Extracting bundle of floating head heat exchanger 3

[Stefan2211]

Dear All,
we are facing the problem that we can't remove the bundle of our horizontal shell&tube heat exchanger. Weight of bundle is 9t. Design floating head (drawing attached).

Problem: hx has severe fouling and is leaking. Leaks are expected to be caused by uneven thermal expansion due to debris collected between the baffle plates and reducing heat transfer (pic), mechanical stress as we have non corrosive fluids (hydrocarbons).

Assumed reason that bundle stuck in the shell:
Solids (coke) has has settled down at the botom of the shell between the baffle plates (high and tolerance of baffle plates to shell is to small. Bundle can't be moved at all.
We tried to pull with 16 t forklift, used hydraulic jack at floating end to push it and to break it loose - no success. There might be also a deformation of the shell due to uneven heat distribution since we expect a lot of residue (coke - pic is from different unit) has settled at the bottom so free tubes have a different thermal expansion than the covered tubes.Chemicals doesn't work as coke is quite inert to most common used chemicals

We will try once again with larger hydraulic jacks to break the bundle loose and then pull (see right side of marked drawing). If that fails we will need to cut the shell. Red marked cutting of a window is not suitable acc to the vendor as the shell will collapse once the section is welded back (heat impact and shrinking - many attempts failed with other clients). They agreed to follow the green marked strategy to cut the shell in 3 pieces. But even then rewelding will be a challenge. Please note that only the pic with tubesheet (leaks) is the particular unit. The other pics are from an other unit but we were able to extract the bundle. Any suggestion? Thanks in advance

 

[TiCl4]

Stefan,

Yes, we use 20k and 40k regularly here for cleaning polymer reactors and Hx tubes. Pressures below 13-15k barely do anything to clean hard polymers off the metal surfaces. The cleaning is done by a professional external company - I do not recommend you taking a "do it yourself" approach when using high pressure water cleaning that is outside your normal area of expertise.

I’ve also used 55k to clean a styrene storage tank - polystyrene buildup is as hard as a rock and chips away like glass. That was cleaning with a hand-operated wand, though - I’m not sure if Hx cleaning is done with 55k.

Ultimately, I think MJCronin is right. I wish you the best, but you I suggest you start collecting design data/quotes for a new unit.

 

[EdStainless]

When you HP hyro-lance a HX make very sure that the wand is not inserted or removed from the tube with pressure on. you will peel the tube ends right out of the tubesheets.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[LittleInch]

But have you managed to get it out of the shell yet?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Stefan2211]

Thanks TiCl4, I am not experienced with such high pressure using in hx tubes since they are only 2 mm thick. A straight jet nozzle will remove any hard coking but using a polishing nozzle(under certain angle cleaning the wall itself) might cause damage to the tube itself. We will try to find a service company to try but it's Batam lol........

 

[Stefan2211]

EdStainless, we use service company they have experience so that risk is minimized but yeah, good to know that issue since Indonesia does not have always skilled workers on such equipment. Thanks for yr input

 

[Stefan2211]

LittleInch, no progress yet. We now try for the last time to make a structure for a stronger hydraulic jack (50t) and tray again to break it loose. I feel the problem is not so much the coke at the bottom of the shell as it turned into hard coke so it is not sticky but the main issue might be the jammed bundle due to the tight baffle plates. Bundle or shell might have experienced deformation. If this fails........well cutting the shell will be the last resort with new headache to reassemble later..

 

[MJCronin]

Please keep us informed by posting more pictures of your progress and your final solution.

Please respect us by keeping us in the thread and leaving information that others will benefit from

Thank you and Best Regards ....

