Soaking solution for inside air compressor tank?

[All]

What is the best practice for dealing with surface rust inside a large air compressor tank?

Is there a common consensus regarding a soaking solution that might "convert" the interior surface rust, to slow the rust growth?

Phosphoric acid has been mentioned in previous threads, along with TSP and muriatic acid (also described by some posters as "pool" acid). Still others recommend simply water and detergent, with the detergent having the benefits of ph balance as well as a potential ability to keep the disturbed and loosened rust particles in suspension. Some posts have described multi step treatments... power washing, followed by soap and water soaking, followed by acid treatments of the various flavors described above.

It is hard for a reader to determine what the long term efficacy is of any of these suggested methods, because posters do not report back years later to say "hey, this worked for me" or "what a waste of time and expense... the tank rusted away anyway." And even if posters DID report back years later, who is to say that a favorable or unfavorable result might not have happened anyway.

Sigh...

Yet, ever hopeful, I'm still wondering if over the years people have figured out what generally works, and what generally doesn't. Kind of like we, as a species, have generally figured out the difference between live oak and poison oak, and know to preserve the former and avoid the latter. Is there a tank treatment process that is generally known to work, and likewise, a tank treatment process that should generally be avoided?

For example, one thread on exploding air compressor tanks referred to a state investigation that discovered that the tank didn't rupture due to corrosion, but actually exploded due to an explosion event inside the tank... from too much carbon generated from what they surmised was the wrong oil in the electrically driven air compressor pump. The manual to one of my air compressors talks about cleaning out the "carbonization" when switching from mineral based to synthetic based oil in the pump. This is a very interesting hazard... which naturally would effect what cleaning solutions and rust conversion solutions should be put into a tank, with some thought as to what residues these solutions might leave behind, despite thorough rinsing.

Some have talked about recoating the tank with aircraft fuel tank lining compounds and epoxies, along with POR 15 type coatings. The question there is, are these coatings simply covering the rust, which continues unabated underneath, cannibalizing new steel deeper inside the tank wall, since the former rusted surface is now unavailable to act upon. Some have said that old RUST is the best inhibitor of new rust, since the oxidation of old rust somewhat serves to protect new rust from growing. I'm not sure about this, since it seems that deep rust pits form when rust is left unchecked.

So what to do?

Replacing the tank is always an option, but what if the tank to be treated is already relatively new? And what if the brand new tank is delivered with rust already inside, just from life between manufacture, distribution, and storage until sale?

Some people talked about putting gravel in their tanks and rolling it around. This seems impractical for an 80 gallon tank with an oblong rectangular compressor motor base that exceeds the circumference of the tank.

Some people have successfully used air compressors dating back 40, 50, and even 60 years old, without issue. It has been postulated that the older tanks were made thicker than the newer tanks of today.

It seems like a good idea to have an ultrasound or some other high tech thickness measuring done, but I'm fairly confident that the Made in USA code welded ASME tank I'm interested in "soaking with solutions" is currently thick enough, and would pass not only a thickness test, but a hydrostatic pressure test as well. What concerns me is down the road, a few years from now.

Is there a known good, as well as known safe, as well as known to be effective tank treatment regimen that can slow down the rate of corrosion inside?

(Thanks for listening to my overly long first post, but I've done a lot of reading here on the Journal before posting... and am just having trouble drawing a consensus based conclusion from the many directions those various threads have turned.)

 

[Garage5.9]

When in doubt buy a new tank , X amount of $ isn't worth your life or getting injured for life IMO

 

[NOZZLEMAN]

If it's an ASME tank; then no hydro testing is required and the material requirements are very ridged. A surface coating of rust would not worry me in the least. If you need some peace of mind find a pal with a fiber-optic inspection camera and have a look inside. At standard pressures around 150psi this thing will last forever.
I come from a fire service background and am familiar with high pressure (6500psi) breathing air fill systems and compressors as well as my last 8 years I ran the FD shops and had shop air compressors to maintain
Just my 2 cents & good luck

 

[sberry]

I agree, find somthing else to worry about.

 

[CNGsaves]

Once you put in anti-rust solution and got any rustflakes out, couldn't you paint interior of tank with appropriate paint??

