Cracked weld Compressor Repair?

Date 06-07-2009 11:59

Sounds like there is a crack around the nut through the compressor wall. A brazing or soldering operation would work as well as fusion welding IMHO, but the coolant will have to be released in any case. Nothing is going to seal the crack with coolant vaporizing and escaping out the crack while welding, brazing, or soldering. You could attempt an epoxy like J-B Weld with the coolant still in the compressor, but I'm not to sure this would work well, I would think the vibration of the compressor sets up cyclic stresses at the nut, probably why there is a crack in the first place. I'm not too sure how well an epoxy would retard further cracking either, but it may be worth a first try, and it is something your neighbor can do himself.

By awspartb (***)

Date 06-07-2009 12:19

Maybe I wasn't clear but it's the air tank that is leaking.

By jarcher (**)

Date 06-07-2009 12:29

Well, ok, same thing, you have to bleed off any pressure or its unmendable by any method. Shouldn't be a safety issue at no pressure, but hey I would have an understanding with the neighbor that if the repair worsens the condition, he sucks it up, you're not buying new parts for an attempted repair that's free to him.

By awspartb (***)

Date 06-07-2009 12:40

I'm more worried about the neighbor getting killed if it explodes. Compressors kill people when they let go. Even at low PSI. Can anyone point me to a code, if any exists, on welding air compressor tanks? OSHA regs?

By jarcher (**)

Date 06-07-2009 13:16

I don't know what kind of pressure we are talking about here, but the preeminent code for welding of pressure vessels is ASME IX. The construction code is ASME VIII. I know ASME VIII goes up to 3000 psi in Division I, I'm not sure that there is a lower limit. If you have reservations about the fitness for this service of the part after a repair, by all means don't undertake the repair. A part as old as you indicate could be ready to let go simply because of internal corrosion. I don't think brazing or soldering rightly done would cause any metallurgical change in the base metal (embrittlement cracking) but then, like I said, the air tank could be on the verge just because of corrosion.

By boiler tube (*)

Date 06-07-2009 14:09

awspartb, compressor tanks are fabricated to sect. VIII Div. I of the A.S.M.E. code and there should be a code stamping on the vessel. The code repairs would fall under the National Board of Boiler & Pressure Vessel Inspectors Code ( NBIC ) and any company doing the repair must have an "R" stamp authorization. I have had several car dealerships call about cracks in their tanks where the legs are attached and after giving them a quote for code repairs it was cheaper just to replace that tank. Most people think it's just a little crack can't you just weld it and I have to explain the code requirements and procedures I have to follow to do a code repair being "R" stamp holder then they understand whats involved and the costs. Just remember if you do a welded repair on any code vessel and don't have the stamp authorization from the national board the vessel will not be a code compliant. Hope that helps.

By tazmannusa (**)

Date 06-07-2009 13:20

If the tank is in good shape "not a rust bucket" weldin up the crack is no big deal, Ive done plenty
Tom

By ramakant_v (*)

Date 06-14-2009 10:15

PL. check for any internal rusts. if so, you have to be very careful. If the shell thickness is in good shape, probably it can be repaired by providing butt weld configurations. However, it has to be re-qualified by hydrostatic test in accordance to the relevent code specification e.g. 1.3 x design pressure in case of ASME Sev VII Div 1.

One may try to re-rate the shell for a lower working pressure also if required. But please do not pressurize the shell with air without giving hydrostatic test.

Hope this is useful....