Questions about LP72 tanks

[guruboy]

Luis calculated his own REE values using measurements he took. None of his numbers will apply to your tanks.

The best you can hope for is to use the "official" PST number and hope that is accepted by your hydro shop for determining if your tanks qualify for the + rating. The PST REE number applies only to PST tanks.

 

[runsongas]

REE varies tank to tank. the 58.4 is a low cutoff intended to cover all tanks eg no tanks would have a REE value lower than 58.4

A tank in good shape qualifying for the + generally does not come close to the REE, so you would not see a 50% failure rate for the + unless if all the tanks were maxing to their REE.

 

[Luis H]

OK I know this is an old thread, but I've recently found a hydro tester that MAY + stamp my old 72's.. I'm a bit confused on the stated PST ree value 58.4 and the values found above which are greater than 58.4. I thought to get the +, the tank had to have a value LESS than the posted value??
In addition to that - if the average REE is determined by a set of tanks, is there a % less than average that allows the pass as using the average, 50% of tanks would fail to pass...

Anyone have time to explain what it is I'm missing? I now have 14 steel 72's. Half are in hydro, half have been sitting for over 15 years but most look great, I just haven't taken them in yet. Now that I may be able to get a +, the ones that have not seen use, may see it soon (as I don't want to re-hydro tanks that are still good for a few years).
Thanks!

Don’t confuse the EE (Elastic Expansion) number with the REE (Reject Elastic Expansion) numbers.

The EE value is what your re-qualifying facility will measure for every cylinder during the hydro test. They will compare the measure EE value with the REE value in order to qualify the cylinder for the “+” stamp.

The REE value can come from the manufacturer as a general value that applies to a large group of cylinders or you can have a more specific REE value that applies to some or just even one specific cylinder.

I calculated very specific REE values for a number of cylinders. To do that calculation I had to take many specific measurements including wall thickness (using ultra sound equipment) and the specific volume of the particular cylinder by measuring the weight of the water that the cylinder held.

The publish REE number is a relative conservative number and as expected it would be a lower threshold than the specific numbers that I calculated for each cylinder.

The measured EE numbers (during hydro) for all of my cylinder were always lower that the published REE number and therefore even lower than the specific REE number for the particular cylinder. They are all “+” stamped.

 

[ScubaBunga]

well so this was in a previous post (and I don't have the CFR to validate against):
"I have not read the CFR in a while, but I think it only required a sample of 3 cylinders (from a particular type: same size, manufacturer, pressure, etc) to determine the REE number for that type of cylinder. From what I recall, it did not required them to be from the same batch or year or anything similar, just the same type."

if that is true, I should be able to use luis's calculations for any pst and norris tanks (up to hydro people's approval)???

BUT if the published ree is "a relatively conservative" number - how can an average from a sample set be used as an average will have 50% not meet the mark unless that value has a buffer/% calculation in it to set it so "healthy" tanks pass.

And in other posts it's stated the tank has to be LESS than the REE? But from above it sounds like tanks would have values HIGHER than the REE.

As I'd love to have all my old tanks get a +, I'm trying to fully understand the details. Thanks.

 

[runsongas]

read luis' post again. during testing, the expansion value you receive from hydro testing of your tank should not be close to the REE if it is in good shape. so the difference between 58.4 and 61.5 or whatever becomes immaterial. just use the published REE unless if you want to measure wall thickness and do the manual computation for the REE. reiterating again, the measured elastic expansion during hydro and the REE are completely separated values.

https://law.resource.org/pub/us/cfr/ibr/003/cga.c-5.1991.pdf

using a simple analogy, the REE is a cutoff value like if we say someone over 250 pounds is fat. but in a random group, that doesn't mean everyone is going to be from 240 to 260 with half less than 250 and half over. you would get a wider range with people ranging from as low as 100 soaking wet up to a 300 pound bodybuilder, but most people would be under 250.

 

[Luis H]

well so this was in a previous post (and I don't have the CFR to validate against):
"I have not read the CFR in a while, but I think it only required a sample of 3 cylinders (from a particular type: same size, manufacturer, pressure, etc) to determine the REE number for that type of cylinder. From what I recall, it did not required them to be from the same batch or year or anything similar, just the same type."

if that is true, I should be able to use luis's calculations for any pst and norris tanks (up to hydro people's approval)???

BUT if the published ree is "a relatively conservative" number - how can an average from a sample set be used as an average will have 50% not meet the mark unless that value has a buffer/% calculation in it to set it so "healthy" tanks pass.

And in other posts it's stated the tank has to be LESS than the REE? But from above it sounds like tanks would have values HIGHER than the REE.

As I'd love to have all my old tanks get a +, I'm trying to fully understand the details. Thanks.

I wrote that statement a while back and I just did a quick search on CGA C-5 2005 edition (that is the newest one I have). I could not find where I got that from. It is buried somewhere, and as far as I remember, it wasn't entirely clear how to implement it.

Here is a quote from section 4 (CGA C-5 2005):

4 Rejection elastic expansion values

Rejection elastic expansion (REE) values may be used from the following sources:

• REE stamped at the time of manufacture. Normally, calculated by the manufacturer during the design of the cylinder;
• REE obtained from the manufacturer. Calculated by the manufacturer. but not stamped on the cylinde
• REE found in Table 1 for the cylinder size and design to be evaluated; and
• REE determined by experimental testing and measurement of typical sample cylinder representative of a manufacturer's design. Average wall stress shall be used. See Section 5 and 6.

