Help me get these studs out 7

[ehbadger]

I have some reactors in a corrosive service. Heads are 30" 300# flanges. They were installed with studs. See attached photos

The corrosion between the stud shaft and flange itself locks it up tighter than whatever you can think of that is really damn tight. We can get both nuts to break free but that is actually not helpful as the stud is still "frozen" in place. Also note, the only way to get the stud out is to remove the bottom nut and pull the stud out from the top. We cannot go down because there is no clearance between the flange and vessel itself. Cutting isn't very useful either.

We tried to weld the top nuts to the studs to create what is effectively a bolt. Tack welds at first (didn't work) so we went back around and did the whole circumference. However we are actually shearing the welds with the big torque gun and still not breaking the stud loose. We've tried a ~3/8" interference pin in a hole drilled half and half into stud and nut, which worked for a few, but that is rolling over and not working in most. We tried a hole and pin though the side of the nut into the shaft (shear pin style) but that is shearing as well.
We've got 32 bolts per flange, 3 on each vessel. Drilling 3/8"+ holes for pins/keys is very time consuming and kills a serious amount of drill bits.

Next plan of attack is some hot welding and trying to burn down between nuts and stud and get more area fused. This is also pretty undesireable (but not impossible) due to it being some serious hot work in the middle of a turnaround.

Any creative ideas out there?



Note: For reassembly I am using PTFE coated BOLTS. (not the crappy ptfe coated ones either)

 

[Screwman1]

Get a heavy sledge and beat them down to break the the rust loose. At that point you can get them to turn the the rest is easy. I just hope that you backed those nuts off a couple turns before welding them all.

 

[byrdj]

many decades past, I had a similair problem, but MUCH bigger studs/flange.
We used a "hole saw" that cut between the stud and flange.

 

[ehbadger]

Screwman1:
We did not back them off (welded while process online so we couldn't loosen them). However as we are shut down now we can re weld them with the nuts backed off since the first welds sheared off anyway.
We prefer not to sledge; as the vessel is brick lined. We did attempt this a bit yesterday with a 4ft sledge and got one to break loose, still takes quite a beating. The other thought was back the nuts off about halfway then seat a jackhammer bit inside and just pound away, but again, not ideal with the brick lining.

 

[ehbadger]

byrdj:
That sounds awful, but effective. Probably a good last resort.

Right now we are having some success with a 1/2" shear pin drilled through the side of the nut. It is just slow going.

 

[Nescius]

Well, I would not have welded all the studs, yet.

If they're really, truly, anomalously stuck...and you have a bunch of them to remove...building a tool could be prudent.

I might try a hydraulic pulling cylinder with a threaded rod end mated to the stud with a coupling nut. Use a section of heavy wall tubing for a standoff, to bear on the flange. Cut a window in the tubing so you can hit the coupling/stud with a big air chipping hammer. Pull HARD, then continuously apply the hammer.

Take significant precaution to guard against the apparatus launching upward when the stud breaks free, or simply breaks.

This is only my opinion, NOT based on experience!

 

[Tmoose]

It is hard to see if the stud body is smooth, or fully threaded.

Regardless, I'd heat the stud to a cheery (maybe even cherry) red glow for a while, so the stud body surrounded by flanges got glowing hot too. Then I'd let it cool off a bit, expecting the corrosion in the gap would have been crushed and broken by the hot swollen stud. In addition, the red hot stud body might "upset" a little and upon cooling be smaller than before.

The theory of heating the nut, hoping to make it expand and be looser and loosenable has not tested well.
For me, loosening more traditionally sized rust seized fasteners by aggressively heating the bolt/stud has been 10X more effective than heating the nut ever was.

 

[MJCronin]

Given the corrosive conditions and the experience with the frozen studs:


1) What should have been a much better flange/fastener design ? (I am thinking longer studs and double nuts on top)

2) What fastener materials/coatings should have been used ?

3) Am I correct to understand that there was widespread flange leakage of a corrosive liquid that caused the problem ?

4) Is there a reason that the traditional removal solution mentioned above( lots of heat) can't be used here ?

5) PTFE coatings have temperature limits. What is the operating temperature of the unit ?

MJCronin
Sr. Process Engineer

 

[3DDave]

Who knows what I would have done, but from my comfy sofa, I'd look at getting a rig that does bolt tensioning on structural bolts and using that on the ends of the studs. The math tells me that a good pull will stretch the stud a bit and that the diameter has to decrease. That should lower the compression between the stud and the holes that the corrosion products have built up to allow one's choice of penetrating lubricant to slip in while also shearing the corrosion.

https://www.youtube.com/watch?v=ef9CXDhzH-U

In your case, remove the nuts so there is access to apply penetrating oil to lube up the corrosion products. Work around the flange from stud to stud until they are free. It may take a few passes. The bolt tensioners are daisy-chainable so the entire flange has a uniform pre-load. It may be worthwhile to do the same to remove these studs rather than just one at a time.