There's lot of YouTubes out there to convert 20 lb propane tank to portable air tank and they traditionally paint the interior of tank. However, lot easier to swirl around paint in a small 5 gallon propane tank (ie convert to air tank) than it is to swirl around liquid paint in 60 gallon air compressor tank (ie to coat the entire interior surface).

Get a fiber optic snake camera and take a look inside tank to see what you really have - - - ie determine whether tank's worth saving.

 

[White 99]

Would a squirt of WD 40 or something like that be good as a PM move now and then?

 

[Fixnair]

Been dealing with air tanks for 50+ years and have never had a request to coat the inside of a tank or to remove the rust.
If you are worried about the tanks reliability you can hydro stat it yourself. All you need is water a grease gun and an appropriate sized gauge. You want to test it at 1.5 times the data plate max working pressure. For a 200 PSI tank take it up to 300 PSI.
Plug all the openings and make sure they are leak tight. Install your gauge on one of the ports where you can see it.
Tip the bare tank so it is standing upright. Now the baseplate is no longer Horizontal but vertical. Most tanks have a large bung plug in the end of the tank. Remove one end before you tip it on end and make sure the open end is up.
Fill the tank with water, completely full. It is important that there is no air in the tank. It will become very heavy so you need some way of securing it. Be careful.
Now, install a bushing in the open bung down to say 1/4" pipe and screw a grease zerk fitting into the 1/4" opening again being sure to fill the added pipe fittings with water.
Good so far? OK! Now here is where the magic happens. Your hand powered grease gun can develop pressures up to 3000 PSI. Now attach your grease gun to the zerk and begin pumping. Remember we are pumping a liquid which is not compress able so should something fracture, the only thing that will happen is a slow release of water.
Pump it up to your final test pressure and let it set for awhile. Any fracture will be evidenced by a water leak.

 

[Chadwilliam1]

that is very clever!

 

[compo7000]

Great post Fixnair

 

[WhoWhatNow]

Fixnair that is awesome. Thanks for ousting the procedure. I never would have thought to use a grease gun to apply the pressure. Why do you say to turn the tank vertical?

 

[fflintstone]

Been dealing with air tanks for 50+ years and have never had a request to coat the inside of a tank or to remove the rust.
If you are worried about the tanks reliability you can hydro stat it yourself. All you need is water a grease gun and an appropriate sized gauge. You want to test it at 1.5 times the data plate max working pressure. For a 200 PSI tank take it up to 300 PSI.
Plug all the openings and make sure they are leak tight. Install your gauge on one of the ports where you can see it.
Tip the bare tank so it is standing upright. Now the baseplate is no longer Horizontal but vertical. Most tanks have a large bung plug in the end of the tank. Remove one end before you tip it on end and make sure the open end is up.
Fill the tank with water, completely full. It is important that there is no air in the tank. It will become very heavy so you need some way of securing it. Be careful.
Now, install a bushing in the open bung down to say 1/4" pipe and screw a grease zerk fitting into the 1/4" opening again being sure to fill the added pipe fittings with water.
Good so far? OK! Now here is where the magic happens. Your hand powered grease gun can develop pressures up to 3000 PSI. Now attach your grease gun to the zerk and begin pumping. Remember we are pumping a liquid which is not compress able so should something fracture, the only thing that will happen is a slow release of water.
Pump it up to your final test pressure and let it set for awhile. Any fracture will be evidenced by a water leak.

thanks for saving me the typing, I have posted this at least twice here.

 

[All]

Thanks for the constructive post Fixnair.

I recall reading other posts in the archives of TGJ over the years that recommend the same procedure.

One cautionary concern that came up in those earlier threads was distending the un-braced sheet steel of the tank up to or beyond it's yield point. The argument, as I recall, was that since water is relatively incompressible, the hydraulic pressure applied to the large unsupported expanse of 3/16" sheet steel is enough to distend the tank at the areas of least resistance. Once the tank has distended beyond the yield stress point where it does not return to it's original shape, then the tank has yielded, and becomes more vulnerable to abrupt and radical tensile failure.

