Muriatic Acid

[salem747]

Hello,

Has anyone used Muriatic acid or Hydrochloric acid to clean out a rusty tank? I understand this is the best way to do it, essentially etching the steel.

How much is required? The stuff I got doesn't have a percentage but it says 20 then a degree symbol. I am cleaning out an old 60 gallon compressor tank.

I plan on dumping some in, rolling it around, dump it out and repeat. I only bought a gallon of the stuff to start. I was planning to rinse out the tank with water after, then slosh some WD-40 around inside to drive out the moisture and protect the steel.

Any thoughts?

[transit]

I don’t think its too good of an idea. You obviously plain on using the tank. Rust has reduced the tanks wall thickness and you will be making the tank thinner. The acid cant choose to work on the rust and not attack the steel. I think it’s too risky, the tank may or may become too thin to be safe.
http://en.wikipedia.org/wiki/Hydrochloric_acid

[rlitman]

The degree symbol refers to "Baume", so it is 20 Baume Muriatic acid, which is also known as 32%.

I will say, that simply dumping and a quick rinse, will not neutralize the acid. After dumping, it could take MANY gallons of continuous rinsing, or better yet, neutralize with baking soda (then more rinsing).

Not commenting on your tank cleaning idea though. I'll leave that to someone else.

[Rojodiablo]

Okay, what will you be holding in the tank?? I'll say that rlitman is correct about cleaning procedure, you will need to neutralize the tank internals with a baking soda mix after cleaning. Any acids left that are on the walls in pores and such will be persistent. On clean, smooth metal, a good rinse will work.
WD40 won't neutralize the acid very well, and it won't necessarily protect from acid continuing to do it's job.

[specter]

How bad is the inside of that tank???

When you turn the tank upside down and tap on it does alot heavy thick bits & pieces of rust come out of the tank? If so a simple fluid cleaning and rinsing will not work.

You may have to try an interior sandblasting then doing a flush followed by your acid cleaning. Some sandblasting shops can do the tank blasting. There is also water blasting.

[salem747]

It is a compressor tank.

I haven't taken all of the inspection plugs out yet to have a better look inside but the inside of the pipe going to the pressure switch is half full of fluffy rust.

Neutralizing the acid is a good idea, I hadn't thought of that.

I wasn't planning to leave the acid to work for long. You'll note in the wikipedia entry that mils etc use HCl to clean rust, "pickling steel"

Wall thickness is an issue though. I should do what I can to confirm that the walls are still thick enough to do what they need to. I wonder how much a quick NDT would be or maybe a normal pressure vessel test.

Gave a quick call to a pressure vessel guy and he suggested doing a hydrostatic test to 150% of operating pressure. He said this is the best because it isn't dangerous and you can probably just do it with a hand pump. Probably not a bad idea to do on an old tank like this even without using any acid!

[MoonRise]

Depending on the laws/rules/regulations in your locale, hydrostatically testing a pressure vessel might not only be a good idea, it might also be a requirement. Depending on whether this is for commercial use or a home garage, depending on tank size vs rules, etc, etc, etc.

[salem747]

Yeah, it doesn't look like anyone around here could do it for me "officially". I will have another look and see if there is a boiler guy or something. At the mine I used to work at in BC we couldn't use Parker Hannifan accumulators on our equipment because they weren't tested the same way a boiler is! Talk about letter of the law, do you have any idea how many hydraulic accumulators there are in this world? We had to use bladder type accumulators because they had been tested.

[pulser]

As others said, it may not be safe to salvge this tank if the rust has thinned it too much.

Regarding the acid techinque, I restored a badly rusted motorcycle tank by using phosporic acid and water, with several cups of pea sized gravel, and shook it like crazy many times, let it sit, and repeat. Rinse with lots of water, dry with a shop vac blower, coat with Sea Foam.

Phosphoric is the acid to use, it is the active ingredient in rust removal products like Naval Jelly, and it produces a protective layer of iron phosphate after removing the rust. I suppose you could use HCl first and finish off with phosphoric.

