Hello
Can someone comment for me on the situation we seem to have ourselves in.
We are working on a project where the end-user has requested the use of a split mechanical seal
We are unsure as to the suitability of a split mechanical seal for this application based on the following conditions and are looking for your thoughts and opinions
The pump is a horizontal split case type pump.
Design conditions are 300 l/s @ 48m TDH
Suction pressure is approx. 2 psi
Because of the control set up, the singer valve will be closing while the pump is still running at 100%. There is no VFD to slow the pump down so the pump will be dead heading for approx 30-45 seconds every time the system is shut down with of course the resulting cavitation.
Can you tell me if you feel that a split type seal (or any type of mechanical seal) would be appropriate for this application. Past experience tells us that this is not the best option but we would like something from others to include in our submission
Ed, an interesting mix of Units there.Originally Posted by EdH
Also you provide nowhere near enough information for an informed opinion, from me anyway. Some pump designs are not overly adversely affected by cavitation, while others will break stuff long before cavitation begins, and then there are others...
unfortunately, those are the units I have to work with. The suction gauge reads PSI while the duty conditions were given by the Engineering Firm - but i suppose I could have converted the psi to meters to maintain commonality.
I'd like to supply you with whatever information you think might be needed to comment
Split seal where? Rod end cap seal?
Like Dave asked... What kind of pump?
Sounds to me like your issue will be more with the type of pump you're using than any sealing arrangement. There's only one style pump I'm aware of that will tolerate flow being slammed to off for indefinite time and not be bothered by the situation at all. Can anyone out there guess what type I'm referring to?
My original post identifies the pump as a horizontal split case and as such has a mechanical seal on each end of the shaft.
The RPM is 1800
ambient temperatures
Starts and stops will vary by demand but likely 2-3 times per day at a minimum
The shaft is sleeved and the diameter of the sleeve is 2.75”
We were given NPSHa of 8.7m by the engineering firm. The pump has NPSHr of 20.9’ (6.4m) so we should be OK there
The control valve (singer) takes approx 7.5 minutes to shut fully. The pump gets noisy and vibrations are noticable when the valve reaches 25 percent open. from that point it takes 30-45 seconds for the valve to reach 0% open (fully closed) As read on the control panel. for that final 30-45 seconds the pump sounds as though it is pumping rocks and vibration increases.
The pump is in use in a municipal setting on potable water drawn from a lake.
Further questions would be: Is this a 'standard' pumping setup for a municipal application? If a mechanical seal had to be used, are we better of with a standard cartridge seal. Since this is just water with no worry about emissions, would packing not be a better option?
Thanks for taking the time to look and help
Last edited by EdH; 08-29-2011 at 01:48 PM.
I suspect that the split seal assembly is being used for manufacturing and assembly reasons. But, without a drawing or picture I'm only guessing based on previous experience..
BTW, maintainability is another reason....
I would say you are correct on the maintainence point.
Our biggest concern is that the vibration caused by the deadheading will lead to premature seal failure, especially with the split seal. with the excessive vibration openingthe seal faces, not to mention the stresses on the bearings
Ok, then moving forward, I suggest that you post the split seal mechanical and performance specifications or a link to it. With detail performance information we can examine the applicability.
I'm pretty sure that matching the design specifications/performance requirements to a speciic seal is achievable..
I think the pump style you are using will have a very short life dead-heading it. If you could find a pressure compensated vane pump with the ratings your after that'd be the way to go. Otherwise you might be able to give the pump you're using a fighting chance by creating a by-pass line for it rather than dead heading it. Divert the flow back to tank or wherever it's coming from as you are shutting down wherever it is that you are pumping two. Even just a short loop back to a Tee at the pump inlet.
"There's only one style pump I'm aware of that will tolerate flow being slammed to off for indefinite time and not be bothered by the situation at all. Can anyone out there guess what type I'm referring to? "
Ok, I will byte.
I am not really sure if you consider this a "pump",
but back on the farm while growing up,
we called them "gravity pumps".
Others called them "siphons".
We used them to get water over the edge of the head ditch
into the various irrigation channels between rows of crops.
Yeah, but that's like saying "this has two doors so it must a car." Is it a submersible? How deep? Above ground? Turbine, Centrifugal, positive displacement, multi-stage?
As for PSI, I fail to see how "suction" can be indicated with pressure. Mutually exclusive.
Even ignoring all of the above, I am with RJ and suggest you do something to allow overpressure during the valve cycle, to return to supply. Providing the split seal is well designed, which in a Municipal water supply, I would assume it is, it should be no more or less effective than a full-circle seal. Ford had them in their 302/351 V8s up until about '98
Is this electrically driven? If so then maybe indicating to the client that the significant rise in load during the shut down is a terrible waste of energy. Then suggest the bleed-back system as a long term power saving method. It would pay for itself in two years.
Dale I forgot about siphon pumps. They are slick aren't they...! I was thinking about the pressure compensated vane pumps. You can dead head them all day and they won't mind. Thought of another type of pump that doesn't mind being dead headed... But can't remember what they're called. They have a plate inside that wobbles around and strokes several pistons as it rotates. As back pressure increases the plate will level out and not wobble which stops the pumping. But I'm also thinking I recall they aren't very sturdy either.
Far as the split case pump goes... That typical design does not like to be dead headed. That scary sounding cavitation you hear is the pump begging for mercy. You might get away with it a time or two but each time runs risk of wearing the vanes or breaking the thing.
The best thing to probably do would be to contact the pump manufacturer and pick their brain some. I can hear the phone conversation now... "Hello. We have this pump we're dead-heading a couple times a day for a few minutes. what do you think about that?" "Well sir I will grab a few more for our inventory. Thank you very much for your business..." (humor... nothing personal.)
One of the key words in the original query that jumped out at me was "municipal." The problem you tend to run into when a situation involves any form of government is that they're spending other peoples money. It doesn't hurt so bad to keep spare $10,000-$15,000 pumps on the shelf when it isn't coming out of your own pocket. If any business ran itself like our various levels of government they'd go under so fast it'd make your head spin. It really is a shame...
That would be a Swash Plate Pump, RJ.
Posting Permissions
Reply With Quote