In further research, I read that in addition to a pressure guage, some testers have added some type of hydrometer that has a graduated scale in a manometer type tube that indicates if the same amount of water returns up to the scale marks on the tube. The goal is to see if the internal volume of the air compressor tank being hydro tested changes after the pressure is released. If the additional water used to elevate the pressure of the tank refills the test tube all the way back to it's original mark, then the original volume of the tank has not changed. If the water level does not restore back up to the mark, then the original volume of the tank has increased due to the test, and therefore the tank has yielded.

Here we have a situation where the tank did not spring any leak, but yielded under hydraulic pressure, which ruined the tank. It seems reasonable to add this volume measuring element to the test.

 

[RCStocker]

Spot on.
New tanks can cost almost as much as a compresor depending on the quality.

 

[cryan]

surface rust on pressure vessels is best removed by mechanical means such as needle scalers however I know that on a small domestic tank that will be impossible. Here in the UK there are quite strict regulations on pressure vessels in industrial use and recommendations for home use. You should try and reduce the water in the tank as prevention is easier than cure. Fit a good dryer and drain the tank daily. There is a coating system called apexior which gives a water resistant rubber like coating but I know that over here surveyors don't like to see it used as it makes it harder to check shell thickness with ultrasonics etc. Be very careful with air receivers as they can be deadly if abused.

 

[Fixnair]

Fixnair that is awesome. Thanks for ousting the procedure. I never would have thought to use a grease gun to apply the pressure. Why do you say to turn the tank vertical?

Because of its shape, turning it end up insures that no air is trapped inside. Any entrained will cause expansion upon fracture. You don't want expansion.

And, Mr. All, not doubting your premise but a couple of government inspection agencies have contracted me over the years to do hydro stats because they don't have the necessary tools. They have never been concerned with this distension you refer to. Not that it doesn't exist. That's a great idea though to install a glass level gauge and mark the level before the test,then comparing it after. You would have to install a shut off valve before the gauge to prevent over pressurizing the glass tube during the test.
That's a good question to bring up with my tank supplier. Think I'll do that next week. I have never known what the ultimate burst pressure was for a given r'cvr but I've always been told to use the 1.5 test ratio.

 

[kams1973]

I've hydro tested many receivers w/ a grease gun. I read a post on another forum a while back where a guy plumbed in his pressure washer pump and used it to pressurize the vessel.

 

[All]

I am mightily appreciative of the comments, and the direction this thread has taken organically on it's own, concerning testing questionable tanks, either through ultrasound or through thoroughly filling with water.

But let's say the tank is still sound, holds hydraulic pressure, doesn't yield, and is of the rated thickness as determined by ultra sound.

And let's say that the tank was well maintained, but unknown to the previous operator, the daily drain schedule was not doing the job that was thought, due to a small amount of corrosion being sufficient to block the drain valve opening inside the tank enough to still retain a gallon of water, despite the drain releasing enough water to give a false impression that the tank was drained.

Now, with the false function of the internally impaired tank drain discovered by the removal of the tank bushing, there may be some rust inside the tank that would not normally have accumulated had that tank bushing not corroded over invisibly inside the receiver, restricting the 1/4" orifice of the drain valve port.

So, the original question to this post still stands, narrowed purposely to only what the best remedial soaking solution might be, and/or if any such solution would be effective at mitigating the rust growth at all.

Suggestions about better practices in operation (daily draining, etc) are very valid and are still appreciated, but in this case, I am told those practices were done, and found that despite those efforts, the internal corrosion at the valve orifice confounded the attempts at maintaining a water free tank.

So now that the water is under the bridge, so to speak, the question now turns to the best mitigation measures. New drainage system design ideas are being explored in a separate thread, because they present separate, unrelated problems to solve.

But what, if anything, should be done inside the tank?

Previous threads about phosphoric acid, TSP, muriatic acid, pressure washing, etc are inspiring ideas, but it has been hard to find any conclusive results from these attempts.

What say you?

 

[kams1973]

The tank will corrode regardless of what you do. Everything you have mentioned is simply ******* in the wind.

 

[metal4130]

I bought an older Saylor Beall with a 120 gallon horizontal receiver. Receiver is made by Buckeye Boiler Corp and says 1988 on the tag. Looks to be a really well built receiver. It has an auto drain on it and when I drained it there was no water. This was factory surplus and I have no idea on how this tank was maintained or even the last time it was used. I plan to run the compressor at 175 psi and don't want any trouble.

