Water quality is one of the easiest and first things to look at when rapid repetitive scaling occurs in seal flush heat exchangers.
Four seal flush plans that immediately come to mind that use heat exchangers; Plans 21, 23, 52, and 53
21- Recirculation from discharge through a heat exchanger to the seal chamber
23- Circulation of fluid in the seal chamber through a heat exchanger. This plan generally employs a pumping ring to help develop head to promote circulation, as well as a bushing in the stuffing box to isolate the cooled fluid from the pumpage
52- Buffer fluid circulated through a tandem seal. The heat exchanger is in the buffer fluid reservoir
53- Barrier fluid circulated through a dual seal. Again the exchanger is in the reservoir.
I'm guessing that you're using the Plan 21 at the moment. The biggest advantage that this plan has include the fact that operators have little to do with the operation of the seal. The higher pressure fluid off discharge shoots right on through the exchanger into the seal chamber and back into the volute. Ideally this plan shouldn't have any valves in the flush tubing, there may be a need for an orifice depending upon the desired flush rate and pressure. The disadvantage is that this plan has the highest dT. The larger the difference in temperature between the flush fluid and the cooling water the faster the cooler will foul. This may be part of the problem you describe.
Plan 23 avoids the large dT because it circulates the same fluid constantly. The seal chamber is isolated from the volute by a bushing so that it is much cooler than the fluid at the pump's discharge. Lower dT- longer seal flush exchanger life. However there is a price- Operators and millwrights are more involved with the seal and flush plan. This plan requires that the cooler is installed a certain way so that the cooled fluid drops- helping to circulate the flush... Thermosyphoning is the term most seal guys will use. Heat rises, cold sinks. Tubing runs should be short within the guidelines of the cooler height requirements, large diameter tubing and large radius bends to decrease friction losses. This is why most seals that use a Plan 23 have a pumping ring of some kind. Even when the pump is not running, the thermosyphon effect will help keep the seal chamber cool. If the seal chamber, exchanger, and tubing are not properly vented at start-up this flush plan may not work properly. Bubbles tend to keep the fluid from circulating, a high point bleed is required... depending upon the pumpage you may need to make arraingments to sewer or recover small amounts of process fluid.
Buffer fluid circulation is generally the same idea as a Plan 23. This time the fluid is circulated between two seals at a pressure lower than the stuffing box and, possibly higher than atmospheric. The idea is containment of the inevitable leakage across the primary seal faces. Generally the reservoir is piped to a flare or knock out system... The same problems of a Plan 23 apply here.
Barrier fluid on the other hand is pressurized higher than the process fluid in the stuffing box. This means that the barrier fluid wets both sets of seal faces, the process fluid is isolated as long as the barrier fluid remains pressurized and as long as there is barrier fluid. Operations should monitor the fluid level in the reservoirs (Both 52 and 53).
Talk to your favorite mechanical seal vendor about the flush plan you are using ask if the heat exchanger is piped or installed correctly. I'm sure that they would be happy to discuss alternatives or how to correct any installation or operations problems.
