Heat Medium Flow through Expansion Tank

[Ahmed Elarbagi]

Hello everyone. I have been discussing a lot with different vendors and very experienced process engineers on hot oil systems. There is a lot of different opinions on whether or not hot oil should be normally flowing through the expansion tank.

Some propose that flowing through the expansion tank regularly is more logical; maintain pressure and allow for expansion as needed and not have to worry about pressure buildup due to thermal expansion as the expansion tank would have a PSV.

Some propose that the expansion tank should be bypassed and only one of the lines to it left normally open to account for expansions as needed.

Any thoughts on what the best practice is, and if reasons for why one approach is better than the other? Which of these valves would you NO and which NC, and why?

 

[georgeverghese]

I prefer a flow through the expansion drum system
a)It allows for better disengagement of vapor bubbles that may be in the HO return stream from any high pressure leaks at user heat exchangers tubes or tube-tubesheet joints. So HO pump does not see vapor bubbles in its feed
b) With a through the drum scheme, there is more thermal inertia ( with drum inventory included) to cushion the HO fired heater from sudden changes in heat demand. This then keeps the fired heater temperature control loop more stable and prevents trip on TSHH - this is one of the primary functions of the HO drum.

 

[pierreick]

Hi,
You may want to refer to this document (link attached) from Eastman supplier of Therminol fluid.

https://www.therminol.com/sites/therminol/files/documents/Thermal_Expansion_Tank_Design_and_Operation.pdf

To me, I like the idea to use it as an expansion drum only, for reasons explained in the paper.
Regarding degradation, temperature and O2 are the catalysts, the reason why N2 blanketing is a must.
BTW, those guys are specialists, and you may want to ring them for confirmation. At least your hot oil supplier should help you.
Pierre

 

[LittleInch]

I think the set up you have indicated allows for both, but "normal" flow should be bypass IMHO.

Why? One thing to think about is the actual location of the nozzles and the flow velocity in the nozzles which could result in significant flow disturbance and jetting, especially if there is any gas or vapour in the liquid and you have low liquid level in the tank. If you want to go through then I thinkyou should have entry d exit into the ends of the vessel.

If you want or need the expansion tank to remove vapour, you prob need to have a small section of larger pipe and have the expansion tank connection on top of say a 12" pipe.

This is a pretty sophisticated expansion vessel so the normal for this tank wouldn't work for a more simple expansion tank which you often see with a bladder or similar.

So there is no one option which works and each system can be different, hence why design engineers exist....

Also that set up of valves allows flow through when needed and bypass when not or if issues arise due to flow through.

 

[Compositepro]

In large, sophisticated process loops, a flow through system is the way to go for all the reasons mentioned. In small, simple hot oil loops, bypass is more often preferable. Mainly because the oil in the tank can be kept much cooler at the liquid/gas interface and nitrogen blanketing of the tank can be avoided. Most small plants don't have nitrogen as a utility, so this is a big deal.

 

[Latexman]

What does your hot oil supplier say? Our hot oil supplier has dedicated resources to provide guidance to us. They have complete P&IDs for their recommendations. They also analyze our oil yearly and give their advice on what to do to extend the oil life. Call your supplier.