Just wondering if I should do any testing on the thing? What kind of things would I see on the inside of the tank to indicate it has rusted beyond safe use?

 

[All]

A pressure vessel's integrity cannot be determined by sight, however, visual cues can be helpful in determining the direction you might take with tank testing vs tank replacement vs just hook up and use.

The presence of an auto drain, as well as the older age of the tank... can be both good and bad signs. Good, based on the presumption that the auto drain worked, and good, based on the assumption that older tanks may generally have been built with thicker steel which has a longer "half life" to corroding away to minimum thickness.

However, what if the auto drain was there, but wasn't working when the compressor was in use? What if rust has built up over the internal drain orifice, or what if the screen inside the auto drain (which should be serviced annually) is clogged, and the fact that no water came out was not a function of no water being in the tank, but a function of the filter screen being clogged?

You can't really know until you see. A big 120 gallon tank is hard to lift up and shake around to hear water. I found this impossible to do even with a "little" 80 gallon tank.

So while looking inside the inspection ports of the tank is not a way to determine conformance with the boiler plate specifications for a pressure vessel, it IS a very valid and useful way to gather facts about the tank condition to determine your next move.

In my case, so far I've removed tank bushings and fittings, and purchased a boroscope like video camera to have a look inside. I'm currently searching for a local tank inspector who has an Ultrasonic Thickness Tester. I finally found the email address of one last night, but have no idea how much it will cost yet.

New tanks are quite expensive (more expensive than the compressor), and it appears that my pump and motor arrangement might require drilling new holes in the base plate of whatever tank I am able to obtain.

After video inspection of my tank, I found that the shell has a lot more vigin bare gray steel than I expected, which confined my concern only to the bottom cap. The caps in vertical tanks are generally much thicker than the shell, as the caps have to have enough material to brace the shell around the circumference. There are ASME standards for the conical apex angle, the elliptical ratio, the hemispherical radius, the radius of the crown and knuckle... all related to the formation of those end caps. Even a ball peen hammer rap test confirms the thickness difference between the end caps versus the shell.

In the case of your horizontal tank, any standing water remaining in your tank (if from a non functional auto drain valve when the compressor was in service, or if from a rusted over and internally clogged orifice or screen when out of service) would remain sitting in a pool on what could be a thinner shell. It is probably worth having a look, to see if there is enough indications to justify a prudent next step.

 

[All]

There MIGHT BE an acceptable allowance for corrosion in a tank. Or there might not be. It depends on the tank specifications. The specs are determined at the time of manufacture, and are filed with the manufacturer's data report of that tank.

The certification records the original thickness as manufactured, and specifies the minimum "corrosion thickness" allowable that will still meet the pressure rating with the temperature range certified for. Sometimes, there is zero corrosion allowance. I had a hard time believing that at first, but after reviewing several tank registrations, I found that it is true.

 

[CNGsaves]

There IS an acceptable allowance for corrosion in a tank. The specifications are determined at the time of manufacture, and are filed with the certification of that tank.

The certification records the original thickness as manufactured, and specifies the minimum "corrosion thickness" allowable that will still meet the pressure rating with the temperature range certified for.

My current intention is to obtain the original tank certification data.

QUESTION . . . . Does the tank in question actually exist??

. . . OR . . . was this a pedagogical exercise???

Writing a paper or restoring an old compressor??

 

[All]

Unfortunately, when I went to school, we didn't learn about how to post pics and videos because personal computers didn't exist. Wang word processors didn't exist. We didn't even have IBM electric typewriters. I got good at slowing down my rate of typing for certain letter combinations so the mechanical key levers on my Smith Corona and Hermes typewriters wouldn't crash into each other on their simultaneous way to the platen.

So, like an old dog, I crutch on words. I just recently joined this site, so I haven't learned about posting photos here yet, and I really would rather just go back to the garage and leave the computer to gather dust. So I apologize for no pics, but I can assure you that I have a tank, and here are some of the details of that tank:

The tank pressure rating is 200 psi. As best as I have been able to find out so far, the tank manufacturer only hydro tests these particular tanks to 260 psi, so it seems like I would want to avoid hydro testing to 300 psi, as that might stretch and stress the steel closer to it's yield point. This point is being made for the archives, to help anyone else who comes along with a similar problem.

