Steel tank and rust

Please register or login

Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.

Benefits of registering include

  • Ability to post and comment on topics and discussions.
  • A Free photo gallery to share your dive photos with the world.
  • You can make this box go away

Joining is quick and easy. Log in or Register now!

I have little comment on the technical accuracy of that graph. But from a practical perspective I can tell you its not correct. If it were I would be pumping free water into my HP steel cylinders 50% of the year or more. And that's not happening; I have ~20ppm H20 in my gas and zero rust issues.

In the US, you can have a steel cylinder bead blasted. But its typically only done on industrial cylinders at hydro shops. Can be used to remove expoxy linings in old steel 72s with so-so success.

The OP's shop needs to be replaced or educated. Then he can fix his rusty tanks.
 
I have little comment on the technical accuracy of that graph. But from a practical perspective I can tell you its not correct. If it were I would be pumping free water into my HP steel cylinders 50% of the year or more. And that's not happening; I have ~20ppm H20 in my gas and zero rust issues.

Actually, that begs the question: How much does that tiny bit of condensation in "good" air matter, if droplets are actually forming? What I'm wondering is if it's cold enough to condense, isn't the rate of oxidation of iron fairly tiny compared to room temperature? And when the tank (or droplet) warms back up, the water would evaporate again since it's so dry in there.

Another way to look at it is I don't seem to recall much rust forming on wrought iron frozen in a block of ice. It's only when it warms back up that it seems to be an issue.
 
Below is the graph in captndale link. I have been searching for a better graph or table. The resolution of the picture doesn’t allow me to enlarge it in order to make it easier to read. It also only goes to 3000 psi.
Do you have a better source?

No, I calculated the 30F myself. This is such a well known standard in Europe that I was not surprised when my calcs showed the same result for CGA Grade E using the revised content of 24 ppm water vapor. The Europeans actually publish two standards for moisture content. Air pumped into 4500 psi tanks is required by regulation to be dryer than that for 2900 psi. The reason is obvious, to meet the 0 deg C saturation criterion for both classes. If that graph shown by Dale had indicated 35F @ 3000 or 40F at 3500 I would not dispute it but 50F @ 3000 is definitely inaccurate. By the way, the CGA "E" is similar to the Euro requirement for 200 bar but the spec for 300 bar cylinders is more stringent than Grade E. If I recall correctly, the 300 bar doc specifies <15 mg/ m3 which is down there in the Sahara.

This is all old news as I have presented the calcs on SB previously along with a ton of other stuff on the subject.

Pesky
 
Last edited:
It has been a while and my memory is not so great so I took another look at Grade E and how it compares to EN 12021 or whatever it is. Again, if I recall correctly, these allow 25 mg/m3(200 bar tanks) and 15 mg/m3 (300 bar tanks). I calculated grade E to be 18 mg/m3, a bit drier than I thought.

The way that this is done is to convert 24 ppm (v) to 15 ppm (w). The total weight of air in a 12 liter, 210 bar tank is about 3000 grams which represents 2.5 m3 of free air. Multiply 3000 X .000015. = .045 grams or 45 milligrams. Divide 45 mg by 2.5 and get 18.

The volume of air (1 atm) inside this tank is about 2.5 m3 and this air contains 0.045 grams of water. The amount of water per cubic meter at 210 bar is 210 X .045 /2.5 = 3.8 grams per cubic meter. This is saying that the air inside the tank contains 3.8 grams per cubic meter at full pressure. This looks like a lot but the tank's internal space is only 0.012 m3, so relax and remember that our itty bitty tank can only hold 0.045 grams as stated above. Disregarding pressure at this point, if you look up this absolute humidity on a humidity calculator, 3.8 grams/m3, it represents a saturation point of just above 32 degrees Fahrenheit. So, the dew point inside this tank is down at the freezing point. Calculator below:

Humidity Calculator
 
Disregarding pressure at this point, if you look up this absolute humidity on a humidity calculator, 3.8 grams/m3, it represents a saturation point of just above 32 degrees Fahrenheit. So, the dew point inside this tank is down at the freezing point. Calculator below:

Humidity Calculator


You can’t disregard the pressure. Air basically holds less moisture under pressure and therefore it reaches saturation (dew point) at a higher temperature. The link you are using is based on the standard psychrometric chart which is only valid at pressures very close to one atmosphere.

