Starting Up a Pump 7

[Milkboy]

Hello

I have to do a shutdown and Start up on a set of pumps
(Centrifugal)

For Start Up, the proceedure says

Open Suction Valve, Vent Air from Pump, Start Pump
Open Discharge Valve.

So, The pump will pump its maximum head
with the discharge valve closed at zero flow right ?

What about the pressure in the suffing box
at the Mech seal ?

How do ensure the seal will be OK

rho g h = p if I know what the max head
of the pump is at zero flow (?)

But am I right in thinking the box
pressure is a third of this pressure or similar

Basically I want to know the pressure
of the stuffing box at closed discharge



TIA


Milkboy







-
Milkboy

 

[uncleal52]

shouldn't you open the discharge valve before starting the pump?

 

[Milkboy]

Arrrrh ... well

I thought that but basically didnt
want to ask in a meeting full of folk !
(no balls I know but I aint worked here long)

I thought keeping the valve closed
was maybe a way of venting/priming (?)

Once running a closed valve head you
can open the discharge valve and away you go

This is why I thought the mech seal
may be effected.






-
Milkboy

 

[DLANDISSR]

Milkboy:'
The pressure at the mechanical seal at shut off head
is not an issue!!

I would not start the pump with the discharge valve closed.
Start the pump with the valve open. If you want to know the max pump pressure at shut off then close the valve slowly so as to not create any shock waved and cause water hammer.

 

[Milkboy]

Thanks

Weird thing is all the pumps I have on the list here
are

Start Pump before Opening the Delivery

except One item which says

Start Pump with Delivery Valve open
(which is listed as a CIRC pump)

(I assumed Delivery was Discharge ! I hope i havent
confussed this issue)


I assume this has always been the way here/site

It there ANYWAY in which this is acceptable
where Pump Start Ups are with Delivery Closed







-
Milkboy

 

[Montemayor]

Milkboy:

The startup and shutdown of a centrifugal pump is specific to the user's application - one can't expect that general directions will apply 100% of the time. That's what engineers are supposed to oversee and ensure.

A pump's proper and recommended startup depends on the type of pump, the fluid, the pump's configuration and piping, the process downstream, the flow and pressure controls installed, and the hazards involved (if any). There may be other factors I left out, but those are just some of the main ones. All manufacturers send you General (un-specific) startup instructions. I presume these are the ones you allude to. Your specific, actual application and configuration are the main and most important items to consider first. You can startup with your discharge open; but you'll impose a heavy current draw on your electric motor and you should be ready to handle the discharge downstream. This is the way I usually prefer to startup because it automatically "vents" the air from the pump if I direct the initial discharge back to the suction source. However, this takes more instrumentation. Also, you must have an automatically flooded pump - i.e., you should have a positive static head over the pump. These are just some of the examples of what I mean by "actual application and configuration" and how they influence your decision on how to startup. It's not like a recipe book.

What about the mechanical seal? What do you mean to say? The mechanical seal is there and will do its job of sealing if you subject it to its function. The manufacturer designed it to seal and you don't explain your concerns as to why you should worry about it. If you don't run it dry (which is Rule #1), you should not have a concern. The pressure at the face of the seal will be characteric of that pump's design - whatever it is. The seal is surely designed to operate at deadhead pressure - although I wouldn't subject a pump to this extreme. Sustained operation at deadhead (as I'm sure you're aware) pressure is just bad, bad practice that puts abuse on a pump - mechanically and thermally. Don't do this unless you want to buy more pumps.

If you want to know the pressure of the stuffing box at closed discharge, inquire with your pump manufactuer. This is not a usual question from a user, but if you're interested in knowing, I'm sure people like Goulds and other recognized pump manufacturers will tell you. You haven't said what kind of mechanical seal you're using so I assume you mean the process face of the seal, directly behind the impeller.

Normally, the pump fabricator does not write up the operating and startup/shutdown instructions for the pump. This is the Owner/operator's job, since he is in control of the application and the need. The pump fabricator furnishes "general" instructions, but these should never be applied literally to a specific application. The fabricator normally knows little, if anything, about how you are going to use his product. Of course, the fabricator can help you in the application - if you share all your basic data and knowledge about the application with him. This is information over and above what is normally furnished in the Specification Sheet. I would certainly use the fabricator as a consultant in special cases like troublesome fluids, difficult NPSHa situations, automatic prime and startup applications, non-newtonian fluids, etc. This is just smart engineering that is available to you if you apply common sense.

I hope this experience helps.



Art Montemayor
Spring, TX

 

[jay165]

Milk:

You normally start a centrifugal pump with discharge valve 10 to 25% open, then open further as pressure builds.

A wide open discharge valve could cause over-amping if there is not enough downstream resistance.

 

[uncleal52]

You should never dead-head the pump. This causes extreme wear on the pump since its trying force fluid out but it has no where to go... if you want to get a header up to pressure its better to recirc the flow on bypass back to tank and throttle a bypass valve to get your pressure.