It was in fact the many threads found here on air compressor uses at home that inspired me to join this site, to ask questions and participate in the community dialog. In more commercial environments, pressure vessels are regulated and there are "people" with hard hats and clip board that manage safety monitoring. At home, we are on our own, so I find that sharing data and ideas such as what occurs here has been quite helpful, and I am trying to "reciprocate" by being a giver as much as a "receiver", all air compressor puns intended.

In an earlier post, I mentioned cap thickness versus shell thickness. I have since discovered the minimum shell thickness for my receiver has to be at least .184" in order to still meet the maximum allowable working pressure of 200 psi at up to 650 F. The heads, or caps, of my tank only have to be .149" thick. This turned out to be the opposite of what I seemed to "hear" with the "hammer" test around the tank, where the head "sounds" thicker than the shell. It may very well BE thicker as manufactured, and as it now stands. But the ratings state what they state, and in this case, apparently the head (and bottom end) can be as thin as .149" (a bit more than 1/8", a bit less than 3/16") and still be within the max allowable working pressure as well as survive the minimum design metal temperature rating (-20F).

I don't live in a cold climate, and never will, so the minimum design metal temperature rating will not effect me, but it is something to consider for those that do. I've read of some very interesting behaviors that weakened metal has in -30 temperatures.

I post the nitty gritty details of ratings, because I think they would be helpful to future searchers. I have found that the ratings on my tank are very consistent with many 80 to 120 gallon air receivers. So when it comes to wondering how thick is thick enough, it appears that .149" for the head and .184" for the shell would be a reasonable guideline. A lot of thread I read here assumed 1/4" plate, and I have not found this to be the case from the tank stampings on air receivers in this range.

The best way to get information about an air tank is to locate the U1 manufacturer data report. Unfortunately, this is not as easy as making a phone call. I contacted the tank manufacturer (Manchester), and their records did not go back that far. I contacted the compressor manufacture (Ingersoll Rand), and their records didn't go back that far either. So now my pursuit turns to paperwork filing, the outcome of which is yet to be known.

While looking into my tank, I also grabbed a data point for the other poster with the Saylor Beale compressor whose tank was made by Buckeye. Buckeye was bought by Manchester, and the receiver making operation was consolidated in Quincy, IL. You will not be able to contact Buckeye any longer, and since Manchester's tank records only go back about seven years, Manchester may not be able to provide any original data about the tank (the type of steel it was made from, the original thickness, the corrosion allowance, etc).

I think having some of this information is useful, especially when finding a score on an older compressor which very well might be made better than equivalently sized replacements in an affordable price range. An ultrasonic thickness test is useless without the data about what type of steel it is (ie "SA-414 Grade G", and what the minimum thickness that steel should be on the different sections of the ASME pressure vessel. Might as well take advantage of the engineering that was already invested in building the thing.

 

[dkmc]

I think it's a good idea to clean the inside of the tank.....maybe with solvent or Purple power to get the oil film out and then pour in some POR15, slosh, roll, distribute, coat, and FEEL BETTER about it.

What can it hurt?
POR15 is good at sticking to rust....or ANYhing.....so I don't think conversion treatment would be needed. POR does make some sort of etching treatment
if that would help it stick. But the way it sticks to my dusty old concrete floor...
I don't see how it could NOT stick to a steel surface.

I have a propane tank conversion to air receiver (horiz) under a 2HP quincy and it has pin holes happening after about 25 years. The propane tank was new.

Also have a 25 gallon ASME code receiver (horiz) under a 7.5HP Hydrovane.
This tank spent too many years on the back of a service truck, it is pitted outside and has a couple holes in the bottom big enough to make it unusable at this point.

Already have 2 different size VERTICAL ASME receivers I have picked up as replacements for the others....not a small project to move the motors, pumps, controls, etc all over to the 'new to me' tanks.
I'd like to think with coating.....I'd never have to go thru this again...but I'm 53....maybe I don't need to coat those replacement tanks anyway.....ha ha....

dk

 

[Fixnair]

Great discussion guys. Lots of good information from very knowledgable people.
Thanks!

