real difference with this setup and the NOAA setup is the oil lubricated compressor. If you migrate to a continuous blending setup for personal use you can easily make do with one bottle of oxygen but it would be wise to get a couple of storage cylinders.

I use an oil lubricated compressor with the Lawrence Factor Hyperpure filter in addition to a typical filtration tower and continuous blend Trimix and Eanx. I also have 4 high pressure (4500 psi) storage cylinders so I only have to start up the compressor on a monthly basis (usually). And then I usually take a full day to mix up a bunch of gas.

I have been in the process of making improvements to my system the last couple of weeks and was just writing it up to send off to folks that I know who are continuously blending Tmx. In short, if you go to continuous blending I would recommend that you get a flow meter to control the output of the gas from the supply bottles in addition to the regulator. In the past I have been controlling the gas output of the supply tanks by adjusting the pressure of the regulators. Some times this is a hit or miss proposition. To make a long story short, adding the flow meter to the output worked like a charm and I had enough control of the flow where I could fine tune the mix with very little wasted effort or time.

In addition, after screwing around with a scuba valve, and then a needle valve to regulate flow on my fill whip I have added a blending panel regulator. While this is a fairly expensive approach the added control and safety definitely out weigh the cost and in my mind should be a necessary component of a fill setup.

omar

 

[Uncle Pug]

Originally posted by omar
In short, if you go to continuous blending I would recommend that you get a flow meter to control the output of the gas from the supply bottles in addition to the regulator. omar

Omar I have been using Dwyer's VFB 4" flow gauges on the O2 and He regulators but I think I want to add one to the Argon regulator as well.... only problem is finding one in the right range... 0~10 CFM...

Your thoughts?

 

[Rick Murchison]

Omar,
Are you mixing your EANx first and adding the He downstream with two Oxygen sensors to indirectly monitor the He content? (It sounds like you have a dream setup... and you've got my little pea brain designing away!)
Thanks,
Rick

 

[omar]

Pug,

I am using the RMB-53-BV now, 10-100 SCFH (~5-50 L/min) for both. These are 8.5" with a 5" scale. I started with the 20-200 SCFH but the flow rates when I was using both were in the 30 to 75 range. I sized them for individual output, not taking into account that they both would be going at the same time. All of these flow meters are rated for air which is close for oxygen and is about half of what the helium going through it is. So these ended up being the right size for my 5 cfm unit.

I haven't played around with a flow meter for the argon because I use the vaccum gauge to set it. I would put a flow meter (ball in a tube) from one of my spare oxygen regs on it, or one of the 20-200 meters to see where it falls out first before I buy anything else. I would guess that this would probably do it for my setup because I have a fair amount of intake tube that acts like a reservior for the input. (Also you rarely get the rated output (volume) of your compressor when it is at the higher end of its pressure output). I know mine drops to 2.5 cfm when I am up around 4K.

omar

 

[omar]

Rick,

I blend He and O2 at the same time. I run it through a blending chamber built by Arnold Henry of Masterline Inc in Kent, WA. He built these for EANx but I added a "T" to the input line to add both gases at the same time. I send it to the compressor, through the filters and to a manifold where I can send it to a fill whip (scuba tanks) or the storage bank. I have a medical oxygen reg picked up from e-bay connected to the manifold also. I am using this type of reg because it has a nice variable area flow meter (ball in a tube) for the flow out. I run the output of the reg/flowmeter to a helium analyzer and a oxygen analyzer. The O2 sensor is fitted into the gas cap for the helium sensor so there is just one line.

After running a bunch of gas through this set up and logging the results I can fire up the system, turn on the gas to predetermined levels and get within 0.1-0.5% right away. I then tweak it to get what I want.

I have also done a very similar type of setup with a nitrox stik which can be seen at:
http://www.divemagic.com/html/trimx__blending.html

If I was going to do another one. I would get a couple of 1" diameter, 12" long PVC inline static mixers for $75 each at cole-parmer. I would add the gases at the mid point of the first mixer. I would stack/connect the mixers with a coupling that had a disk with holes in it (similar to a shower drain) in between them. And then run it to the compressor. The rest of the setup would be the same.

If you do not want to get a helium analyzer you could do this in the manner you suggested. Add the O2 to the first mixer and run it by the first O2 meter. Than add the helium in the second mixer and run it by the second O2 meter. You would need to have a chart to aid in a quick check of the mix percentages. Or determine up front what your target is. Trying to do it while you are mixing is not a good idea because it is so easy to get distracted.

omar

 

[Uncle Pug]

Originally posted by omar

I haven't played around with a flow meter for the argon because I use the vaccum gauge to set it. I would put a flow meter (ball in a tube) from one of my spare oxygen regs on it, or one of the 20-200 meters to see where it falls out first before I buy anything else.

I took your advice and installed a vacuum/pressure gauge on the intake but I haven't used it since to pump Ar....

I just thought that perhaps having a flow gauge would help me get it set. I did install a ball valve in the intake stream right after the filter so I can shut off the air intake....

Your point about the compressors slowing down cfm as the pressure rises is a good one.... do you notice the change in your vacuum gauge and have to adjust the Ar feed?

I'm not really sure what is happening with the RIX but with the same O2 flow setting I start out at 31.5% and end up around 32.5% when I get to 3500 psi so I don't know how much the cfm is changing. Unlike you.... I am definitely not a math wizard ;-)

 

[omar]

Pug,

Try running just air and monitor the O2 while you run it up to 3500 psi. When I did this I found that the O2 content increased about 1%. I have found that I need to run my compressor for about 15 minutes prior to doing any mixing. I set my oxygen analyzer to this air and then start mixing.

Another little trick that I do is put a large plastic bag over the top of the intake of the blender. I noticed that I was getting a little blow by and pulsing at the air intake when the compressor gets up to higher P. I morphed this idea from the oxy hacker book where it suggested to use a bag around the intake for a blending chamber. (I do NOT recommend doing this). The bag is open on the bottom and acts like an umbrella and captures the blow by. I have found that I use less helium and the mix remains more stable while blending. Which suggests that I am getting blow by all the time. This may or may not work with your blender I don't know how Scott put it together.

omar