A feature to note is the studs -were- stretched by the nuts and are now both shorter and larger in diameter than when the corrosion formed, so the friction in there goes up when the nuts are loosened.

 

[byrdj]

expanding on 3DDaves comments. I have used hollow cylinder rams, with a pipe sheeve around the stud and a piece of thread rod attached to the end of the stud. In my case the stud's were centered drilled and end tapped. thread rod could be welded to the end. a blast shield is needed in case (when) the rod snaps prior to pulling the stud

 

[Artisi]

Have you considered drilling into each stud with the thread tapping size drill after the necessary smaller pilot hole/s, followed with appropriate tap. A major undertaking but it would seem you have a big job ahead of what ever path you take.


It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Sparweb]

Hard to position the necessary drilling tool under the pipelines that connect to this flange at 90 degree elbows. Unless that part of the pipeline can also be removed. They say that drilling 3/8" holes has been difficult, so I expect drilling 1" holes will be proportionately more difficult. The bit will want to wander off the axis of the studs, cutting into the lower flange's holes. They probably don't want to risk damaging the boiler or its flange.

STF

 

[Artisi]

Assuming you can access the studs, cut the studs off flush at the top of the flange, carefully centre punch the stud, drill a small hole initially and increase the drill size in a couple of steps. This is not a job for your average boiler-maker / welder, you need a fitter or machinist with experience. Correctly done there is little chance of the drill running off axis.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[ehbadger]

Thanks for all the help; sorry on the delayed responses but I will try to answer everyone in order:

If they're really, truly, anomalously stuck...and you have a bunch of them to remove...building a tool could be prudent.

I might try a hydraulic pulling cylinder with a threaded rod end mated to the stud with a coupling nut. Use a section of heavy wall tubing for a standoff, to bear on the flange. Cut a window in the tubing so you can hit the coupling/stud with a big air chipping hammer. Pull HARD, then continuously apply the hammer.

Building a tool is definitely on my list. We already have a custom "wrench" of sorts to fit down through the gap in the grating and curls underneath the vessel flange to hold that bottom nut.

That coupling nut sounds interesting; to be clear, is this an extra nut threaded onto the top of the stud? Interesting idea, I will note it down.

It is hard to see if the stud body is smooth, or fully threaded.

Regardless, I'd heat the stud to a cheery (maybe even cherry) red glow for a while, so the stud body surrounded by flanges got glowing hot too. Then I'd let it cool off a bit, expecting the corrosion in the gap would have been crushed and broken by the hot swollen stud. In addition, the red hot stud body might "upset" a little and upon cooling be smaller than before.

Stud is currently fully threaded.
I'm not sure heating the stud is an option. Due to the brick lining, the flange insert and brick acid membrane is susceptible to damage at temps >250F. I think heating the studs that hot, even one at a time, would ruin the membrane at the bolt holes.

1) What should have been a much better flange/fastener design ? (I am thinking longer studs and double nuts on top)

2) What fastener materials/coatings should have been used ?

3) Am I correct to understand that there was widespread flange leakage of a corrosive liquid that caused the problem ?

4) Is there a reason that the traditional removal solution mentioned above( lots of heat) can't be used here ?

5) PTFE coatings have temperature limits. What is the operating temperature of the unit ?

1. Not sure. Vessels were built in the 60s I think, so short of unbricking 4 layers, cutting out the existing nozzles and extending them, I think we are stuck. Plus all the piping on top would need reworked (not cheap; most is tantalum).

2. Cannot use SS fasteners due to SCC risks. So studs are B7 carbon steel. We are going back this time with PTFE coated bolting. Probably still not the best all things considered but it's all we can do in our short turnaround (this time)

3. Not really leakage at the flange. Reactor have hot HCl (think >140C) and we don't really detect anything, but permeation or some other mechanism we definitely get corrosion.

4. See previous response to Tmosse; in short, vessel is brick lined and cherry red studs would ash the pyroflex membrane requiring a nozzle rebrick.

