Welding on compressor tank

Date 06-09-2011 22:00

From what you say, looks like the tank is an old one and maintenance hasn't been that good. Compressed air tanks use to hold some water within them that should be drained daily. Has it in your tank? You don't know, so you don't know either in which condition the inside surface of the tank is.
Does the outside surface look like it has been just painted? If the previous owner kept hidden the conditions of the mounting legs, chances are that he painted a rusted otside surface for you not to see in which condition it was.
I would say that before attempting to perform any weld on the tank you make an inspection on it. Remove a small portion of paint from the outside surface with a wire brush and check its condition. Measure the tank actual thickness with ultrasound and calculate if it's still able to withstand a pressure of 130 psig. Use the ASME VIII Section I formula to do the calculations. Assume a corrosion allowance (factor "c" in the formula) of 1/16 of an inch.
Compressed air tanks are designed and built to ASME VIII Section I. If the calculated thickness is higher than the actual one including corrosion allowance, don't hesitate: scrap the tank. If, on the other hand, the tank is in good (or at least acceptable) condition, tell us and we'll tell you how to weld on it.
Giovanni S. Crisi
Sao Paulo - Brazil

By Metarinka (****)

Date 06-10-2011 04:30

maybe I'm jaded but 130 PSI is not that high. The tank is probably cheaper low carbon mild steel.

In terms of code, there's nothing specific to say without knowing the alloy. Pre/heat post weld heat treament etc are designed to ensure certain metal properties or prevent certain defects. General requirements such as cleanliness, grinding back a certain distance and weld size or defect size are different issues.

For a cheap 30 gal compressor, you could try over testing it and taking it to 1.5x it's design pressure of 130 so about 195 psi in a safe area. Chances are there's such design margin that welding on legs wouldn't damage it that much.

I'm jaded as in the aerospace industry we would routinely make pressure vessels in the 3,000-20,000 PSI range with wall thicknesses under .150"

By phaux

Date 06-10-2011 05:54

I was running it the first several times at 180psi due to a faulty gauge. It has since been regulated to 130psi.

By Northweldor (***)

Date 06-10-2011 12:09

I would still do a hydrostatic test to at least 200 psi before and after revision.

By rlitman (***)

Date 06-10-2011 15:07 Edited 06-10-2011 15:10

> 130 PSI is not that high

I'm not normally one to link to another forum, but this is something you should have a look at:
http://weldingweb.com/showthread.php?t=48232

Are the pressure vessels you make made to store air, or hydraulic fluid? There's a big difference in their relative safety (although loosing hydraulics in the air isn't a good thing either . . . )

I would look up a fire extinguisher shop in the yellow pages. They often have the certifications for hydro testing this type of cylinder.

> the previous owner had cut off the mounting legs and welded on new ones at the base

Now I have to ask, did he weld the new legs right to the tank? That would set off red flags for me. Did he grind on the tank, or can you see undercut on the weld. If so, don't bother having it tested, just get a new tank.
If he cut the feet off the legs, and welded to the legs, I could see that.

By phaux

Date 06-12-2011 13:54

No, he cut off the old legs and welded new ones in a completely different spot, looks pretty ugly. I found a local compressor shop that can test it.

By G.S.Crisi (****)

Date 06-10-2011 19:28 Edited 06-10-2011 19:47

I would say that the tank material is A 285 Gr B, i.e., low carbon steel for moderate pressure service. In this case, pre and post weld HT wouldn't be necessary because the tank thickness is certainly less than 3/4 of an inch.
I agree in doing a hydrostatic test at 195 psig, i.e., 1,5 times the pressure of 130 psig, taking the necessary safety precautions, as Metarinka suggested.
Giovanni S. Crisi

P.S.
Metarinka,
a 20.000 psi pressure vessel with a wall thickness of 0,150 inches? I confess my ignorance: what material is the vessel made of?

By Kix

Date 06-16-2011 16:23

You're right 130psi is not that high, but it is very dangerous in a 30gal tank. It's not the pressure that kills, it's the volume escaping all at on time that does.

By qcrobert (***)

Date 06-10-2011 16:50 Edited 06-10-2011 21:12

We are an ASME certified ("U", "PP" & "R" stamp) fabrication facility. But your compressed air tank does not fall in the category of Section VIII Pressure Vessel code nor B31.1 Power Piping code.

As someone has mentioned, a dive shop that certifies compressed air tanks would be more appropriate.

I personally would hydrostatic test the tank to 1.3 times MAWP (as in pressure vessel testing) or 1.5 times MAWP (as in power piping). In order to accomplish this, you should take care to vent all air from the tank while filling with water and make means to relieve pressure when draining. I have seen numerous cases of imploded tanks.

