Air Fill Protocol

[mictrik]

It does take time for heat energy to transfered to the outer part of a metal cylinder. I usually try to make sure the tank is actually cool when it is finished the filling process. I say process because I might stop filling or slow down the fill if it is getting what I consider to be too warm.

---------- Post Merged at 05:11 AM ---------- Previous Post was at 05:01 AM ----------

The engineers have done a good job in devising the service pressure and service life of these cylinders. I believe it was expected that these cylinders would be overfilled from time to time. My point is that exceeding the service pressure by lets say 10% when it is hotter than 70º F is not forbidden by DOT guidelines. I would expect that the engineers placed this process into their calculations when designing these cylinders. Judging by the service life the steel 72's have provided thus far they did well, for these cylinders were commonly over filled not just when at higher than room temp (during fill process to achieve rated service pressure at 70º) but also to greatly increase air volume at room temp for extended diving.

---------- Post Merged at 05:13 AM ---------- Previous Post was at 05:01 AM ----------

I think allot of short fills happen from this or slightly leaking valves. We try to test each cylinder that goes out when the customer picks it up. The customer sees the pressure and then knows what he purchased. Keep a pressure gauge behind the counter just for this purpose...

---------- Post Merged at 05:39 AM ---------- Previous Post was at 05:01 AM ----------

Here is a video that shows a steel cylinder whose rated service pressure when in good condition is 227 BAR (3292) being filled until it bursts. The cylinder was failed inspection due to corrosion that is evident in the video and the cylinder does seem to rupture in area that seems to be most corroded but not until it reached a fill pressure of 580BAR (8412 PSI). Would it of withstood 10000 PSI if it was in good shape without corrosion?? Who knows but I am sure of one thing it can handle a slight & very temporary "overfill" when above room temp in order to avoid a short fill...

 

[NAM001]

DFX

I for the life of me do not understand why this is so difficult. Every tank is given a psi rating at a given temperature. I must assume thatt if a tank was filled in the winter to 2950 and it was 40 degrees f out side you would saya this is ok. Well it is not because when the tank increases tomp to 70 degrees it will be higher than 3000. The 3000 psi rating takes into account psi increase and decrease due to environmental temp. So long as it is 3000 at 70 the tank design will handle any temp variation with out straining the tank material and risking its integrity. Your ludricris example of 3500 cold (70) is too high. if the temp goes to say 150 in the car than then you are 80 degrees over rated temp and the tank increases psi by 4-500 psi. that is the problem. 3500 at 150 degrees is with in the proper fill presure when adjusted to 70 f. Now lets take that tank in your posting and fill to 3500 when it is 40f. and then let it heat up in the car to 150. Lets see 110 increase and 5-6 per thats 600psi so the tank will now be at a min of 4100. You dont do that and especially to an al tank. this scenerio is why the tank is temp rated for the psi. Lastly tanks are psi rated with a safety fator. I believe it is a 4x factor. so an al tank is rated at 3000psi based on its 12000 psi rutpure ability. The 4x factor makes the tank capable of 10000 fill cycles without failing hydro.

Why aren't AL80s filled to a "cold" 3500 psi to begin with then?

 

[spectrum]

I do wonder how many "short fills" are the result of people accidentally cracking the valve a smidgen when putting the tank into or out of their car.

I have wondered that a few times. That's one reason I now always check all fills when I accept them. If I have a significant short when I tag them at home then I know I rolled the valve or perhaps a valve is not seated right after a VIP.

 

[DA Aquamaster]

Let me start with "this thread generally pisses me off" I couldn't stomach much more than about half the first and last pages of the thread.

1) There is general confusion about what "full" means with a scuba tank.

A full tank is one that does not exceed it's service pressure (or the 10% overfill allowed with a plus rating) once it has cooled to room temperature - generally regarded as 70 degrees F. This means the shop can fill a 3000 psi tank to 3200-3300 psi as needed on a hot fill so that it will cool to 3000 psi. This means that their is no need to worry about filling the tank and then allowing it to cool, then topping the tank - just use the same 500-600 psi per minute rate of fill and go to 3200psi to 3300 psi (depending on how warm the shoulder feels) and with minimal experience, you can get very good at nailing the full 3000 psi "full" pressure when the tank cools.

Shops that can't manage this simple skill need to hire smarter staff and/or generally get a fricking clue.

2) In most cases, however bad fills are just due to ignorance of the DOT regs and a well intended but very misguided belief that if you ever exceed the fill pressure you've "overfilled" the tank, and that this is "dangerous". But consider this - if you slowly fill a tank to be exactly 3000 psi at room temp, then set it out in the hot sun (especially a black tank) or leave it in your car, it will heat to more than room temp (say 105 degrees) and the pressure will go up. So when it does this and the tank pressure is now 3200 psi, did the tank grow more gas? Obviously, not. And when it cools to room temp it will be back to room temp and will again be just "full", not "overfull", at 3000 psi.

Oddly enough the DOT understands both physics and the real world use of tanks, so they define "full" based on an understanding that a tank that is "full" at room temp and 3000 psi will have a pressure of about 3350 degrees at 130 degrees - a temperature that can easily occur in some tank storage situations. Yet the DOT also recognizes that the tank at 3350 psi @130 degrees is not overfilled, nor is it dangerous as the engineering standard used takes that real world condition into account.

