Homebrew Nitrox?

[pescador775]

I did some calcs but all my assumptions are hypothetical. FWIW, I suggest that a tank which is being filled from a common compressor be cleaned at a time between 25-100 fills, eg, for purposes of PP mixing.

Assuming that one milligram oil/hydrocarbon is deposited in the tank per fill and assuming that the threshold of safety is one drop (25 mg) of oil/hydrocarbon, the limit of safety is 25 fills. Thus, clean the tank after 25 fills.

I don't know if this is supportable by any studies or experiment but anecdotal evidence suggests that it is conservative. Lots of guys have PP filled and topped off with common compressors using the PO and similar filters which some people like to hate. Mostly, these are the shamans of the diving community who incant spells while waving invoices for $1500 filter canisters.

The 25 mg oil would likely be accompanied by 20 X that much water and the presence of water which does not support combustion particularly well should extend the proposed cleaning cycle up to 100 fills or thereabouts.

Although Grade E has an allowable limit of 5mg/m3 oil I believe this is obsolete. When the moisture limit spec'ed under grade E was dropped the oil limit was not. This ignores the fact that methods currently used to reduce moisture to 24 ppm (v) also scrub oil more efficiently.

 

[cool_hardware52]

I did some calcs but all my assumptions are hypothetical. FWIW, I suggest that a tank which is being filled from a common compressor be cleaned at a time between 25-100 fills, eg, for purposes of PP mixing.

Assuming that one milligram oil/hydrocarbon is deposited in the tank per fill and assuming that the threshold of safety is one drop (25 mg) of oil/hydrocarbon, the limit of safety is 25 fills. Thus, clean the tank after 25 fills.

I don't know if this is supportable by any studies or experiment but anecdotal evidence suggests that it is conservative. Lots of guys have PP filled and topped off with common compressors using the PO and similar filters which some people like to hate. Mostly, these are the shamans of the diving community who incant spells while waving invoices for $1500 filter canisters.

The 25 mg oil would likely be accompanied by 20 X that much water and the presence of water which does not support combustion particularly well should extend the proposed cleaning cycle up to 100 fills or thereabouts.

Although Grade E has an allowable limit of 5mg/m3 oil I believe this is obsolete. When the moisture limit spec'ed under grade E was dropped the oil limit was not. This ignores the fact that methods currently used to reduce moisture to 24 ppm (v) also scrub oil more efficiently.

I agree. Most modern filtration will produce better than grade E, and many are PP blending without a hyperfilter.

OTOH, reports of wet fills are not unknown, and a "wet fill" occurs after the descant has failed and the moisture washes the hydrocarbons out of the AC.

PP blending without a hyperfilter using a compressor well maintained by the blender is one thing. If I was to rely on air tops from a compressor I didn't maintain I want a very good relationship with the fill station operator.

Tobin

 

[Cnotthoff]

The other problem you may run into with this method is that not all fill systems prevent gas in your cylinder from flowing back into their system or into other tanks being filled. If fill station operator is clued in and aware that you are doing a partial-pressure blend, your cylinder can be isolated so this doesn't happen. If not, you can end up with less oxygen than you hoped for and someone else can end up with more.

You got the last part right, for sure. Always analyze.

Good Dives,

Charlie

 

[Fl_Key_Guy]

If the local dive shop is using the same air filtration to partial pressure mix their own Nitrox that they use for regular air fills, then they are comfortable with the quality. That being said, they may or may not top off YOUR tank if they know you have YOUR pure O2 in it. Ethically, they should be made aware- maybe they fill onto pure O2 more slowly or something to keep the heat down? (More likely, a fill jockey is slamming the air into everything and the tanks are too hot to touch- a dumb move with just air) Some shops may not top off your O2 just because they would not make so much $$$ as selling you a mix.
I've been getting my O2 primed tanks topped up for quite a few years now, but the 4 shops I've used all know me well and trust my work.
Ask around to see if you can get the tanks topped off before you spend any money, then get Airspeed's Oxyhacker book for a guide if you are going to rig something up. It's cheap and easy enough- just keep it safe.

