Tag Archives: plan 53

Cooling Water on Seal Pots

Before retirement, I was regularly asked “Does that seal pot really need cooling water?”  The answer then, as now, usually was “Yes”.  Here’s a quick back-of-the envelope evaluation of a seal pot operating without cooling water and a simple guideline to when cooling water is needed.

The convective heat transfer coefficient to air is in the range of 2 to 5 Btu/hr sq ft°F depending on whether the wind is blowing or not.   This is very low heat transfer coefficient.  In comparison, water cooling coefficients are on the order of 50 to 300 Btu/hr sq ft °F.

If the surface temperature of the seal pot gets too hot then there can be a safety issue.  Many End Users do not want the pot to be at temperatures much higher than 150 °F or so.  This evaluation is based on a maximum seal pot (that is, buffer/barrier fluid) temperature of 150 °F.  Will the seal operate satisfactorily at a temperature hotter than 150 °F?  Almost certainly – the pot will just be hot.  Is the buffer/barrier fluid suitable for temperatures hotter than 150 °F?  That depends on the fluid but probably so.

Remember that the surface temperature of the pot will be hotter than that of the surrounding air.  If the pump and seal are to be operating in the summer then a summertime air temperature should be considered.  Also, for most places, sometimes the wind is not blowing.  Therefore, the base case is

Heat transferred = area x temperature difference x heat transfer coefficient; that is
Heat transferred from pot = pot area ft2 x (150 – 100) °F x 2 Btu/hr sq ft

The external area of a typical seal pot is (very roughly)

  • 2 gal –> 3 to 4 sq ft
  • 3 gal –> 4 to 5
  • 4 gal –> 6 to 7
  • 5 gal –> 7 to 8

So the base case heat transfer from an air cooled pot is roughly

  • 2 gal –> 300 to 400 Btu/hr
  • 3 gal –> 400  to 500
  • 4 gal –> 600 to 700
  • 5 gal –> 700 to 800

This is not very much heat transfer.  On a 50 °F day, with the wind blowing, those heat dissipation values could be multiplied by 5 but that is still not much heat transfer.  Of course, whatever heat is generated will be transferred by virtue of the seal pot temperature becoming as hot as necessary.

In addition to heat generation by the seal faces, there is heat absorbed from a hot pump into the buffer/barrier fluid; this is called heat soak.  For reference, according to API 682, the heat soak from a 200 °F pump to a 150 °F buffer/barrier is 1200 Btu/hr for a 2″ seal size pump.  Therefore, an air cooled pot should not be recommended even for a warm pump having small seals.

For tandem (unpressurized, Arrangement 2) seals, the usual requirement is that the seal pot has to dissipate the heat generated by the outer seal plus heat soak, if any.  As a point of reference, the outer seal of a tandem arrangement generates roughly 100 to 300 Btu/hr per inch of seal size at 3600 rpm (depending spring load, face width, materials, etc.).  Therefore, it turns out that the limits for an air cooled pot with Plan 52 on a hot summer day, no wind, 150 °F maximum allowable pot temperature  would be something like

  • 2 gallon pot –> 1.25″ seals and smaller
  • 3 gal –> 1.625″ seals and smaller
  • 4 gal –> 2.125″ seals and smaller
  • 5 gal –> 2.5″ seals and smaller

For double seals (Arrangement 3) using Plan 53, the usual requirement is that the seal pot must dissipate the heat from both seals plus heat soak, if any.  Therefore, an air cooled pot will almost never be adequate for Plan 53 since the minimum pot pressure must be at least 25 psig and heat loads would be more than double that of the tandem seals (Arrangement 2).  That is, heat loads for Plan 53 would be about 400 to 1200 Btu/hr per inch of seal size.

All the above calculations are very rough and very conservative but still representative.  In fact, the rough calculations readily demonstrate that most dual seal applications need water cooling and show why this is so.  But the rough estimate also shows that there is a place for air cooling and can provide some guidelines for defining those limits.

Based on the above analysis and estimates, here are some guidelines for application of cooling water:

  • Pump temperature above ambient — use water cooling
  • Plan 52, when pump temperature is ambient or cooler – use air cooling
    • 1800 rpm — pot size should be 1 gallon per inch of seal size (2 gallon minimum)
    • 3600 rpm — pot size should be 2 gallon plus 1 gallon per inch of seal size
  • Plan 53 — use water cooling

And a reminder:  Seal pots will always run hotter with oil (even the low viscosity synthetics) buffer/barrier fluids than with glycol/water solutions.

For those who really want to further investigate this question using more sophisticated calculations:  Good Luck!  There is not much information available.  A conservative approach is to order the seal pots with cooling coils but do not connect the cooling water to the coils unless proven necessary by actual operation.