[William McCormick Jr]

In a Hazardous Waste yard where I worked for many years. The owner was a friend of mine, and suggested that I learn the plant from the ground up.

My first lesson, I was taught to triple rinse. A black guy named Elija Boe, yelled at me one day. I was filing up containers contaminated with ammonium hydroxide solution, with fresh water, almost two thirds full, with water, to rinse them out.

He snatched the drum away from me, and said "Man what the hell you are doing"? He said "you are going to fill up the rinse tank in a few minutes like that". And you won't get them cleaned.

He said its math. You take the drum and put a little bit of water in it, shake it all around, it cuts the mix, to 1/100th the strength, then you dump that and do it two more times. So that the next time you rinse you get 1/10,000 strength solution left. And finally the third time you have it down to parts per million.

Wow does it save time and energy and resources.

I am not big on putting acids into a tank. They can lay under the gasoline in a gas tank and continue to work, like rust itself, rust generates hydrochloric acid. That is why it acts like a disease even to stainless steel. Most good compressor guys say get a new tank.

It is like if your car is in a really bad accident and you twist the frame or uni-body. Most really good auto body guys will condemn the car for a friend. Even if they could straighten it out. It is just never the same.

Rust can ARC and possibly cause you grief. So if you do use it make sure you only use synthetic compressor oil. It will not let the carbon build up in the check valve, like regular oil does.



Sincerely,


William McCormick

[steve45]

Agreed, a 60 gallon tank wouldn't be all that expensive. You can get one from Granger's or other places.

If you pressure test it, make sure you get ALL of the air out of it and test with water. After it's full, you can use a PortaPower pump, grease gun, etc. to build up the pressure.

[denrep]

It is a compressor tank.

I haven't taken all of the inspection plugs out yet to have a better look inside. . .

That's where you should start, a close visual inspection.
Then you can move on to ultrasonic testing, with a 3# cross-peen.
Any problem will likely be at the low point.

What's the tank's build date?
Does it really look all that rusty inside?

If you expect that the tank has such a small amount of rust that it will
be sound after de-rusting, what's wrong with leaving it alone?

I suppose you could consider treating it with a rust neutralizer.

Good Luck

[Oldiron2]

As stated above, hydrochloric acid will also eat the metal, not just the rust. A better solution is either the Phosphoric acid, or Oxalic acid which is also sold (at painting stores) as a wood bleach. The H3PO4 will leave a phosphate surface which later protects the surface, at least for a while. You could also try the electrolysis procedure, if you have an appropriate power supply. A thread here gives details; do a search for it.

You need to first clean the inside well with something like Gunk to remove oils, otherwise the aqueous solutions don't work well; they won't penetrate an oil-rust crust.

Have fun getting the inspection plugs off; sometimes they forget they really aren't part of the welded tank.

[awright]

A pressure test is an absolute necessity, especially for a tank built in 1958 and known to be rusty. I have used a local fire extinguisher service company to pressure test several salvaged air receivers I wanted to put to use.

I'm not confident that 150% of working pressureis an adequate proof test. I would recommend finding out what the required proof pressure is from the ASME Pressure Vessel code or the Canadian equivalent. This, and the appropriate cleaning procedure, are not things to guess at. A 60 gallon tank exploding at over 100 PSI will definitely blow the walls out of your shop and probably kill anybody nearby.

awright

[salem747]

I haven't gotten the inspection plugs out yet, they are in there good but I did drain it and the water came out pretty clear so maybe it is less of a problem than I thought. I will check on the proof test value. I think I will just do it with an enerpac like the pressure vessel guy suggested but with the enerpac I can just borrow it from work and don't have to buy anything. I will fill the tank to the top with water, then pump it up to the proof test pressure with hydraulic oil. It will probably only take about 5ml of oil anyway and then I don't have to put water in the pump and have to clean it all out.