That is the reason the European require dryer air for higher pressures. I do have one article on this subject (coincidentally from Europe), but I am looking for more generic dew point versus pressure table, similar to the picture shown.

Most of the psychrometric data is only available for low air pressure since it is mostly used in the HVAC (heating ventilation and air conditioning) business.

My thermodynamics is fairly rusty…it has been years since I have done any work with it, but it has always been a interesting subject.
 
You can&#8217;t disregard the pressure. Air basically holds less moisture under pressure and therefore it reaches saturation (dew point) at a higher temperature. The link you are using is based on the standard psychrometric chart which is only valid at pressures very close to one atmosphere.

That is the reason the European require dryer air for higher pressures. I do have one article on this subject (coincidentally from Europe), but I am looking for more generic dew point versus pressure table, similar to the picture shown.

Most of the psychrometric data is only available for low air pressure since it is mostly used in the HVAC (heating ventilation and air conditioning) business.

My thermodynamics is fairly rusty&#8230;it has been years since I have done any work with it, but it has always been a interesting subject.

The Euros require drier air at higher press because the higher the final pressure the more water could potentially be pumped into a given tank volume. They are trying to control the total amount of water pumped into the tank so as not to exceed the total weight of water allowable for that volume. It has nothing to do directly with pressurized air not being able to hold moisture. High pressure air has the potential to impart energy to water molecules which would lower the dewpoint if anything but the effect should be small due to declining tank press at low temps. Relative humidity is based solely on water vapor per unit volume and temp, at least at these pressures. That is why RH tables are volumetric, weight per unit volume (absolute humidity vs temp). However, this definition may have a slightly ragged edge to it, something which I don't regard as being of practical import.

I've not discussed enthalpy because this is Scubaboard; but I have noticed that some Europeans have mentioned a dewpoint of -11C in regards to high pressure cylinders in spite of the fact that the EN 12021 target is about 0C. I can only speculate that they are implying that the actual dewpoint is lower than the target for some rather obscure reasons indirectly related to pressure. I think that this is optimistic but have not looked into it primarily because it has been demonstrated that Grade E and 12021 standards positively prevent rust in tanks within the normal range of temperatures down to the freezing point.
 
Last edited:
Here is one article that I mentioned:
http://www.hse.gov.uk/research/rrpdf/rr427.pdf

Here is a quote from it:
1.2 Water condensation with pressure

In a gas mixture of which water is a constituent, the water vapour, as with any other
gas or vapour present, will comply with Dalton’s Law; the water vapour pressure, as
a partial pressure, will increase with increasing absolute pressure of the mixture.
However, as the pressure increases, the water vapour pressure (partial pressure)
will eventually equal the saturated vapour pressure and water will start to
condense., Further compression of the mixture (increased pressure) will not raise
the water vapour pressure and excess water will condense. If the ambient
temperature is below the freezing point of water it will be deposited as ice – it
should be noted that the freezing point of water reduces with increasing pressure;
falling from 0 °C at 1 bar to -9 °C at 1000 bar.
To avoid the formation of liquid water or ice within a pressurised system the
specified water content of the gas (e.g. compressed air) needs to be such that, at
the intended working pressure (charge pressure) and temperature of use, the
vapour pressure will be less than the saturated vapour pressure. On this basis, the
specified level may be allowed to vary with the charge pressure, the required
(specified) level increasing with a reduction in working pressure. Similarly the
specified level may also vary depending on the minimum expected ambient
temperature; the higher the ambient temperature the greater the permissible water
content.


Take a look at the graph in page 7, also look at the text and tables in pages 8 and 9.


