Letting it settle

[Kev]

Is there any science behind the practice of letting a tank full of nitrox 'settle' so as to produce a homogeneous mix of oxygen and air, or is it just nonsense?

Any views?

[Roger Cardy]

IMHO I cant actually see it making any difference. The air going in will stirr things up, just like this post .

However I know people who say thay have anaylsed a fill later and got a different result so who knows. I normally check it at the shop, and again at home, and allowing for anaylser differences I have not seen any change.

Regards

Roger

[Waskey]

There is a science, but it's to do with turbulence and the maths gets a bit crazy.

[Mal Bridgeman]

Originally Posted by Kev

Is there any science behind the practice of letting a tank full of nitrox 'settle' so as to produce a homogeneous mix of oxygen and air, or is it just nonsense?

Any views?

As I understand it, it's not letting it settle but more to do with letting it mix evenly.

When you fill the cylinder from empty, it first contains pure O2, then Air is squirted in....of course it starts to mix with the O2 but as you get towards the end of filling, the rate of change of pressure reduces and I think that reduces the "mixing" effect leading to a slightly more nitrogen rich mix at the top of the cylinder.

The valve and drain tube will contain only air at this stage which is why when you analyse a tank immediately after filling you get a low reading that then goes higher as the nitrox starts to come through.

In practice the variance is very small I have found .... there is no need to shake the cylinder to assist the mixing unless it's your first day working in the diveshop ....when it becomes absolutely essential

It is a good practice to analyse your cylinder just before you use it though and that's the really important part.

HTH
Mal

[jamesp]

Considering the turbulance generated by the gas entering the cylinder, things should be well and trully mixed. However I have actually seen a cylinder analyised, by two people independantly, rejected as incorrect. Shaken up, re-analyised and found to be correct.

I would have not believed it if i had not seen it, still makes bugger all sense though!

[WillD]

It can make quite a big difference - especially if you add the air very slowly. That's why you sometimes see a quick blast being added at the start and/or end of the air being put in to create turbulence in the gas to help it mix.

Think of it this way. Make yourself a cup of coffee and leave it for 5 minutes. then pour in some milk really slowly. It can take quite a while for it to all spread through. Same think happens in the cylinder, and rolling it is like stirring.

One thing that can help is to let a couple of quick blasts out of the cylinder before analysing. This removes any air that is still sitting on top of the mix and can cause turbulence to help mix it up.

W.

[warmwaterdiver]

Originally Posted by Mal Bridgeman

In practice the variance is very small I have found .... there is no need to shake the cylinder to assist the mixing unless it's your first day working in the diveshop ....when it becomes absolutely essential

funny you should say that, I've actually seen this work

[Mal Bridgeman]

Originally Posted by warmwaterdiver

funny you should say that, I've actually seen this work

Really? ... how much did the analysis change?
Mal

[warmwaterdiver]

Originally Posted by Mal Bridgeman

Really? ... how much did the analysis change?
Mal

about 10%, was on an 80% deco gas

O2 added, air added, left to settle, anaylsed as being rich
air topped, anaylsed as being lean
rolled about for 30 seconds, analysed as spot-on

[Woz]

oxygen is heavier than air. Lie the cyl on its side, gases will mix up. Then it will analyse spot on.