9 replies to this topic

[curious_cat]

Generally welding Titanium to dissimilar metals like Carbon Steel or Stainless Steel is not possible (or at least that is my impression). 

 

If so, what are typical design options for attaching a Titanium tubesheet to a less exotic shell in an application where the tubeside fluid demands Ti but using Ti for the  Shell Side  (steam) would be prohibitively expensive. 

 

U-tubes are not feasible since this is a plugging prone evaporator. Normally, I'd go for something like a TEMA fixed tubesheet exchanger (BEM) with shell welded to the tubesheet. 

 

Is there such a design where the shell doesn't need to be welded to the tubesheet on either end but is simply gasketed? 

 

T & P are relatively modest and there's no need for absolute / critical isolation of shell and tube side fluids. Single pass shell & Tube. Shell side is clean and no shell access is needed for cleaning. It is a small HEX, total Heat Transfer Area approx. 30 m2. 

Edited by curious_cat, 01 September 2015 - 01:55 PM.

[srfish]

You could use a design like that used in the outside packed floating head heat exchangers. In TEMA page 2 refer to the type P head types. The split ring shown would not be needed.

[curious_cat]

You could use a design like that used in the outside packed floating head heat exchangers. In TEMA page 2 refer to the type P head types. The split ring shown would not be needed.

 

 Thank you!

 

I think that's a great idea. It would obviate the need to weld the Ti-tube bundle to anything else. 

[curious_cat]

You could use a design like that used in the outside packed floating head heat exchangers. In TEMA page 2 refer to the type P head types. The split ring shown would not be needed.

 

How about a TEMA Type T? Would that work too? It looks like that saves on the use of the exotic tubesheet MOC. 

 

In Type P the bonnet seems welded to the tubesheet? If so the whole bonnet / channel would need to be of Titanium?

[PingPong]

How about a TEMA Type T? Would that work too?

Although not knowing the details of your process, I would think yes.

 

The exchanger can then include a metal bellows coupling (expansion joint) at the floating head as in this drawing:

 

 

I don't know for sure though whether bellows come in titanium.

A point of concern might be that the grooves in the bellows get filled with solids thereby hindering its function (just thinking the worst now).

 

Such are often used in vertical thermosyphon reboilers if it must be possible to pull the tube bundle for cleaning shellside.

[curious_cat]

Thanks @PingPong

 

The process is rather simple. Low pressure steam (1 barg max) is used to boil a saturated NaCl brine. The steam is on shell side. Brine (max pressure 3 barg) is on tube side. Both side single passes. 

 

The tube and tubesheets will be Titanium. The shell and other components are preferably cheaper MOCs, not titanium. Ideally the bonnets / channels etc. can be rubber / epoxy coated Carbon Steel. 

 

The challenge is to restrict the use of Ti only to tubes and tubesheets if possible. Since welding Ti to CS / SS etc. is difficult hence the search for closures at both ends that won't need welding of the tube bundle to anything else. 

 

The fluids are not toxic etc. and some contamination of one with the other won't be a problem. 

Edited by curious_cat, 03 September 2015 - 07:58 AM.

[curious_cat]

@PingPong

 

Does a TEMA Type T have to have bellows? Titanium bellows would make the design more difficult. 

 

How about an AES type? Would that fit the constraints of not needing to weld the tube bundle to anything else?

[PingPong]

The tubes and shell will expand differently, so without fixed tubesheets that difference in movement has to be absorbed in some way, but maybe you can also account for that in some other way outside the exchanger, as you would also have to do if you used a P shell.

 

Difference between AES (or BES) and AET (or BET) is only the backing device, which allows the S to have a smaller diameter shell than the T. This is especially important if the shellside h benefits from minimising the bundle bypass flow and maximising crossflow velocity, but with condensing steam that hardly matters.

 

You stated before that it is a small exchanger of only 30 m² so I begin to wonder whether it would really be prohibitively expensive to make the shell also titanium and use BEM.

 

Or otherwise use steel shell BEM with steel tubesheets that are titanium clad, or if possible coated with ebonite or epoxy, on the sides that are in contact with the brine.

Edited by PingPong, 03 September 2015 - 10:55 AM.

[curious_cat]

The tubes and shell will expand differently, so without fixed tubesheets that difference in movement has to be absorbed in some way, but maybe you can also account for that in some other way outside the exchanger, as you would also have to do if you used a P shell.

 

Would a expansion joint on the shell take care of that? 

 

 

Or otherwise use steel shell BEM with steel tubesheets that are titanium clad, or if possible coated with ebonite or epoxy, on the sides that are in contact with the brine.

 

 

 

Yes, I think clad is a good idea.  Ti-clad is what I was considering. 

 

Epoxy clad is also a great idea & much cheaper than Ti-cladding. Have you come across an epoxy clad tubesheet before? I wonder if that would be do-able. I'm already considering epoxy coating for the channels / bonnets. 

[PingPong]

Would a expansion joint on the shell take care of that?

I don't see how that makes much difference. A one-tube-pass BES or BET needs a connection from the floating head cover through the shell cover to its nozzle. So you will have a sealed sliding connection through the shell cover to allow the expansion of the tubes.

 

Ti-clad tubesheets are in use, so if you use one-side cladded tubesheets you can weld the steel tubesheet to the steel shell without any problem to obtain a BEM. If necessary for temperature reason the shell can have an expansion joint (bellow).

 

Another option may be to use a BEW type with an externally sealed floating tubesheet inside. Seal is by two O-rings.

Suitable up to 375 ^oF = 190 ^oC according to: http://www.auroraind...w-or-aew-or.htm

 

I have no experience with hot brine service, so I don't know whether epoxy or ebonite coating can withstand the temperature for a long time.