Process leakage into barrier fluid

[Al109]

Hi
We have a pump which pumps 23% NAOH (sodium hydroxide solution) It is protected by a double seal and barrier fluid system (seal pot) system. The pump has the double seal to protect for leakage to the environment and acts as a lubricant to the seal. The mechanical seal is changed out on LIFE every 24 months to avoid seal failures. In general, this works very well for us and we seldom have unscheduled failures of these seals. The seal pot fluid used is filtered water which is piped in to the pot and we have have flow and level change detection inside the pot. This provides information and alarms to our control system to detect abnormal usage of the filtered water which would indicate loss of barrier fluid into the process i.e. a seal leak.
Recently, one of our technicians noted a small drip from the drain on the pot. He noted that the residue form the drip on the ground was white in colour. I took a sample of the seal pot water and had it analysed. pH was 9.4. for reference, i pulled a filtered water sample further back on the filtered water header and it was pH 6.8. I shut down the pump and flushed out the seal pot and resampled for pH. It came back at a pH of 8. I put the pump back on line and ran it for 24 hrs and resampled. The pH was 8.4 (slightly up on the previous result) Everything seems normal with the filtered water pressure to the pot so I am wondering how we are getting contamination back into the seal pot water. Anyone out there have any view or experience of contamination getting back into a barrier fluid system? Appreciate any views on this
Al.

 

[JJPellin]

We are struggling with the same problem on a delayed coker charge pump with an API Plan 54. I assume your piping plan is 53A. Ultimately, a lower pressure fluid is not going to flow into a higher pressure fluid under normal circumstances. So, either continuously or intermittently, the pressure in the process is higher than the pressure in the seal pot. Are the operators dropping the pot pressure to add fluid? Do you have a pressure tap on the seal chamber to verify the pressure there? What style of pump is it? Ours is a two stage, between bearings pump with a balance line. A plugged balance line or a failed throttle bushing can cause the seal chamber pressure on our pump to rise above original design conditions.

Johnny Pellin

 

[SB1976]

You have not stated what type of sealing configuration it is. My guess is that it is likely to be a back-to-back OR face-to-face configuration. In either type, there won't be any primary pump flushing plan like plan 11, 13, etc. and the probable causes of the sealing pressure exceeding the barrier fluid pressure can be attributable to either of the following:-
(1) There could be throat bushing failure or maybe wrong throat bush design (such as an inverted L type).
(2) Improper material selection of pump parts near the stuffing box area can also lead to heavy corrosion resulting in higher pump sealing.
(3) Positive pressure might not be maintained in the pump seal support system (due to failure of one or more system components).