 

[All]

Well, I'm becoming a bit discouraged about the prospects of "cleaning" an air tank.

The ASME inspection plugs are ON there. I was able to get the bottom one out to remove the drain, but in order to actually "pressure wash" the bottom, I need to get the top one out, and it isn't budging.

First, I used dental picks to extract all sealant and paint film out of accessible threads of plug and receiver. Second, I filled the helical thread valleys opened up by the removal of paint and sealant with alternating soaks of PB Blaster and ZEP 45, for about 72 hours. Third, I've busted out all the big tools... 3/4 drive breaker bars with American made sockets, 4 foot long cheater pipes, 24" Crescent wrenches due to the flat bar stiffness of the handle, a 50 year old adjustable bar wrench whose handle is 1.5" thick, and a dead blow hammer to "break" the seal. The only tool I can think of that I haven't tried is an impact wrench. Which, of course, I would need this air compressor I'm working on to operate, ironically.

So, what I have been able to do is rinse whatever loose rusty scale I could out of the tank from the center discharge port half way up the tank, which affords a 3/4" opening, without disturbing the 2" inspection plug. I saved all the rinse water I put through the tank, every drop, for a total of 33 gallons.

I ran a rare earth magnet through each gallon of effluent, to pick up all the ferritic particles, and in 33 gallons, I collected less than a teaspoon of magnetic material.

It appears that the interior surface would need to be abraded with either high pressure applied to every square inch inside, or the tank tumbling methods described in other threads.

The drain water collected was also oily. It continued to be oily through the entirety of the rinsing phase. The oil residue on the interior of the tank surface may influence the efficacy of any planned application of diluted phosphoric acid. Typically, the instructions of most box store phosphoric acid solutions suggest pre cleaning the surface, while leaving a little bit of rust for the phosphoric acid to "convert." Oily film doesn't seem to be a part of their plan... although I do not know this.

Or it may not. I think of some bolt coatings that are stated to be "phosphate and oil", or "P&O" for short. Perhaps there is some compatibility.

The saga continues...

 

[Danglerb]

You washed out the anti rust protective coating of oil?

 

[All]

Yes... well, I attempted to from the limited access I was able to obtain. While waiting for an ultrasonic tester, I wanted to gather as many clues about the tank's condition as possible. Collecting and settling out the drain water to run through a magnet to determine the quantity of loose scale was one of the efforts at fact finding.

The oil will return when (and if) the compressor is brought back into service.

In this case, the oil did not appear to be an effective rust preventative anyway.

Too much oil presents a spontaneous explosion hazard in the presence of heated compressed air. That's how diesel engines make power. These catastrophically explosive events from too much oil residue inside a compressor tank (from using the wrong oil and/or overfilling the air pump crankcase) are well documented on various OSHA and MSHA safety investigation reports that I've recently researched.

 

[3944r]

A question for Fixnair,

How long do you let the test go for at %150 of the rated pressure of the tank?

In reading other forums and doing my research it seems that some test for hours where as the shortest test i've read about has been 5 minutes.



regards all.

Carl

 

[stratman977]

Just so you know the requirement for hydrostatic test from the ASME code is 1.3 times the working pressure.

The code generally limits the hydrostatic pressure test to be 90% of the yield stress for the "membrane stress" which is a average stress. So the tank may only have 10% additional margin from yielding the tank at 1.3 times the working pressure. The manufacturer many not design the tank all the way up to the limit but you won't know that.

As for the question about test time, the stress in the tank due to a hydrostatic test does not change over time so if it did not fail immediately it will not fail after hours. You do have to allow enough time for leakage to be detectable. If you just have a hairline crack and it only leaks a drop every hour you will not notice it in 5 minutes time.

Another point to make is that a pin hole surrounded by full thickness material will only leak and will not blow up. If the pin hole is small enough in diameter a tin foil thickness left will hold pressure and not leak.

I am an engineer and I don't worry about the wall thickness in my air compressor. I think you are reading the ASME code way too much and need to go outside and enjoy the weather.

 

[3944r]

Thank you stratman977

That clarifies things for me.

I have a pressure tester available to me so I may as well do it.
The tank I have is rated at 200psi, and is dated from 1978, so id
rather just get on with it and see how it holds up.

