Are there formulas for Blending Nitrox and how to achieve a certain percentages?

[Kevrumbo]

Well stated, nice jab. I'll entertain your question. But first, lose the "us". Nobody cares but you.

Mathematical modeling is an interesting subject to me. I had the good fortune to be educated by a math teacher (in my far distant past) that truly understood math. He passed on to me the fundamental idea of what math really is and how to use it. That remains with me today.

If I don't model compressibility my real-time simulation gives me the wrong answer and that gives me an incorrect understanding of how to blend accurately. So then why bother? Just pump the tanks, analyze, and adjust. Diving is my recreation and I enjoy the math in it too. Probably the best answer to be had to your challenge.
It's called "humor", dude.
Yep, but you get into problems with helium. I'm planning on going there shortly. Now there is the REAL "geezer gas".

- D

Provided your tanks aren't rated to higher pressures than some 230bar/3300psi, that is.

On this side of the pond, we've got 300bar/4350psi rated tanks, and compressibility/non-ideal behavior quickly becomes an issue when you pass 230bar/3300psi pressure. A 10Lx300bar tank holds about 90% the amount of gas compared to a 15Lx200bar tank, even though the nominal capacity is the same. That can give some... interesting results if you PP blend and don't analyze afterwards.



--
Sent from my Android phone
Typos are a feature, not a bug

Some more specific examples please: So how will compressibility affect a particular mixture's accuracy like Trimix 10/70 in a twinset, compared to just Nitrox32 in a recreational single cylinder?

What are these problems & "interesting" results that matter to you? Is it economics (i.e. wasted gas), or more importantly --The MOD/END of your Trimix blend would turn out grossly & dangerously incorrect? Please elaborate. . .

 

[Storker]

Some more specific examples please: So how will compressibility affect a particular mixture's accuracy like Trimix 10/70 in a twinset, compared to just Nitrox32 in a recreational single cylinder?

I have no idea, since I haven't read up on helium. I don't do He. If I were planning to get trimix certified, I'd check the non-ideal behavior of He, but as it is I don't need to.

What are these problems & "interesting" results that matter to you? Is it economics (i.e. wasted gas), or more importantly --The MOD/END of your Trimix blend would turn out grossly & dangerously incorrect?

Take a guess.




It's the MOD, of course. Don't be silly. If I were concerned about the cost of wasting a few liters of oxygen, I couldn't afford to dive.



--
Sent from my Android phone
Typos are a feature, not a bug

 

[lowviz]

... So how will compressibility affect a particular mixture's accuracy like Trimix 10/70 in a twinset, compared to just Nitrox32 in a recreational single cylinder? ...

Will you knock it off?!!! What are you fishing for, the Buckingham Pi theorem in the restriction...

 

[Kevrumbo]

I have no idea, since I haven't read up on helium. I don't do He. If I were planning to get trimix certified, I'd check the non-ideal behavior of He, but as it is I don't need to.


Take a guess.




It's the MOD, of course. Don't be silly. If I were concerned about the cost of wasting a few liters of oxygen, I couldn't afford to dive.



--
Sent from my Android phone
Typos are a feature, not a bug

I've never had issues with grossly out of tolerance MOD/END because of compressibility concerns.

So what specific gross example did you (or you Lowviz) experience in a mistaken Trimix blend because of compressibility? Elaborate.

 

[runsongas]

I've never had issues with grossly out of tolerance MOD/END because of compressibility concerns.

So what specific gross example did you (or you Lowviz) experience in a mistaken Trimix blend because of compressibility? Elaborate.

it won't be grossly out of tolerance because the difference in compressibility is not very large between oxygen and helium in the grand scheme of things. But if you want an exact mix within +/- 0.5 percent rather than +/- 3 percent, then you have to consider compressibility.

BAUE High Pressure Gas Mixing

there is a spreadsheet at the bottom of the page that you can plug and chug with

 

[lowviz]

... So what specific gross example did you (or you Lowviz) experience in a mistaken Trimix blend because of compressibility? Elaborate.

You understand the concept of error bounds and cumulative errors, I would presume.

When did I say anything about a gross error. That is a subjective statement, anyway.

When modeling, one strives to get the very best model possible. If done properly, the model becomes accurately predictive. If I'm going to fudge it, why model in the first place? Just pump and adjust. Nothing wrong with that either if you just want to dive.

Open your mind a bit and allow for the idea that some people just like to figure things out for the sake of understanding the forces at play. Even common stuff provides surprises when accurately modeled: Discover Brookhaven: Scientists Create and Manipulate Nanoscale "Water Wires"

 

[SubAquatix]

Dear Lord!!

"Welcome to ScubaBoard, where the simplest questions become obfuscated by the id."

Might have to become my new tagline!

 

[lowviz]

Re: "Are there formulas for Blending Nitrox and how to achieve a certain percentages?"

Ans: Yes.

 

[waterpirate]

I am an admitted math idiot, however I was curious about blending and taught myself to do it sans the graduate level formulas. This curiosity led to the purchase of my own compressor, not for the faint of wallet. It is not complicated, you prolly will not die if you are a sane adult with some modicum of sense. I am now blending my own trimix, and never been happier!
YMMV
Eric

 

[Storker]

what specific gross example did you (or you Lowviz) experience in a mistaken Trimix blend because of compressibility? Elaborate.

Kev, do you mind actually reading posts before replying to them? I'll repeat, in very simple words:

I. Don't. Do. Trimix.

Do you get it now? I can't think of any way of saying it more clearly.



---------- Post added February 15th, 2015 at 01:48 PM ----------

Just for the sheer geekness of it, I made a spreadsheet for nitrox PP blending calculations. It has one column for ideal gas behavior, and one column using the compressibility factors given here.

Although you don't get gross errors by assuming ideal gas behavior, there is a difference if you go from 200/232bar (3000/3350psi) rated tanks to 300bar rated tanks. With the low-pressure tanks, you generally stay within the accepted error of ±1 percentage unit in oxygen content. With 300 bar tanks, you end up outside the ±1 percentage unit error range accepted

The error increases at richer mixes and is largest if you start filling a tank with some 50-80bar (750-1200psi) residual gas, due to the shape of the compressibility factor vs pressure graph, up to an error larger than 2 percentage units off for EAN36-EAN40. That's enough to make sensible people start questioning their analyzer or their calculations if they're not aware of the effect of compressibility.

Disclaimer: These numbers depend, of course, on the compressibility factors I used.