Here is the graph from page 7.
It shows the dew point going up with pressure for the same amount of moisture per air volume.


PressureDewPoint.jpg





Here is part of the conclusion:
5 Conclusions
Proposed guidelines for acceptable water content in compressed gases have been
derived using the Magnus equation by relating volumetric humidity to saturated
vapour pressure and pressure dew point.
The water content currently specified for compressed air at pressures greater than
40 bar (BS EN 12021) is likely to result in free water condensing within a
compressed gas system at the expected ambient conditions of use.
Consideration should be given to reducing the maximum permissible water content
in compressed air to a maximum of 20 mg·m-3 (at 1.013 bar, 20 °C) for air at
pressures up to 200 bar and to 15 mg·m-3 (at 1.013 bar, 20 °C) for air at pressures
up to 300 bar.



It looks like a very good report, but I am still looking for more data.
 
As one can see, the water content was expressed at atmospheric pressure. Naturally, as air with variable water content is pressurized, the dewpoint will be raised accordingly. Not exactly rocket science but these are the kinds of power point presentations that were used to convince the Brits and the Americans to lower the permissible water content of compressed gas. The reduction to 15 mg/m3 was pretty radical and entailed some industry cost so the case had to be well supported. Now, do you honestly think that the presenter was about to argue for a terminal dewpoint of 50F? Not likely. I would not wish to be the one to field the group's questions after they heard that. Giving your attention to the above graph one can see that a moisture content of 15 mg/m3 yields a dewpoint of 0C (32F) at 4400 psi. From the same chart, the dewpoint of Grade E (18mg/m3) is about 2C at 4400 psi, slightly better than I thought.

A dewpoint of 50F is what they had been dealing with up to that time, that is why the moisture criteria were lowered in EN 12021 and CGA-6.(I calculated the dewpoint of the old Grade E standard to be 53F @ 3000 psi). As a concession to cost and practicality, the Brits split the maximum water contents to allow for cylinders rated for different pressure ranges. However, I doubt that the 0 deg C dewpoint was negotiable, not to the upside anyway. The Americans opted for one set of numbers to cover pressurized cylinders subject to DOT. It is more stringent than the European high number and a little less stringent than the low number, but close.
 
Last edited:
I am going to jot down some thoughts on this subject of moisture in tanks. I understand that most people don't understand what I am talking about but perhaps if this information is held in SB's memory it will provide some kind of reference for the future. The discussion above is more important but less interesting than this question; who decided that CGA-6 needed to be changed and why?

At first blush it would seem that divers would be most interested in this and would be the community to provide leadership. After all, wasn't it the divers who conducted the tests which demonstrated that water inside tanks could be extremely hazardous and was it not the divers who started the visual inspection programs?

Well, after looking at the history, it was the firemen who pushed for modifications to breathing air standards. When one ponders this, firemen are naturally disposed to seek out and rectify safety issues. As far as supplying compressed air for breathing units, it is the firemen themselves who are responsible for charging and maintaining breathing apparatus. Anytime that their old steel SCBA tanks had to be tumbled or scrapped due to rust it represented a cost and potential hazard to their membership. Importantly, they are organized in the Firemen's Association.

On the other hand, diving organizations profited as an unintended consequence of pumping wet, dirty air. These organizations focused on fixing problems by conducting inspections, tumbling tanks and selling new tanks. One might presume that it never occurred to them to fix problems at the front end. It could be argued that diving is different in the manner in which the equipment is used and the environments to which the equipment is exposed. Indeed, the first destruct tests of SCUBA tanks included filling them partially with salt water and compressed air; then sitting back to see what happened. What happened wasn't pretty. So, we got the inspection regime but who was looking at the quality of air that the shops were delivering? Who would benefit and who would bear the cost of cleaner air? Diving organizations are diverse, uncoordinated and profit driven, and do not necessarily speak for the "membership". For the most part, they represent industry and treat divers as "consumers".