I'll do the test at 260psi and post the results when Im done


Carl

 

[Engineer61]

Back to the original question - I would investigate the various products made to clean gas tanks in older cars. They are both made of mild steel and have rust problems caused by water condensation, so the same cleaner should work.

 

[All]

I decided to soak the tank with Evapo-Rust.

Chelation of the rust seemed to be less risky than phosphoric acid treatment, which was the other choice that was narrowed down.

I've already tried 1 gallon of Evapo-Rust on 50% of the tank shell (rolling the tank around every few hours for 3 days.)

Ultrasonic thickness testing has been at the very top of the wish list of how to test the tank, but it has proven to be impossible to find anyone qualified with the appropriate equipment interested in testing a homeowner tank, even if brought to them.

Research into the affordable range of ultrasonic thickness testers has revealed that the cheap models are not accurate enough in the narrow and thin range of thicknesses typical of an air compressor tank.

Hydrostatic testing is definitely on the list of things to do, but not before thickness testing. The reason being, is if the tank at the original thickness will yield at X psi, then presumably the tank at some lesser thickness will yield at Y psi, where Y is less than X. How much less psi it might take for the tank to yield under hydrostatic pressure would depend on how less thick the tank might be.

Nothing is ever certain, but at least an approximating calculation could be attempted ahead of time if the actual current thickness were known prior to subjecting the tank to hydraulic pressure and yielding the steel.

So for now, to answer this thread's facet of soaking solutions, I am trying Evapo-Rust. I'm just now pressure washing the tank after emptying the first trial gallon out of it, and am planning on pouring in 5 new gallons (plus the original old gallon poured back in after filtering through a coffee filter) so that the meniscus of the soaking solution will lap up both end bells of the tank simultaneously during rotation.

Since I don't have a rotisserie large enough for this size of tank, I am rotating by rolling the tank on the floor, which is padded with a moving blanket. The welded on base plate to the head of the tank prevents even rotation, hence more fluid is needed to make a larger pond inside. Given that Evapo-Rust is reusable, it did not seem wasteful to deploy 6 gallons of it for this application, as I will be reusing the fluid for many other de-rusting projects over the next few years.

I've had great success de-rusting with vinegar and salt, which is a LOT cheaper, and a lot more fun to watch work (very dramatic crustations form at the surface of the tubs), but the lack of full accessibility inside a compressor tank had me concerned about how well any remaining acid could be fully neutralized, and since acid might be indiscriminate it what it eats, I didn't want to weaken the original steel.

From what I was able to find, after much reading, the chelation process is not supposed to cannibalize or continue to eat the original un-rusted steel. I'm not at all satisfied with my newfound and anemic knowledge about this process, but every other process had some detractors... even phosphoric acid, which "converts" the rust to a black protective film.

It was reported that residual phosphoric acid fumes contaminate air tools and fittings in the air system down line. I'd like to learn more about this... and whether it is indeed true or not, but in the meantime, it didn't seem like the risk was worth challenging.

Despite trying very hard, I couldn't find any downsides to Evapo-rust, other than it not working as quickly as some were expecting. That is not a concern of mine at this point. (Obviously, lo these many weeks later!)

 

[Fixnair]

The inspectors had me hold pressure for 15 minutes. The state of California has different test criteria than any other place or at least did when I was actively engaged in this sort of thing. Use whatever criteria you think is appropriate.

. There is a trick they taught me on removing the inspection plugs. Use an oxy/aceteline torch to heat only the plug to a dull red then hose it down with water. When cool the plug will almost screw out by hand. Do not reuse the plug.

 

[All]

Fixnair thank you for your periodic contributions to this thread. I appreciate your tips and experienced perspectives.

One criteria that I am NOT going to do is test to 1.5x MAWP, as has often been bandied about on various threads I've been reading about the topic. Not even the manufacturer of my compressor tank tests tanks that high, preferring instead to only go as much as 1.3x MAWP.

Therefore, the absolute highest pressure I would hydrostat to would be 260 psi, but given there is some chance of a lowered yield strength from a dozen years of pressure cycles and any corrosion, I was considering whether a test to around 225psi, held steady for a period of 15 minutes, would be adequate for a 200 psi rated tank operated at 175 psi.