5. Operates just under 150C (about 300F). PTFE is OK here.

 

[ehbadger]

Who knows what I would have done, but from my comfy sofa, I'd look at getting a rig that does bolt tensioning on structural bolts and using that on the ends of the studs. The math tells me that a good pull will stretch the stud a bit and that the diameter has to decrease. That should lower the compression between the stud and the holes that the corrosion products have built up to allow one's choice of penetrating lubricant to slip in while also shearing the corrosion.

https://www.youtube.com/watch?v=ef9CXDhzH-U

In your case, remove the nuts so there is access to apply penetrating oil to lube up the corrosion products. Work around the flange from stud to stud until they are free. It may take a few passes. The bolt tensioners are daisy-chainable so the entire flange has a uniform pre-load. It may be worthwhile to do the same to remove these studs rather than just one at a time.

A feature to note is the studs -were- stretched by the nuts and are now both shorter and larger in diameter than when the corrosion formed, so the friction in there goes up when the nuts are loosened.

Good thought. We are a bit crunched for time right now - I will have to look into the time component of stretching each bolt. I think the time can be scheduled around (in the future). It is just hard now to know whether we can get these covers off in 1 day or 6. If it is reliably say 4 days, yeah it is a long time but at least consistent and can be planned for.

To your last point, that is definitely true. Kind of a catch-22; to remove the stud you need to remove a nut, but by removing a nut the stud gets stuck more.

expanding on 3DDaves comments. I have used hollow cylinder rams, with a pipe sheeve around the stud and a piece of thread rod attached to the end of the stud. In my case the stud's were centered drilled and end tapped. thread rod could be welded to the end. a blast shield is needed in case (when) the rod snaps prior to pulling the stud

Make sure I get this right; you drill and tap directly into the top face of the stud and thread in another rod, one end of which is connect to the ram? Sounds like it would work. Still requires a lot of drilling though. Hmm

Have you considered drilling into each stud with the thread tapping size drill after the necessary smaller pilot hole/s, followed with appropriate tap. A major undertaking but it would seem you have a big job ahead of what ever path you take.

I am not sure what you are going for here.

Hard to position the necessary drilling tool under the pipelines that connect to this flange at 90 degree elbows. Unless that part of the pipeline can also be removed. They say that drilling 3/8" holes has been difficult, so I expect drilling 1" holes will be proportionately more difficult. The bit will want to wander off the axis of the studs, cutting into the lower flange's holes. They probably don't want to risk damaging the boiler or its flange.

Definitely don't want to damage the lower vessel flange.

 

[ehbadger]

Thanks all for the responses. Definitely some things/ideas to consider. In summary, and not to rain on any ideas but just to clarify some constraints:

1. Right now anything with drilling is not ideal as it takes a long time.

2. Heat is not ideal. The vessel flanges have an acid membrane that gets ruined around 250F.

3. Excessive pounding is also not ideal as the brick inside can fracture off.

4. Access to the lower side of the flange is minimal due to vessel construction.

In the future, I imagine I may have to concede in some fashion on one or more of these points. Whatever solution developed may be able take some time as long as it is consistent, as that can be planned for. Nowadays we never know if this job will take 1 day or 5+. I like the idea of a hydraulic ram type solution if we can figure a way to make it work without requiring a lot of drilling or helped along by pounding.

I'm also considering bolting MOC change to something less susceptible to corrosion if the PTFE coating doesn't work. However, any alloys would definitely be costly. I've got 18 of these flanges on site

Again, thanks all for the input so far. Much appreciated!

 

[3DDave]

You may consider the liberal use of polysulphide rubber to exclude the ability for corrosive materials to enter the threads in the holes in the first place. It will protect the surface of the fastener and the hole. If you use hydraulic tensioning on the studs on the flanges then you can apply it to the portion that is in the nut threads without affecting the torque levels, which you will no longer use/depend on.

 

[ScottyUK]

If beating it with a 16 pounder eventually loosens it, is there any possibility that sustained use of a large electric breaker would vibrate it loose? Something like

https://www.hilti.co.uk/drilling-and-demolition/demolition-hammers-and-breakers/r4783

with an adaptor perhaps. Not sure if that would also damage your brick lining.

 

[moon161]

Have you thought about a cutting torch? In my experience, the burn usually is usually interrupted as you go from one piece to the next. You could burn through each nut from top to bottom and that would relieve the tension. Top to bottom on both sides and it should just split off. The boundary provided by the washer should protect the flange. If you start your cut at the top and burn down, you will minimize heat transfer to the flang. Ice or wet rags will keep the area from heating as well.

 

[itsmoked]

Many times I've had "Liquid Wrench" work when all hope was lost. Have you tried anything like that?

Keith Cress
kcress -

http://www.flaminsystems.com