air compressor tank bung

[alabamatoy]

So I have a old (circa 1988) Sanborn compressor with 80gal tank. The tank has 2 - 2" dia bungs, each with a cast plug screwed in. These are leaking significantly, but have rusted themselves in place. A 3/4" drive ratchet with a 3' DOM tube as a cheater will not budge either of them. (Aside, yes, the tank is vented of all pressure!) I have PBBlaster'ed the bungs extensively, no budge. I weigh about 250, Im hanging on the end of the 3' cheater plus the ratchet (24" or so?), so Im probably on the order of ~1000LbFt on these bungs to no avail.

SO.....can I just weld these threaded plugs into the bung to stop the leaking? Yeah, I know, welding cast plug to dissimilar steel....but can I do it without wrecking the integrity of the tank?

 

[FMB4]

Do not, repeat, do not weld on that tank. You might try heating the tank around the bungs with a heat gun (not a hair dryer).

 

[seber]

Have you tried alternate heat and quench? That followed by impact should break anything. If you do decide to use welding, I would suggest brazing instead. HAZ won't be nearly as bad.

 

[rslaback]

If it were me and I didn't have much confidence in my ability to get them out I would probably opt for brazing.

However, I am pretty sure I could get them out with a last resort of using a hole saw to take out the majority of the plug and then snapping the rest of the thread by using a punch toward the middle to break the cast ring.

 

[joe_padavano]

Use an impact wrench, not a breaker bar. The constant jarring impacts will loosen the rust. Of course, without a functioning compressor, using the impact might be an issue.

 

[rlitman]

... Of course, without a functioning compressor, using the impact might be an issue.

Thumbs up for your post as a whole. LOL at this part.

 

[Rinspeed]

I would start with a 5 foot cheater, be careful though as something might give other than the plugs.

 

[Under_Pressure]

As noted, technically you should not weld to the pressure boundary. It certainly voids the ASME certification, though I doubt that is a major concern for you (it wouldn't be for me). Now in this case, welding a small fillet between the plug and the relatively heavy fitting? The risk of causing an issue with the tank integrity is minimal. That said, I'm not sure how successful you would be. Welding around a circumference like that, with one material that may not be easily weldable, over threads, with moisture and sealant and whatever else is between the plug and the fitting, and getting the result to be pressure tight is a challenge. The leak might be smaller when you're done, but there is a good chance there will still be a leak. Those are inspection plugs, and on a tank that age it's probably not a bad idea to do what you need to do to get them out and use them for what they're for- take a look at the inside. Generally the area of concern is the bottom (low point) of the tank.

 

[engineer2]

Know anybody with a battery powered heavy duty impact?

 

[Davefr]

Use an impact wrench, not a breaker bar. The constant jarring impacts will loosen the rust. Of course, without a functioning compressor, using the impact might be an issue.

I bet he could still use an impact even though the tank leaks. I don't see a big hole. It just means the pump would run longer. Sometime if a fastener won't budge in the removal direction you could see if there's any movement in the tightening direction and then work it back and forth.

 

[The Cobbler]

they may have had lock-tite on the threads at assembly too. NPT usually isn't an issue to remove .

is it leaking out of the square socket?

 

[Jswain]

Give it a few whacks with a hammer first then try again. If you can't get a 3/4" impact on it then I would try again with your previous setup with a friend giving it a good whack with the hammer when you're pulling on it.

 

[goblue1998]

I got mine out after heating with a oxy acetylene torch. Then it came right out with a breaker bar.

I tried a long cheater pipe before the torch and got nowhere. After the torch, just needed a normal breaker bar.

 

[Rinspeed]

they may have had lock-tite on the threads at assembly too. NPT usually isn't an issue to remove .

is it leaking out of the square socket?

I've found some pipe dopes set up like cement.

 

[sweetk30]

if that tank is like mine there is a huge o-ring between the fitting and tank .

and cast to reg steel might be bad stuff .

i have drilled a few stuck fittings before and took a die grinder and opened up the holes in a stright line and then split the fitting in 2 and got them out . but in your case you need that fitting or another like it . . . how clean is the inside of the tank for real ? drained a lot so no rust build up ? or like most and ready to blow out ? might not be worth the hassle to fix it . i have gotten lucky a few used OLD tanks / thick wall built and also tossed a few . i got one now the inside looks like the titanic . . . the ship might be better shape even .

 

[Jarhead0408]

This is a joke, do not do this:

Fill it up with air…hook up a 3/4” impact…and see if it comes loose a millisecond before the impact looses air haha.

Hey, it could work!

 

[Jswain]

Whatever you do do not hit it with an impact under while the tank is under pressure. There is a reason the manufacturer felt it was important enough to spend the extra money on that plug specifically labelling that. Things like to keep unscrewing when you have them under pressure and get them going in either direction

 

[Wrench97]

Vent the tank and heat with a torch.

 

[MacMcMacmac]

SWENCH — Power Hawk Technologies, Inc.

www.powerhawk.com

 

[matt_i]

I did this job and the bung was a reducing bushing. I had the same 5 ft bar on it and it wasn't moving. Basically just broke off the hex.

I did not want to use oxyacetylene and burn the reasonably good paint job.

I used a very small hacksaw blade in a supersawzall and sawed two slots from the center->out, stopping before I hit the minor diameter of the internal thread crests.

Then used the cold chisel to carefully crack out pieces so they wouldn't go into the tank. Once the first piece is out then they easily collapse back toward the center as the kerf bends.

 

[Walkers]

I just heated mine up, then they came out with a breaker bar.