I bought a 1 foot standpipe yesterday to screw into the top of the tank. I was thinking of filling from the bottom, while attaching a vacuum fitting at the top of the standpipe, and running a vacuum from a brake bleeder kit (or heck, a shop vac) while the tank is filling. I would continue to fill all the way to overflowing the top of the standpipe, then cap off. Black iron pipe and fittings are being used. Hope the China cast fittings can hold up to the pressure. I picked up a yellow tube of RectorSeal for the threads.

I found a 1/8" pipe thread zerk fitting at the same hardware store, and was able to bounce from department to department eventually obtaining all the bushings needed to to step from a 3/4" NPT outlet down to 1/8". I'm not willing to pump grease into the tank, so it looks like I'll have to buy a grease gun that I can dedicate to being a water blaster. Not sure how that would work though... wouldn't water run out of end caps and spring retainer?

After cleaning the tank again yesterday, I wasn't satisfied with EvapoRust performance. Sometimes my visual assessment seems to depend on the light (sparse) and the state of wetness of the tank. I decided to run a separate Evaporust test coupon for a timed evaluation. In the 12 hours since... the jury is still out.

I was thinking about the economy of time in just getting a brand new tank, but was reminded once more that part of this project is to repair my ignorance with knowledge, and to replace paranoia with procedure.

 

[kams1973]

You may as well experiment w/ electrolysis as well.

 

[CJKaz]

All, where are you located? I might be able to connect you with an inspector that can UT your tank.

 

[All]

THANK YOU CJKAZ!

However, I'm posting here right now jubilant with joy! I finally got my tank UT tested. YIPPEE!! HOORAY!!! In fact I just this moment got home from the inspector's office.

I don't know why something so trivial is so exciting emotionally... like winning the lottery or something. But I suppose when one has tried so hard for so long to get something so simple done with no success, when it FINALLY does happen, it's overjoying!

Here's how I did it. It's summer. It's hot. It was after 12 Noon. I looked in the fridge for something to drink. (I just recently retired I think for good, so I was home).

My darling cute wife of 27 years had brought home some Belgian White "Blue Moon". I figured, what the heck, and cracked one open. I'm retired, it's afternoon, and it's 5 o"clock somewhere. I was just getting ready to touch up paint the moving and handling scuffs on my compressor tank with the factory paint that I had ordered, and after the morning of prep with Sem Solve and Acrylic Clean 330, was taking a solvent break inside.

I'm a lightweight, so it only took 12 ounces of liquid courage to strengthen my resolve to once again telephone the oil refinery inspection firm I found online a couple of months ago. I asked to speak with the specific engineer by name whose curriculum vitae I had found online, which revealed vast and varied experience with UT testing of all manner of pressure, fuel, and oil vessels.

He answered live. Here's the deal, I pleaded. I need to know how thick this tank is, I don't need any certifications or recertifications. I'm just a homeowner, the tank is stripped, I can throw it in the truck and be at your office in 20 minutes, I'm the same guy who left a detailed message on your voicemail a month ago (with no response). He said hang on a minute, let me call you back.

That was enough hope for me to start loading the tank in the back of my truck. As I was closing the tailgate, the phone rang... come on down!

Oh yeah!

I fired up the truck and headed to the land of "Test, don't Guess."

They trundled out the big old $5,000 Panametrics that I had been drooling over on EBay for the past couple months, along with a tube of couplant goo. I had the tank position so that the bottom, where the water ponding is, was facing them with easy access right at the end of the tailgate.

I asked them to make a series of repeated measurements circumferentially around the drain plug bung, in concentric circles matching the pond pattern of standing water in the tank.

Most of the measurements were in the .142" to .153" range. (target is .149" as manufactured). The lowest measurement recorded was .137" in one spot. So I was right to suspect some material loss. Now, the question is what to do about this material loss, as it is not that much.

 

[metal4130]

What brand and size tank do you have. It sounds like the only way you will be satisfied is to just buy a brand new tank!!

 

[CNGsaves]

So now you've tested it . . . . sounds like it's plenty safe to use.

. . . . still the question begs . . WHAT COUNTRY are you in ??