CGA C-5 is available on line and it is not that expensive if you just buy a downloadable PDF copy.

The equations are fairly straight forward, but there are some judgment calls on how to obtain some of the data. It seems straight forward at first, but for some may think that it is like reading a foreign language.

I totally understand your question about taking an average doesn't seem right in order to determine a threshold for a pass/ fail criteria. I agree that it doesn’t seem right. I can’t find exactly where I read that and I don’t remember if there was a way of dealing with the low numbers, or if it just assumes that the calculation is conservative enough. I will try to look more into it when I have a lot more time.

I have the max/ min values and the standard deviation of my sample of cylinders. All of my calculated REE values are higher than the published value by PST.

Also all my measured EE values are lower than the PST published value.

Since you have the PST value, I would not worry about my calculated numbers.


This statement puzzles me:

And in other posts it's stated the tank has to be LESS than the REE? But from above it sounds like tanks would have values HIGHER than the REE.

What values are you talking about? Are you confusing EE values with REE values?

EE values are measured during the hydro test.

REE values are calculated based on wall cylinder stress.



Again, if you have a technical background, I would recommend you buy a copy of CGA C-5. But, before you do that, you can download:

• CFR – title 49 section 178.37 (Specification 3AA and 3AAX seamless steel cylinders)
• CFR-title 49-vol 3-sec180-205 (General requirements for requalification of specification cylinders)
• CFR –title 49-vol 3-sec180-203 (Definitions)
• CFR –title 49-vol 3-sec180-213 (Requalification markings)

CFR: Code of Federal Regulations (These are the government regulations used by DOT)

CGA: Compress Gas Association (This is an industry association and some of it documents are invoked by the CFR)

The CGA can charge for its documents. The government cannot charge for the CFRs.

 

[ScubaBunga]

so for the PST, I will use the published value - but I also have several Norris tanks. Is there a published value?? I see you posted 60.1 in a very early post. Is there a document like the pst one I can use for my Norris tanks?

@Luis - THANK YOU for the detailed responses and I may have to go get those documents.

With regards to the numbers the published, PST is 58.4, but your chart show ALL PST with a REE greater than that?

But in post #28 you have these values which show REE > the pst and norris values
1 / Norris >> 0.177 inches / 30.26 Lbs / 72.2 cu ft of air (@ STP) / REE = 61.3
2 / PST >>>> 0.171 inches / 28.88 Lbs / 70.5 cu ft of air (@ STP) / REE = 58.6
4 / PST >>>> 0.181 inches / 30.11 Lbs / 70.8 cu ft of air (@ STP) / REE = 61.5
5 / Norris >> 0.179 inches / 29.73 Lbs / 71.4 cu ft of air (@ STP) / REE = 60.7
6 / PST >>>> 0.173 inches / 30.43 Lbs / 70.9 cu ft of air (@ STP) / REE = 58.7
7 / Norris >> 0.178 inches / 31.01 Lbs / 70.3 cu ft of air (@ STP) / REE = 58.4
8 / Norris >> 0.175 inches / 29.73 Lbs / 71.4 cu ft of air (@ STP) / REE = 61.2

Is it that you calculated a REE (use tank 6 as an example) which came to 58.7 - but when the tank is actually tested, the tested value is (for a tank that passes) generally well below 58.4(the published value? Therefore is the average is allowed based on a sample set (if it can be found that that is allowed), then the average should be good because it is well above a fail rate?

 

[Luis H]

I calculated the REE number each one of my tanks based on the average wall thickness that I measured. The average wall thickness of all my cylinders is higher than the minimum allowed.

PST would probably have calculated their REE value based on the minimum wall thickness allowed for that class of cylinders.

The most recent EE values I have for those same 8 cylinders are as follow (most recent hydro test was in 2015):
1) 57.4 cc
2) 55.8 cc
3) 53 cc
4) 54 cc
5) 56.3 cc
6) 53.4 cc
7) 54 cc
8) 55.8 cc

The data collected during hydro test is the "total expansion" and "permanent expansion" (also known as residual expansion due to plastic deformations).
The difference between those two numbers is the elastic expansion (EE).

So here are all the numbers for the 2015 hydro test (all values are in cc):
"Total expansion" >> "Permanent expansion" >> "Elastic Expansion

1) 57.6 >> 0.2 >> 57.4 cc
2) 56.4 >> 0.6 >> 55.8 cc
3) 55.4 >> 2.4 >> 53 cc
4) 56.2 >> 2.2 >> 54 cc
5) 58.5 >> 2.2 >> 56.3 cc
6) 57.0 >> 3.6 >> 53.4 cc
7) 56.4 >> 2.4 >> 54 cc
8) 57.8 >> 2.0 >> 55.8 cc

My personal records have been verified, but I typed this numbers above by hand so there may be some typos.

Notice that even the total expansion in all cases is numerically less than the REE numbers.

Also you may notice that cylinder number 6 has the highest permanent (plastic) expansion as well as the lowest calculated REE number. That is probably not a coincidence. Its average wall thickness is a little bit less than most, not all.

 

[ScubaBunga]

I calculated the REE number each one of my tanks based on the average wall thickness that I measured. The average wall thickness of all my cylinders is higher than the minimum allowed.....

I appreciate all that info and I think I now get it, Do you or anyone have any published Norris info? I"m still waiting on the hydro place to see exactly what they need. Apparently they do many old tanks and have a list (maybe similar to yours) so I've passed on my info to see if they can match it and + my tanks!