 

[metaldad]

I did this job and the bung was a reducing bushing. I had the same 5 ft bar on it and it wasn't moving. Basically just broke off the hex.

I did not want to use oxyacetylene and burn the reasonably good paint job.

I used a very small hacksaw blade in a supersawzall and sawed two slots from the center->out, stopping before I hit the minor diameter of the internal thread crests.

Then used the cold chisel to carefully crack out pieces so they wouldn't go into the tank. Once the first piece is out then they easily collapse back toward the center as the kerf bends.

called 'caping' a fitting out.
this, is the best suggestion.

 

[joe_padavano]

I bet he could still use an impact even though the tank leaks.

Pressure inside the tank high enough to run that impact will add significantly to the friction on the threads, and more importantly, will turn that plug into a projectile as it comes loose. Really really bad idea.

 

[pcmeiners]

With everything the OP stated he obviously has brain cells, suicide or a Darwin award is not on his bucket list.

 

[jerry67]

You say that compressor is from 1988, 35 years old, I would seriously consider treating yourself to new one its not worth taking chances

 

[seber]

ASME tanks do not do that. They are heavy enough to just leak when they get rusted.

 

[rlitman]

You say that compressor is from 1988, 35 years old, I would seriously consider treating yourself to new one its not worth taking chances

Is it just me, or does that bird-**** weld around the drain look like it had some home-made repairs from previous leaks?

 

[Jswain]

Definitely looks like some bird ****

 

[Under_Pressure]

Is it just me, or does that bird-**** weld around the drain look like it had some home-made repairs from previous leaks?

Which indicates that there was severe corrosion in the bottom of the tank. The warning signs were ignored and patched up. While it's not really a code thing, most tanks will indeed give some indication there there is serious corrosion via a pinhole leak of some kind before completely unzipping like this one. The answer then is not to simply do whatever it takes to make that leak stop and keep going- you need to do some internal inspection. If the interior of the tank does in fact look good and it is just a pinhole in a weld or something that opened up after several years, then OK, it might be fixable. But if there is significant corrosion, the answer for pretty much any air-compressor type tank will be to scrap it an buy a new one. It would almost never be worthwhile to invest what would be necessary to do an extensive repair on a compressor tank.

 

[Davefr]

Pressure inside the tank high enough to run that impact will add significantly to the friction on the threads, and more importantly, will turn that plug into a projectile as it comes loose. Really really bad idea.

Good point but the objective would be to only use a couple trigger pulls to get some movement and then revert to manual tools.

 

[rlitman]

Good point but the objective would be to only use a couple trigger pulls to get some movement and then revert to manual tools.

I get it, but once it's broken loose, it's going to move REAL fast.

 

[jerry67]

Here's the follow up to the first video. After watching these I checked mine over and did routine maintenance that was long overdue.

 

[Jswain]

It's a terrible idea, imagine if cast fitting breaks apart and you have 150psi in the tank and are standing directly in front of it.

You ALWAYS bleed pressure off entirely before working on pressurized systems, that rule is written with blood.

 

[Davefr]

It's a terrible idea, imagine if cast fitting breaks apart and you have 150psi in the tank and are standing directly in front of it.

You ALWAYS bleed pressure off entirely before working on pressurized systems, that rule is written with blood.

Good point

 

[Citation]

You say that compressor is from 1988, 35 years old, I would seriously consider treating yourself to new one its not worth taking chances

Looking at that video did you see the welds around the drain plug? That tank failed because someone decided to "fix" it once it started to leak. 35 years old isn't bad so long as it wasn't "fixed".

 

[PoorUB]

Which indicates that there was severe corrosion in the bottom of the tank. The warning signs were ignored and patched up. While it's not really a code thing, most tanks will indeed give some indication there there is serious corrosion via a pinhole leak of some kind before completely unzipping like this one. The answer then is not to simply do whatever it takes to make that leak stop and keep going- you need to do some internal inspection. If the interior of the tank does in fact look good and it is just a pinhole in a weld or something that opened up after several years, then OK, it might be fixable. But if there is significant corrosion, the answer for pretty much any air-compressor type tank will be to scrap it an buy a new one. It would almost never be worthwhile to invest what would be necessary to do an extensive repair on a compressor tank.

When I had my small engine shop I had a customer ask me if I would sell him and engine, then mount it to a air tank and compressor he provided and get it running. Sure, no problem. I got is ll together and ran it and it wouldn't make pressure. I shut the unit off and could hear air hissing. The tank was cracked around the leg near the bottom. The customer asked me to weld up the crack and I refused. He gave me some **** about it not working, I told him it wasn't my fault the equipment he provided was defective. He paid me and took the unit and left. He called me a couple days latter and told me he had a friend weld it and it works fine! I told him, good luck and never heard from him again.

 

[Rinspeed]

That tank failed because someone decided to "fix" it once it started to leak.

Just to live in the real world you have no clue if that is an accurate statement or not. That's just a guess on your part.

 

[Citation]

Just to live in the real world you have no clue if that is an accurate statement or not. That's just a guess on your part.

The guy in the video did say he welded it when it first started leaking. Had he dumped it when it first told him it was worn out it won't have burst.

 

[quickfarms]

This is the best way I have found to remove them

using a cheater bar just deformed the socket

drill holes around the perimeter as close together as possible

drill and tap a hole in the center and inset a long bolt

use a chisel to connect the holes and remove the center

use the chisel to collapse the remaining outer ring

 

[Rinspeed]

The guy in the video did say he welded it when it first started leaking. Had he dumped it when it first told him it was worn out it won't have burst.

It very well might have burst if he didn't weld on it though, correct.