MJCronin
Sr. Process Engineer

 

[0707]

Try to hang your shell with a crane and take proffit of the weight of the tubes bundle maybe it works!

luis

 

[Stefan2211]

0707
Already considered but as the 16 t forklift did not work at all, the 9 t bundle will not help much especially crane costs USD 2000/day
I will go with hydraulic jack 50 t and support-frame and will push it from the floating head side. Once it can be moved 100-200 mm then we can pull it. But anyway....this is just the beginning. We will have to clean the bundle and investigate the cause of the leaks (so far 23). Anyone has an idea what can cause such leak as shown in pic? Pressure not exceeding 9 bar. Tubes are 25.4 x 2 mm SA 179
This is just 1 pic with left side large hole.

I am still thinking of mech. stress due to uneven thermal expansion but side walls are looking like overpressure inside of tube. Also leaks happened almost same time. Problem was only recognized when level of 1. column could not be maintained but level of 2. column increased. That indicated the leakage of more than 1 0r 2 tubes. (Still need to check trends which might identify wrong operation)

 

[0707]

Corrosion can be due to localized overheating by uneven thermal expansion caused by the coke fouling. Wish you good luck|

luis

 

[pierreick]

Hi Stefan,
Consider to contact Uraca Indonesia ,they should be able to give you a contact (contractor)

https://www.uraca.com/en/contact/sales-worldwide/

Another Link :

http://sbsindonesia.com/products/equipment/

Not far you may consider to check with Singapore .

Sore
Pierre

 

[Stefan2211]

Thanks Pierre,

Right now all is difficult since we were unable to generate revenue since 2 years. We are bleeding due to some design faults.

 

[Stefan2211]

One of the problem seems to be obvious at the 3rd unit we could actually pull. The baffle plates were made with too small tolerances. The shell has a light deformation on the top where the lifting lug is welded at the shell. When the manufacturer installed the bundle, the baffle plates were already bent (Easy to see since the direction shows it)

Lesson learnt. Keep 3 mm gap between shell and baffle plate. At least if you run with products which require regular cleaning of shell side.

 

[Stefan2211]

So here we go, finally I had time to loosen up the bundles. How did we do it?
1. Pulling didn't work so I decided to push it out with a 100t hydraulic jack
2. Built a structure with a 20 mm plate to protect the tube sheet, and a reinforced 24 mm plate as counter part.
3. Used 8 long stut bolts to balance the forces on the shell flange
5. Welded a support on the plate to support the 30 kg jack
6. Installed marker for the progress of movement
7. ....starting pumping up. Once the pressure was to high the bundle made a 20mm jump and was freed.

Costs:
Steel plates from junkyard incl cutting, drilling, stut bolts, u chanel, welding, rental jack: ~ USD 800

Need retubing now since many damaged tubes can be seen. Corrosion might be the reason due to chem reaction of solvent. Acidity might have not been considered correctly in the design of the heat exchangers. Need to send samples to metallurgic lab to confirm.

Video:

https://files.engineering.com/getfi...ile=WhatsApp_Video_2021-07-12_at_14.20.09.mp4

 

[Stefan2211]

 

[pierreick]

Hi,
Good job and thanks for sharing your progress and video.
Bagus,bagus.
Pierre

 

[don1980]

I used to work in a plant that had similar problems with coke-like deposits accumulating in the shell side. When cleaning was necessary, we sent the exchangers to to be "baked-out". This was done in an exchanger service shop that had an oven for this specific purpose. For coke-like deposits, which seize the tube bundle to the shell, this is a practical and cost-effective solution.

 

[TiCl4]

Stefan - to clarify, you used a hydraulic jack to push the bundle towards the shell side to loosen it, and after it moved you were then able to freely pull it all the way out?

 

[Stefan2211]

TiCl4, I havn't pulled the bundles yet but it will be done this week. It is loose and easy to pull now. It has tremendous coke deposits and was actually cemented in the shell.

Don1980. Unfortunately we do not have such facilities here in Batam and it requires a high temperature I am sure.

 

[MJCronin]

Stefan2211,

Thank you for sharing your field experience and detailed reasoning in solving this challenging HX problem.

Your extra effort by completion of this thread certainly helps us all ..

Thanks again

MJCronin

MJCronin
Sr. Process Engineer