In the case of the firemen it was clear that they would suffer a cost from use of humid air but that the cost of upgrading air compressors would be balanced by reduced maintenance and capital costs of SCBA tanks. With respect to SCUBA shops the cost of cleaning up air would be mostly on the side of upgrading and maintaining their compressors. On the benefit side would be rental tanks but the solution to the original rust problem was already being implemented, to replace steel with aluminum which is less subject to corrosion. So, little change there. Also, more aluminum tanks were already being sold to customers and this continues today for various reasons, mainly price and due to the perception that aluminum is less subject to corrosion. Back in the day, the deal was something like this, "buy an aluminum tank and it won't rust after we pump it full of wet air." After improvements to the air supply, all tanks continued to be subject to inspections and this was emphasized after the neck crack scare. Aluminum tanks became very cheap due to the huge number produced and sold even though the original argument for them is less clear due to cleaner air at the shops and in general from the change in the Grade E standard. One could speculate as to whether aluminum tanks would have become so wide spread if shop air had been cleaner in the first place.

The sequence of events is a bit more complex than I've described so far. The fact is that compressor filtration technology had begun to evolve as far back as the 1970's. By the early 1990's, the actual compressed air delivered by many shops and individual machines was already superior to the existing standard. This may have been driven by occasional reports of storage banks exploding. Liability concerns could have surfaced among the compressor manufacturers. However, IMO, the basic thesis, that the dive organizations were more interested in triage than preventive medicine, is intact.

In conclusion, differences in economic interests and different approaches to safety concerns put the firemen out front in the initiative to upgrade compressed breathing air and to modify the moisture content of Grade E.

Next, I'll talk about today's sources of rust and corrosion and how visual inspectors, compressor operators and hydro facilities may be causing some of this.
 
Last edited:
The steel SCUBA tank is intended for containment of dry gas. Any violation of this principle leaves open the possibility of wetting the inner tank walls and resultant rusting of the metal, a condition which is potentially dangerous including explosions and failure of the diver's air supply. The presumed purpose of conducting period inspections which breach the air tight seal of the tank is to promote safety. The inspector looks for contamination and rust, especially the form which includes pitting of the steel. Any other considerations are secondary to that.

To this end, to prevent rusting, some positive air pressure is maintained in the tank at all times and when refilled, the tank is charged only with dry air or gas. However, during inspection the valve is removed and atmospheric air may enter. A typical Summer dewpoint for this air would be 60F which is equivalent to 18g/m3 @ 80F. This means that the air inside the tank now has a dewpoint of 60F instead of the recommended 30F. After filling to 3000 psi, the total moisture is 21.8 g/m3 equal to a dewpoint of 72F. When diving chilly seas some of this internal water condenses.This amount of moisture (.25 gram inside the tank) could be expected to keep the tank wall damp and to produce some minor browning or powder which would be barely perceptible. On the other hand, during the dive, some saturated air is blown off. It is difficult to know how much of the total moisture leaves because some of it is adhering to the tank wall, but some of it leaves. After the dive, all or part of the neat water evaporates and the dewpoint rises but is less than before. That is modified by incremental additions of moisture from the compressor to a tank which already has near saturated air . Each time that the tank if filled about .045 g water is added to the humid air and each time that the tank is emptied a little water condenses during the dive and some humid air leaves. If the air had not condensed in the first place, or if it had condensed and subsequently evaporated, there would be no problem. It depends on the dewpoint of the air during the inspection, the sea temp and the air temp. When the air is not saturated before and after filling, normal discharge of the air pressure would just carry away the water vapor. Not so if there is neat water inside, however small. Even so, over time, the humidity inside the tank should be a little less after each dive and the dewpoint a little lower. One question is "why put up with this in the first place"? The solution to this is to prevent the tank's inner volume from ever being exposed to a condition of saturation and preventing the incremental effects.

It's called purging. Following an inspection, pump the tank to 200 psi with dry gas and empty. Refill to rated pressure with dry gas and this problem, however rare, will not occur.
 
https://www.shearwater.com/products/swift/

Back
Top Bottom