[fredf]

how about a cheap grease gun filled with water???
fred

[Oldiron2]

I will fill the tank to the top with water, then pump it up to the proof test pressure with hydraulic oil. It will probably only take about 5ml of oil anyway and then I don't have to put water in the pump and have to clean it all out.

Even high-pressure cylinders stretch when filled; I would expect a 60 gallon tank like this to stretch significantly more than 5ml. I would not use the oil.
Also, I believe that tanks being tested are immersed completely in large tanks of water to contain things if the one being tested does rupture. The water does compress some, the metal stretches, and that energy does work if released in a failure.

[William McCormick Jr]

Water and oil under pressure is not a good idea either. You remember the oxygen tank regulator and oil contamination being a bad thing? Well, water is 8/1 oxygen by weight over hydrogen.

They take out all the water from high pressure air before they compress it. For many reasons.

Steam pipes blow, sometimes because of the sudden compression and condensation of high temperature steam gas. As the steam condenses very quickly it can liberate many thousands or millions of BTU's in an instant. Causing any trapped water to instantly turn to steam. You can get instant detonation.

You can also cause water to detonate upon sudden compression.

Water can also combine with carbon dioxide under pressure and cause the carbon dioxide to turn into carbon monoxide, while burning the hydrogen in the water. This is a common accident with water.

Let an expert look at the tank. If he says its no good. I would listen to him. I will ask my friend if they de grease the compressor tank before testing.


Sincerely,


William McCormick

[steve45]

We're not combining oil and oxygen here, which will definitely burn. We're combining oil and water. Two different things.

We're not doing this at an elevated temperature, either. Nor are we combining it with carbon dioxide, which can form carbonic acid.

And, yes, expect the tank volume to grow a LOT more than 5 ml. Also, when you fill the tank with water, let it stand for a couple of hours and let the air entrained with it form bubbles. Then, bang on the side and swirl a stick around in it to get the bubbles out.

[awright]

I think (but do not actually know) that part of the pressure proof test is measurement of the amount of expansion of the tank under test. More than some specified amount of expansion indicates weakened walls and constitutes a failure of the proof test, even in the absence of bursting or leaking. The amount of expansion is measured by the volume of water that overflows from the tank in which the tank under test is submerged. I have no idea what the permissible expansion is or if my impression is even correct. If correct, I imagine it is some fixed percentage of the volume of the tank under test.

awright

[MoonRise]

awright,

Yes, a hydro test does include a measure of the expansion of the pressure vessel under test. The fail criteria vary depending on the test specs.

[salem747]

I'm not afraid to pay to have it tested, but I have checked with the fire extinguisher place here and they can only do small tanks.

I have read that the measure of expansion they use is basically to make sure that most of what you put in comes back out, therefore an elastic action, versus a permanent deformation => yield of the steel. Anyway I haven't given up trying to find someone to test it but the guy I got the original advice from builds pressure vessels so he would know what it takes.

[hockeyman]

I think that William McCormic Jr. needs to reread what he stated as facts in Post 18 and maybe , deleting that BS.

[William McCormick Jr]

If the water and oil become heated, by any means. Mechanical pump, an electrical ARC to the outside of the tank during testing. A cutting torch against the tank. A chemical reaction in the tank. Oil and water can detonate. An electrical charge to the water in the tank, could also cause detonation.

When bearings in pumps go, and the pump overheats, while pumping contaminated, oil and water, it can burn the mix, with oxygen and acetylene like force.

But I understand the years of testing tanks with water pressure, and the safety, of the tank fragments not leaving the body of water if it blows. Much like hydraulics.

Air, hydraulic fluid, and water, can detonate. When air is suddenly compressed, by a mechanical movement of a cylinder. Causing amazing pressure and heat through condensation in a cylinder, you can detonate the water and oil in that cylinder.

I would let someone that does that, do it.

Sincerely,


William McCormick

[salem747]

I was thinking to ask my LWS where they have bottles tested. It's likely that there is no one around here that does it though. I may have to go down to Toronto or Barrie.