Seals for Caustic application 1

[RGRAJANI]

For 5% and 10% Caustic @ 40 Deg C application which is the best (Optimum) flushing plan ?

We can not use plan 32. Plan 11/62 are in use but still problem of crystalisation and seal failure exist.

 

[JJPellin]

Depending on the cleanliness of the stream, I would like a sealless pump in this service. We have one set of caustic pumps that have had seal reliability problems. We are converting them to an "upstream pumper" seal design. This seal is very much like a gas seal. It uses spiral grooves to pump water from an atmospheric tank into the process. Normally they would not allow this because they don't want to water down the caustic solution. But the rate of water injection is tiny (a few ounces per day) and insignificant. We had previously tried plan 11 and a modified plan 54 with very poor results. The particular seal we are installing was described in a paper presented at the International Pump User's Symposium last year.

 

[RGRAJANI]

JJPellin,
Thanks for valuable input. Have you installe these seals ? Share your experience.

 

[longeron]

Right on the money...

We have several in use at the refinery that I service and they are performing quite well. These are much nicer than a single seal with steam or water quench because the upstream operates like tandem or dual seal and eliminates the housekeeeping issues of a quench. No more questions from operators about the fluid comming out of the seal.

Be sure to mount the reservoir properly, as the tubing to and from the Tee or cross in the flush plan is what acts like a heat exchanger for the buffer fluid. In some applications finned tubing and insulation on the return leg might be a good idea. The reservoir should be high enough to offer enough resonance time and head (a little back pressure) for the system.

Look for an article in an upcoming issue Pumps and Systems about upstream pumping seals.

 

[JJPellin]

Unfortunately, our upstream pumper seals are not in service yet. I think the seals are on order, but I am not sure about the project to build the support system. This may have been bumped off to 2007 because of budgetary constraints.

 

[Milkboy]

Just an additional thought to this issue, RGRAJANI.

You say Plan 11/62 is still giving you crystallisation
Has the flow rate into the box of the P11 been confirmed as being sufficient to prevent crystalisation/sedimentation?

HTH

-
Milkboy

 

[RGRAJANI]

Milkboy, Flush rate mentioned in seal dwg. is 5 lit/min. I understand this is standard rate and does not indicate special consideration for Caustic application. One seal vendor is saying increasing the flushing fluid temp. by steam tracing will help. ; But we are convinced with Upstream pumping and is specified for upcoming projects.