Tank Storage.

[The Chairman]

I have seen tanks left in the sun and either the burst disk went or the o'ring under the valve extruded.

Hmnnnnn, a burst disc is rated to 5/3s of a tank's pressure. Ergo a 3000 psi tank's burst disc should not rupture until it sees 5,000 psi. A tank's internal pressure will increase proportionally due to increased absolute temperature, Rankine or Kelvin (not C or F). Consider a tank filled to 3000 at 80F. It would have to be heated to 440F in order to reach that 5000 psi. I've never seen the sun heat a tank that high. Hell, I've never seen one hot enough to boil water (212F)! Here's my math. Please check it to see if I've made a fool of myself:

• Convert Fahrenheit to Rankine
  • 80F+460F=540R
• Multiply Rankine by 5/3s
  • 540Rx5/3=900R
• Convert Rankine back to Fahrenheit
  • 900R-460F=440F

Of course, there are some burst discs that are compromised and will not take the full 5/3s. They need to be replaced anyway!

 

[JackD342]

Hmnnnnn, a burst disc is rated to 5/3s of a tank's pressure. Ergo a 3000 psi tank's burst disc should not rupture until it sees 5,000 psi. A tank's internal pressure will increase proportionally due to increased absolute temperature, Rankine or Kelvin (not C or F). Consider a tank filled to 3000 at 80F. It would have to be heated to 440F in order to reach that 5000 psi. I've never seen the sun heat a tank that high. Hell, I've never seen one hot enough to boil water (212F)! Here's my math. Please check it to see if I've made a fool of myself:

• Convert Fahrenheit to Rankine
  • 80F+460F=540R
• Multiply Rankine by 5/3s
  • 540Rx5/3=900R
• Convert Rankine back to Fahrenheit
  • 900R-460F=440F

Of course, there are some burst discs that are compromised and will not take the full 5/3s. They need to be replaced anyway!

Sounds about right to me. A hot car failure of a burst disc that is installed/replaced correctly and according to manufacturer directions, is far less likely to occur than is condensation inside a cold storage pressurized cylinder.

I knew that was a theoretical, but until I did the math to reply earlier in this thread I didn't have a solid concept of the specifics. Be careful of the guidance that simply says "here's what I do." Those cylinders may be routinely collecting condensation inside without any way of knowing that it is happening. The physics on the dew point if not a "maybe" scenario, it is a definite, with the moisture content of the air inside the only unknown variable that adds some fuzziness to the math. For most folks, that top-off fill at the start of the season is going to cost the same whether the cylinder has 2500 psi in it or it has 200 psi in it. So no harm in keeping just a few hundred psi when storing cold rather than trying to decide what the "right" low pressure is.

 

[The Chairman]

The dew point increases dramatically with pressure. I don't have the formula for it, but it has to do with absolute pressures and temperatures.

 

[JackD342]

The dew point increases dramatically with pressure. I don't have the formula for it, but it has to do with absolute pressures and temperatures.

There was a link to a calculator in that other thread I linked to. It agreed with the dewpoints listed for moisture ppmv at ambient pressure on my compressor test results, so I felt okay with using it as a "source of truth."

I would be curious if anyone else has any past samplings of moisture content from compressor tests to understand how much that might vary. I'm thinking a compressor that is not in an A/C space may be leaving more moisture in the flow, especially on a warm humid day. Of course, if you are diving frequently it is less likely that you are cold storing a cylinder at the end of the season that was filled from air banked on a warm humid day. Or maybe it is run in a space that also contains a pool and the intake has not been routed elsewhere?

Here is the calculator to save a couple of clicks:
FREE Dew Point Calculator | Alpha Moisture Systems

 

[TrimixToo]

There was a link to a calculator in that other thread I linked to. It agreed with the dewpoints listed for moisture ppmv at ambient pressure on my compressor test results, so I felt okay with using it as a "source of truth."

I would be curious if anyone else has any past samplings of moisture content from compressor tests to understand how much that might vary. I'm thinking a compressor that is not in an A/C space may be leaving more moisture in the flow, especially on a warm humid day. Of course, if you are diving frequently it is less likely that you are cold storing a cylinder at the end of the season that was filled from air banked on a warm humid day. Or maybe it is run in a space that also contains a pool and the intake has not been routed elsewhere?

Here is the calculator to save a couple of clicks:
FREE Dew Point Calculator | Alpha Moisture Systems

A compressor will usually have fully- to super-saturated air reach the coalescer no matter whether the input air is air conditioned or not. My compressor is pretty consistent in the moisture content shown by testing no matter when I run it. Hot days, cool days, low humidity, high humidity...none seem to make much difference on test day. What did make a difference was turning up the turnoff pressure and upgrading the coalescer, which taken together made the test dew point drop by over 10F, IIRC.

Someone else has posted that they added a refrigerated cooler, which got his compressor into "extra dry" territory. I don't recall the dew point but it was pretty impressive.

 

[bamafan]

Hmnnnnn, a burst disc is rated to 5/3s of a tank's pressure. Ergo a 3000 psi tank's burst disc should not rupture until it sees 5,000 psi. A tank's internal pressure will increase proportionally due to increased absolute temperature, Rankine or Kelvin (not C or F). Consider a tank filled to 3000 at 80F. It would have to be heated to 440F in order to reach that 5000 psi. I've never seen the sun heat a tank that high. Hell, I've never seen one hot enough to boil water (212F)! Here's my math. Please check it to see if I've made a fool of myself:

• Convert Fahrenheit to Rankine
  • 80F+460F=540R
• Multiply Rankine by 5/3s
  • 540Rx5/3=900R
• Convert Rankine back to Fahrenheit
  • 900R-460F=440F

Of course, there are some burst discs that are compromised and will not take the full 5/3s. They need to be replaced anyway!

They will blow. I have seen it several times. Generally it is cave filled steels that I recall. Like I said both the burst disk and o'ring extruding out under the valve.

 

[The Chairman]

They will blow.

Only if they are already compromised. I've had two blow while the tank was being filled. Both were LP steels (a 72 and an 85). They were no where near their 5/3s pressure limit and no sun was involved. Poof. I think my underwear still has the skid marks. I still remember those incidents vividly.

 

[michael-fisch]

They will blow. I have seen it several times. Generally it is cave filled steels that I recall. Like I said both the burst disk and o'ring extruding out under the valve.

Over 20 years ago I was trained that a cave fill usually involves a diver with enough common sense to replace the burst disk with a piece of stainless steel shim stock that will not fail. If the tank gets really hot the tank neck O-Ring should melt and fail long before a new burst disk will fail.
Since the move away from tape-sealed taper thread valves to O-Ring sealed tank necks over 50 years ago, there is no sane reason to use a burst disk, except for 50 year out of date DOT regulations.
In europe, nobody uses burst disk valves and our tanks still don't detonate in a house fire.

Michael

 

[martincohn]

My tanks live in the house in the corner of the spare bedroom with all of my other dive stuff...much to my wife's butthurt...er dismay.

 

[Diver below 83]

All my stuff stays inside the house in ac.