Exchanger Tube Thickness and HTC

[robhanson]

Hi All,

I'm replacing a tube bundle in shell and tube exchanger which has cracked with a new, slightly thicker one. The old tubes were 1.2mm thick, new 1.6mm. I'm wondering (and can't seem to find anywhere) how much this will affect heat transfer. Looking at Fourier's Law for conductive heat transfer it says that the relationship is linear between different thicknesses (i.e. it will be 33% harder for heat to travel through this extra 0.4mm which does make sense), however saying that the HTC will be 33% reduced just doesn't feel right to me. It feels like there's something big I'm missing, but just can't put my finger on it. If anyone has any advice it'd be much appreciated.

Thanks,
Rob

 

[SNORGY]

I would suspect that, if the flow rates do not change, you will realize at least some offsetting of the increased conductive resistance by the improvement in the inside film coefficient. If this is the case, since the film coefficients otherwise often tend to dominate in the calculation of OHTC, the reduction might indeed be less than 33%.

 

[TD2K]

What you aren't considering is the heat transfer through the other resistances between the shell side bulk fluid and the tube side bulk fluid, those dwarf the resistance through the tube itself. Or in other words, the dT across the tube is quite small so 33% due to the increased thickness is still a very small temperature drop.

 

[robhanson]

Thanks for both your replies, after a lot more thinking that is what I started get to as well. Going to head home and dig through old uni notes, I'm sure I can remember doing some equations on inside and outside coefficients sometime...

 

[MJCronin]

TD2K is correct...... as usual.

Investigate the "overall heat transfer coefficient" for shell and tube heat exchangers.

You will find that the increased wall thickness will only be a small part of the coeffiecient.

Chapter 11... Perry's handbook..... and

http://web2.clarkson.edu/projects/subramanian/ch302/notes/shelltube.pdf

 

[robhanson]

Lovely, thank you. Found it in my old notes too, feel a bit of a dunce for not remembering...

 

[rmw]

I think you have probably found out what you need to know about the HTC. If you don't change your pumps or whatever motivates the liquid, the flow through the Hx will be reduced (assuming the tube O.D. remains the same). If you maintain the flow rate by whatever means - higher discharge pressure from the pumps for example, the pressure drop through the bundle will be higher than before. If you maintain the flow rate, with the resulting higher tube velocities, that too will affect your HTC.

Be careful not to exceed the allowable tube velocities for whatever tube metallurgy you have.

Why pray tell are you increasing the tube wall thickness?

rmw

 

[robhanson]

We've got a shell and tube HX which has ruptured - have found cracking in a number of the tubes. When sent away for repair the guy doing the repair said he would never have made the tubes so thin (1.2mm) and would go 1.6mm as a minimum. Nothing more technical than that I'm afraid!

 

[MJCronin]

Rob,

Since you are suspect of the original design of your HX, I suggest you consider getting a quote for a new replacement HX from a fabricator.

I would use the original TEMA data sheets, if still avaialble from your records.

Compare the recommended tube thicknesses (and many other design details) of a replacement unit with your existing one.

Suggest that you also compare performance results in an "off original design" condition........ This is simple to do in most modern HX design programs.

Best regards

 

[robhanson]

Hi,

I have the quote from a fabricator to replace the entire unit, which is what we'll end up doing. I was just trying to find how much the resizing will affect our operation, but from everything I've done (and other engineers) all seems fine.

Thanks everyone for your help,
Rob

 

[SNORGY]

What tube material was involved?

For example, if, say, SA-178 had been specified before, could one now specify SA-179?

Might save you a new unit...

 

[robhanson]

Material is 316 SS and needs to stay as is (GMP reasons, etc)

 

[SNORGY]

I apologize, I should have elaborated.

If tube cracking was the prevalent concern, you might have had the option to review seamless tubes in lieu of welded tubes.

 

[rmw]

What is the service for this Hx? How high is the pressure or how corrosive is the service that requires 1.2mm thick tubes in 316SS?

Did your vendor take the increase in tube thickness into account in the new design or is it a replacement in kind with thicker tubes? If the former, then you are probably OK, except that the Hx is larger somewhere - longer, larger diameter, etc. If the latter, then your performance is going to be affected as per my previous post.

And to my colleague Snorgy, that would have been the last thing I would have suggested - changing from welded tubing to seamless assuming that is the direction change would occur. When I use welded tubing, I expect the strongest part of the tube to be the weld. Seamless Hx tubing hasn't been a friend of mine.

rmw

 

[SNORGY]

rmw,

Fair comment. Without knowing more about what was meant by "cracking", and not knowing the mechanism at play, perhaps it was premature to suggest that maybe it might be a seam problem. Would you think it would be worthwhile to see where the cracking is occurring, as well as whether they are transverse, longitudinal or random?

 

[Christine74]

Your overall heat transfer coefficient probably won't change much at all but note that your pressure drop will certainly increase.


-Christine