3) Burst discs are mandated by law to blow between 90% and 100% of the tank's test pressure. A 3 AL aluminum 80 with a service pressure of 3000 psi has a test pressure of 5000 psi. So the burst disc will not blow at less than 4500 psi. Now, as they age, flex, corrode, etc, the pressure at which the burst disc fails will decrease, but assuming you replace them every 5 years when you have the tank hydro tested, you won't have to worry about a burst disc failing at the 3300 psi or so you'd need to get a hot filled AL 80 actually full. You've still got a 1200 psi margin to play with with over and above the 3300 fill pressure with a fresh burst disc, and probably a 600 psi margin to play with on a 5 year old burst disc.

4) Water baths make sense in cave country where you are overfilling a 3AA steel tank to 150% of it's service pressure. Most often this means filling a 2400 psi service pressure tank to 3600 psi when the test pressure for that tank is 4000 psi. In this case, over shooting the fill pressure by a large margin may not make much sense, and the water bath is the lesser of the two evils.

The downside of a water bath is that care has to be taken to ensure that water is not lodged in the valve or the fill whip where it could then be blown into the tank. Both the valve and the fill whip should be cracked open on every fill to expel any water, but it's even more vital with a water bath given the close proximity of water and the potential for splash into the valve or fill whip. Consequently, if I am filling in any other situation, I see no benefit to a water bath as I can just pad the fill pressure as needed to ensure the tank is "full" when it cools to room temperature.

5) misquoting the CFRs and referencing sections that refer to actual hazardous materials rather than gases such as air and nitrox does not help. I realize it is confusing but you have to pay close attention to the sections and sub sections you are in to properly interpret the meaning and interpretation of a specific regulation.

 

[boulderjohn]

1) There is general confusion about what "full" means with a scuba tank.

A full tank is one that does not exceed it's service pressure (or the 10% overfill allowed with a plus rating) once it has cooled to room temperature - generally regarded as 70 degrees F. This means the shop can fill a 3000 psi tank to 3200-3300 psi as needed on a hot fill so that it will cool to 3000 psi. This means that their is no need to worry about filling the tank and then allowing it to cool, then topping the tank - just use the same 500-600 psi per minute rate of fill and go to 3200psi to 3300 psi (depending on how warm the shoulder feels) and with minimal experience, you can get very good at nailing the full 3000 psi "full" pressure when the tank cools.

Shops that can't manage this simple skill need to hire smarter staff and/or generally get a fricking clue.

In agreement with this and in full disagreement with the article that people referenced without apparently reading, I got my PSI certification along with just about all the instructors associated with our shop in a class held in our shop. The PSI instructor gave rough guidelines as to how much a tank could be overfilled so that it will cool down to its service pressure.

I also find it amazing that this article is on the Fill Express web site, since I have used that shop often and know darn well they do not follow its dictates.

 

[flyandive]

I.

I also find it amazing that this article is on the Fill Express web site, since I have used that shop often and know darn well they do not follow its dictates.

Seriously! I no longer live in Florida but when I did, I stopped going to them for fills when they filled our tanks fast and hot enough to blow two burst disks on a buddy's tanks. My disks did not burst but they were so hot to the touch I had to wait for them to cool down before I could carry them out to the car.

The best part was they refused to replace the burst disks because they were HP burst disks installed on LP tanks!

 

[boulderjohn]

Seriously! I no longer live in Florida but when I did, I stopped going to them for fills when they filled our tanks fast and hot enough to blow two burst disks on a buddy's tanks. My disks did not burst but they were so hot to the touch I had to wait for them to cool down before I could carry them out to the car.

The best part was they refused to replace the burst disks because they were HP burst disks installed on LP tanks!

The shop has new ownership since the spring of 2011, so the situation is completely different. I have never experienced what you describe.

 

[DA Aquamaster]

Seriously! I no longer live in Florida but when I did, I stopped going to them for fills when they filled our tanks fast and hot enough to blow two burst disks on a buddy's tanks.

I never really minded hot trimix fills there, but I was waiting for fills for my stage, back gas and O2 tanks. I talked to someone for a couple minutes and then turned back around and noted that many of the tanks had been cleared out so I started to move my (I thought still near empty) LP 45 closer to the fill station. That's when I noted it was almost hot enough to burn my hand on it, and sure it enough it had been filled in about 2 minutes. Scary.

 

[flyandive]

The shop has new ownership since the spring of 2011, so the situation is completely different. I have never experienced what you describe.

I certainly hope so. My last experience with them would have been at least five years ago.

 

[spectrum]

I get the impression that shops consider that higher pressure air needed to overshoot as compensation for cooling to be more expensive and don't want to deplete their banks.

As much as possible I leave them so I can get a final top-off before I take them. They are getting to anticipate this and often have them topped off and waiting when I return. This comes at the cost of some time and mileage but it not a huge inconvenience and where I get a good end result I can't begin justifying a compressor. Once upon a time I had a neighborhood shop that would do a nice creep fill to 3700 on my HP cylinders and life was easier for all.

I always check them before I leave and I have done a side by side verification against the shops tank checker gauge. It's not that I don't trust them but things happen, like a partially open tank valve that gates off some of the fill pressure. Even with perfect fills I believe in the past I may have rolled a valve and let it bleed in transit. With the fill verified I know who to blame if they don't check out later.

Once I have had the tanks is the dive locker, usually over night I will check them and mark the pressure on masking tape. This lets me cherry pick the right cylinders for the dive planned. This is especially helpful if I have settled for some insta-fills which will suffer from cooling. I have a thermometer in the location if it's significantly different than 70F I adjust my reading by #50 per 10F. Yes, I'm anal but when I get to the dive site I know what I'm working with.

If you work with the shop and manage the situation you can usually get good results but you often need to make it clear that you care. Of course there is the place across town that can do a 30 second fill!

Pete