Definition of 'clean' fluid 1

[TPL]

Does anyone have a reference which defines how 'clean' a 'clean' fluid for use in a Plan 11 system should be?

I have a system where a pump fitted with a Plan 11 flush (single seal bellows) precedes a Plan 53B (pressurised dual seal bellows) and I'm not entirely sure that the Plan 11 is appropriate in this application - the fluid is at ambient temperature but is known to contain varying amounts of sand (typically 5mg/kg under normal conditions), although there are filters/strainers designed to remove everything above 10microns at strategic locations in the system - I would have thought that in terms of damaging the seal faces, the smaller particles (<10microns)would have been a greater threat especially during initial startup and following trips/process upsets when much greater (but unknown) quantities of sand are likely to be present - I'm questioning the seal application and taking the view that the selection process should take account of worst case operating condition (ie. lots of sand <10 microns).

Anyone been here before and wiling to share their thoughts?

 

[Pumpsonly]

The correct description of your seal should be pressurized dual seal with Plan 11/ 53B flush system.
Your concern about the smaller particles <10microns damaging the seal faces is unfounded as the Plan 53B will ensure only clean liquid is seen by the seal faces.

It is very seldom Plan 53B is used for a water services at ambient temp. A single seal with Plan 31 or 32 would normally be suffice. Perhaps they provided this to cater for the worst case scenario as you described..lots of sand.

 

[flexibox]

TPL,
it would be good if you can explain what the problems are that you currently experience. As Pumpsonly says Plan 53B ensures that a clean barrier fluid is at the seal faces and not Dirty process fluid. I dont understand why you have a metal bellows anyway, and suspect that sand is being trappend in the bellows convolutions. I suspect Plan 11 was recommended with Plan 53B to encurrage circulation at the inboard seal but this could also pose a problem with your current seal type.
If you provide more information shaft dia, seal arrangement, attach drawing perhaps then we can make more informed recommendations.

 

[TPL]

Thanks for the your responses and apologies - I should have written things a little more clearly.

The system is MEG reclamation: the first pump in the flow line is fitted with a single cartridge seal and a Plan 11 flush. The next pump in the flow line is fitted with a pressurised dual seal bellows and a Plan 53B flush with water/glycol as the barrier fluid. Alkali is added downstream of the first pump to precipitate corrosion products (mostly scale, but including carbonates of iron and calcium) so I can see why a pressurised dual seal with bellows and a Plan 53B flush has been selected for the second, downstream pump.

However, sand is present at both locations and the fluid going into the first pump and through the Plan 11 flush is not exactly clean, as it is known to contain varying amounts of sand (typically 5mg/kg under normal conditions), although there are filters/strainers designed to remove everything above 10microns at strategic locations in the system and as before, I would have thought that in terms of damaging the seal faces, the smaller particles (<10microns)would have been a greater threat, especially during initial startup and following trips/process upsets when much greater (but unknown) quantities of sand are likely to be present

Bearing in mind the seal /flush selected for the second pump, I am questioning the seal (single seal cartridge) and Plan 11 flush that has been selected for the first one, with a view that the selection process should take account of worst case operating condition (ie. lots of sand <10 microns).

So I guess my question boils down to ‘how clean must a fluid be in order for a Plan 11 arrangement to succeed?’

As an afterthought, I know the environmental issues with a Plan 11 in this application are wrong, but that’s another can of worms

 

[flexibox]

Normally filtration on wet seals is 10 micron or better. In your case if you need to Use hard faces and if you can change to plan 31 see if it works for you then try plan 32 if you can afford dilution. Can you tell us if you are experiencing regular failures and what you see?

 

[pumpkind]

I subscribe to the answer you have received from Pumpsonly - the combination shall work fine - the seal faces do not get process fluid at all because the barrier fluid has a higher pressure , therefore the fluid passes the seal faces in the opposite direction (against) to plan 11 flow

 

[Airsporter1st]

Beware that although in principal the condition of the Plan 11 flush flow should not matter, in practice there are several potential problems:

a) If the plan 11 flow contains sand then it can cause erosion of the seal components over which it flows - especially softer materials such as carbon. It can also erode the orifice, resulting in ever increasing flow.

b) Presence of sand in the flush flow can cause clogging of springs or bellows especially if the latter is stationary.

c)Presence of sand can result in clogging of the flush line such that there is insufficient flow.

 

[1gibson]

If you add a 31 you also need to add a 13 or there will be no flow. This gives 2 paths back to suction (the 13, and the dirty side of the 31) so you may lose a tiny amount of efficiency. Depends on the conditions, you won't even notice it unless very high pressure/low flow.

Also might review the location of the plan 11 flush, most cases it is desired to "aim" the flush directly at the seal faces for optimum cooling. This usually means flush connection in the seal chamber, rather than the seal gland. Not a good plan with solids, if that is your configuration you may be able to install a flush diffuser.