Hope this is beneficial,
QCRobert

By G.S.Crisi (****)

Date 06-10-2011 19:45

Robert,
in my opinion, phaux can not run a hydrostatic test at 1,3 (or 1.5) the maximum allowable working pressure because for this he should know the actual vessel wall thickness, which, as I said on my first posting, he doesn't know. The original thickness may have suffered corrosion beyond the original corrosion allowance.
He should know also the tank material, although, as I said on my second posting, he can assume that it's A 285 GrC, i.e., low carbon steel for moderate pressure service.
So, I'm with metarinka when he suggests to run the hydro test at 1,5 times the maximum pressure the vessel has withstood (130 psi).

Now I must confess my ignorance: why doesn't that tank fall under ASME VIII? Would you clear it up to me? Thankyou.
Giovanni S. Crisi

By qcrobert (***)

Date 06-10-2011 21:11

I agree with your reasoning that the thickness may have surpassed the corrosion allowance of the tank and not knowing the construction material is of concern. Still I personally would proceed with a hydrostatic test for my own assurance as you have stated 1.5 X the 130 psig with a calibrated gauge.

Our facility is fortunate to have UT to check the remaining thickness of "used" pressure vessels.

I may have spoken too quickly in that I have never seen, heard of, or dealt with a compressed air tank constructed to VIII Div 1. However, any type of tank or vessel may be built to the Section VIII Div 1 code, even a steel coffee cup.

ASME Section VIII Div 1 is Rules for Construction of Pressure Vessels and the Introduction states that hydraulic and pneumatic cylinders are not included in the scope of this Division. Sec VIII Div 1 primarily deals with steam retaining vessels, heat exchangers, condensate flash tanks, fired boilers, and some hydropneumatic tanks, etc..

Air receivers, vacuum tanks, surge arrester tanks, air surge tanks, expansion tanks, air separators are also not included in the VIII Div 1 scope. When these types of tanks are constructed, many Engineers rely on Sec II for (applicable Part A or B) material specifications & properties along with (Part C) Specifications for Welding Rods, Electrodes, and Filler Metals.

Compressed air tanks can be fabricated to API, NFPA-22, AWWA-D100, UL-58, UL-142, UL-1746 and UL-2085, including ASME.

Robert P.

By Tex

Date 06-16-2011 22:32

I have to respectfully disagree with some of the earlier post regarding UT of the tank and the inherent dangers. Having welded on and tested tanks upon completion Im not seeing the concerns with a 130 psi air tank. We had to weld end caps on the rolled steel to build the tanks. For testing we used a gauge and filled the tanks with air and then looked for leaks with a spray bottle of soapy water. On the rare occasions that there was a major failure, the weld would split out. As the pressure was released (and scaring the heck out of everyone) the pressure on the remaining weld was reduced, therefore holding the end caps in position and not letting them fly off through space. Your situation sounds as if there were a failure all escaping air pressure would be directed down toward the bed allowing the other three bolted legs to hold the rapidly diminishing 43 psi per leg. One thing you may consider if you are worried is bending some 3/16 strap to the radius of your tank and after welding it on then weld your legs to that. I think this should do the trick safely. If I misunderstood your original question I apologize. Also I meant no disrespect to the previous posters. It is always better to err on the side of caution. Have a good day.

By aevald

Date 06-17-2011 00:42

Hello Tex, fortunately I have never been the recipient of any explosive compressor vessels. However, on one of the other welding forums there was a gentleman who included a number of pictures and narrative about his experience with a "simple" air compressor tank failure. His son had received this compressor from a friend and "dad" decided that he should probably check this unit out before letting his son loose with it(he said that the son wasn't really very mechanically inclined). Short story: the pressure cut-out switch wasn't set properly and the relief valve wasn't functioning correctly. The center seam on the bottom of the tank cut loose and as the sides opened up they hit his hand, leg, and knee. Broke the hand in many places, broke both bones of the lower leg, and broke his knee. The force also rearranged his garage, blew the door off of his refridgerator and generally raised havoc and all of this from approx. a 20 to 30 gal. tank. You spoke of leak checking tanks that you fabricated, I'm guessing that you were probably using around 3 to 7 psi. of air and some soapy water to find leaks as that is relatively common. Yet, if the tanks are very large at all, even 3 to 7 psi. can be very scary if something actually cuts loose. Much of the time a doubler is the way to go if you are going to be welding braces, legs, or other attachments to tanks, that way if there is a crack or something from these appendages they will generally only affect the doubler and not the tank itself. Regardless of the repair or modification, proper testing of the finished product is really the proper way to deal with a situation such as this. For what it's worth that's my take on this situation. Best regards, Allan