 

[Safety Stop]

Instead of calculating everytime, is there a pre-made chart that we can use???

OK, I get it, and the question (is there a question?) is fairly simple. Clean and dry the inside of the tank. Disassemble and clean the valve. Partial pressure fill the tank slowly with oxygen using oxygen safe fill whip. Starting from empty, use this formula:

(FO2 mix - FO2 air)/FN2 air X final press = psi O2

IOW, percent mix minus percent air (21%) divided by percent nitrogen(79%) multiplied by final press.

(32%-21%)/79% = 13.9, 13.9 X 3000 = 417 psi

Inject tank with 417 psi oxygen and top off with air.

To answer your question, yes, be careful, no smoking or drinking whilst doing partial pressure filling.

 

[Charlie99]

Instead of calculating everytime, is there a pre-made chart that we can use???

Here's a table that tells you how much O2 to put into an empty tank that will be filled to 3000psi to make various mixes:

EAN32 --- 417psi O2
EAN36 --- 570psi O2

What else do you need?

In practice, you would end up tweaking these numbers slightly after doing it a few times, to adjust for what the LDS actually provides as typical fill pressure, and for slight gauge inaccuracies and deviation from ideal gas.

===========================================

I'm not keen on the ethics of getting an LDS to top off a tank that they don't realized has 100% O2 in it. Oxyhacker's suggestion of using another tank of air and a transfill whip to add enough air to get the mix down below 50% O2 is a really good idea in terms of reducing the safety hazards at the LDS.

 

[Safety Stop]

I'm using a twinset of HP 130 but something I'm using single HP 100 so that's why I'm asking if is there a pre-made chart available.

Here's a table that tells you how much O2 to put into an empty tank that will be filled to 3000psi to make various mixes:

EAN32 --- 417psi O2
EAN36 --- 450psi O2

What else do you need?

In practice, you would end up tweaking these numbers slightly after doing it a few times, to adjust for what the LDS actually provides as typical fill pressure, and for slight gauge inaccuracies and deviation from ideal gas.

===========================================

I'm not keen on the ethics of getting an LDS to top off a tank that they don't realized has 100% O2 in it. Oxyhacker's suggestion of using another tank of air and a transfill whip to add enough air to get the mix down below 50% O2 is a really good idea in terms of reducing the safety hazards at the LDS.

 

[Safety Stop]

I'm problably not good in math but following this formula for 32% and 36%, I have :

(FO2 mix - FO2 air)/FN2 air X final press = psi O2

IOW, percent mix minus percent air (21%) divided by percent nitrogen(79%) multiplied by final press.

(32%-21%)/79% = 14%, 14% X 3000 = 420 psi O2

(36%-21%)/79% = 19%, 19% X 3000 = 570 psi O2

 

[Charlie99]

I'm using a twinset of HP 130 but something I'm using single HP 100 so that's why I'm asking if is there a pre-made chart available.

It's not the SIZE of the tank that counts, it's the final fill pressure. So here's the revised chart assuming that your LDS fills both the twinset HP130 and the HP100 to 3442 psi:

EAN32 ..... 479 psi of O2 to start, top off with air to 3442psi.
EAN36 ..... 653 psi "" ""

Or if you want a more general table:

EAN32 ....... start with O2 fill of 13.9 psi/100psi of final fill pressure
EAN36 ....... start with O2 fill of 19psi/100psi of final fill pressure

As always, analyze, that's the ultimate test to catch math errors.

 

[pescador775]

Safety Stop, I think you get it, but here is your table anyway:

32% mix @ 3000 psi = 417 psi O2
36% mix @ 3000 psi = 570 psi O2

32% mix @ 3500 psi = 486 psi O2
36% mix @ 3500 psi = 665 psi O2

32% mix @ 3442 psi = 478 psi O2
36% mix @ 3442 psi = 654 psi O2