Thermoshyphoning in sealing

[mec5216]

hello

According to API 682 (shaft sealing systems for centrifugal pumps), the API plan 23 requires the heat exchanger to be placed some high above the mechanical seal in order to provide thermoshyphoning, what would help the pump ring to delivery head to make the sealing system work. Although API says such thing, I have made some calculations and got the conclusion that the head delivered by the thermoshyphon effect is much smaller than the head delivered by the pump ring. That can be even 100 time smaller!
Some people say thermoshyphoning has the only purpose of cooling the seal faces when the pump is out of operation. But if it's not operating, why the need to cool them,as no heat will be being generated?

I would like you to make some comments about my opinion.

Thank you

 

[FredRosse]

"head delivered by the pump ring"
Is this head always defined? Natural circulation, though small head, does give enough flow for many conditions.

 

[Compositepro]

Convection never provides much head but it can create high flow if there isn't much resistance to flow.

 

[Ripz]

I've always heard that the thermosiphoning effect creates higher flow than a pumping ring. Is that wrong?

 

[1gibson]

Yes you need head to elevate a fluid in an open system, but in a closed system it also drains back down, which assists the flow. So, if the HX is 2 feet above the seal, you don't need to generate 2 feet of head before there is circulation. The thermosyphon effect due to elevating the HX is more than enough to make up for the difference in height.

 

[mec5216]

But the difference in density of the fluid changes very little along with the change in temperature. And even if we consider this difference in density and a weight of column of 2 feet, the system would get very little head from thermoshyphoning. Two 90-degree bend may be enough to overcome this head and stop thermosyphoning.

 

[mbt22]

Thermosyphoning does not have to overcome the weight of a 2ft column of water, it just needs to provide kinetic energy for circulation. Bends would slow down the circulation but not stop it (counter-example - if flow is stopped, you have no pressure drop due to the bends and less dense fluid under more dense, so there's nothing to stop flow)

Matt

 

[mec5216]

ok, I agree with you. But I meant something different. Let's suppose there are two tubes lengthening 2ft. The first tube is filled up with a fluid in a known temperature. The second tube is also filled up with the same fluid, but in a lower temperatur (let's suppose 47ºF lower compared to the fluid in tube 1). As the fluid in second tube is colder, its density is higher, thefore the pressure at the bottom of tube two is higher than the pressure in tube one. This difference in pressure is what would power the thermoshyphon effect (considering the heat exchanger 2ft high). But if you make some calculations, you will find out this pressure is too small compared to the pressure given by pump ring in a sealing plan 23.
That's because I said thermoshyphon head is much smaller than the head given by a pump ring.