 

[vpl]

Sometimes the discharge valve is closed when the pump receives its start signal to avoid column-rejoining water hammers. However, usually the discharge valve opening is slaved off the pump start so that the valve starts to open as the pump comes up to speed, resulting in the situation that jay165 recommends (that the pump starts with the valve only 10-15% open).

Sometimes you have to ask the question. There may be a reason why your company starts the pump with the valves closed. And you won't know the specifics until you ask. And being new, it gives you the "out" to ask, especially if you start with a disclaimer, "Being new here, I want to know why... so that I can learn."

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[Milkboy]

Thanks All
Hopefully I can use my new found knowledge
this Weds at the next meeting and ask
why we start up with closed discharge

From what I have read I am thinking that
maybe there is a recirc to suction, as per Montemayor, as it seem they have done this
before

Thanks

Milkboy

-
Milkboy

 

[quark]

Why can't you go with a balanced mechanical seal?

Eng-Tips.com : Solving your problems before you get them.

 

[jet1749]

Generally speaking for pump start up reasons, the initial discharge valve position is related to the type of impeller fitted to the pump. A simple radial vane or francis vane impeller draws low amps at dead head, so start pump with discharge closed. A mixed flow or axial flow impeller draws high amps at dead head, so start with discharge open. In both instances this prevents overloading of the motor during start up. It is possible to start a pump with a radial vane impeller with the discharge open providing you have a non return valve at the pump discharge, and the header or manifold is at pressure. The seal chamber pressure at dead head depends on the type of impeller fitted. An enclosed impeller with front and rear wear rings, and that also features balance holes between the rear wear ring and impeller eye, will generally give a seal chamber pressure of slightly above suction pressure. A semi open impeller, or an enclosed impeller with only a front wear ring and no balance holes, will subject the seal chamber at dead head condition to a pressure of around 60% of dead head pressure. Although as previous posts have mentioned you may have bypass or reirc lines that bleed off some pressure with the dicharge valve shut at start up, this is probably due to local operational conditions. Best advice is to follow the pump manufacturers recommendation, if in doubt or just out of interest, look at your pump curve at the absorbed HP curve at dead head. A low draw will indicate starting with valve shut and a high draw will indicate starting with valve open.
Best of luck
John

 

[rmw]

Milkboy,

Go read thread 237-90230 to find out all you wanted to know about pump motor starting current, etc, and more. Just remember that you want a minimum amount of flow at start up to prevent the pump from thrusting at a dead head condition, hence the recommendation of the discharge valve to be just a little open.


rmw

 

[tgreif]

I'm not sure why everybody is saying to not start the pump against a closed discharge valve. It is common to start centrifugal pumps against a closed discharge valve, especially large pumps. Take a large water pumping station for instance, where multiple pumps are piped in parallel to discharge into a large distribution system. These pump can't use mechanical pump check valves because they're too large and they ould allow the flow to accelerate too rapidly, causing a surge problem. They use power actuated discharge valves. The pump starts against the closed valve, then the valve is slowly opened to accelerate the water slowly into the system, avoiding water hammer or surge. If you open the discharge valve first, you have reverse flow through the pump and you'll burn up the motor trying to start it against a reverse running pump.

 

[Montemayor]

tgreif:

Everybody is not saying that a centrifugal pump should not be started against a closed discharge valve. What I wrote -and I repeat in the following - is: "A pump's proper and recommended startup depends on the type of pump, the fluid, the pump's configuration and piping, the process downstream, the flow and pressure controls installed, and the hazards involved (if any). There may be other factors I left out, but those are just some of the main ones."

Without knowing the specifics of the application, or having all the detailed basic data, no professional engineer can make a definitive recommendation. I agree that it is common in some applications to start the pump against a closed discharge, but you also know that this is not the rule 100% of the time. There are situations where the pump suction is not flooded and the pump must be primed before starting up. Under these conditions it would be folly and totally bad engineering to attempt a startup without first evacuating trapped air and priming the pump. One cannot simply assume that a pump will be primed, flooded and prepared to startup without first checking all the basic data.

It is not true that if you open the discharge valve first, you have reverse flow through the pump and you'll burn up the motor trying to start it against a reverse running pump. A customary check valve in the discharge prevents reverse flow; this is normally done in countless process applications in industry with appropriate valving and instrumentation - both manually and automatically.

Since the original post does not furnish detailed and complete basic data, we cannot furnish detailed and complete recommendations on the pump's startup. To do so would be to advise the user on a situation that we are ignorant of and one that may cause repercussions. Please be aware that this response is not a critique against you or your statements; I want to make sure that the user, Milkboy, understands that we are working with very little basic data and he alone holds the key to the specific needs and safety of his application.

Art Montemayor
Spring, TX

 

[tgreif]

No offense taken Art. However, I've seen very few instances of powered discharge valves in series with conventional check valves. If you have a check valve, why would you need to open and close a discharge valve, other than surge avoidance. If it is for surge avoidance, we're probably talking about some pretty big pumps. I work in the water and wastewater industry where valve sizes 24" and larger are common. Check valves in series with power actuated discharge valves are cost prohibitive at those sizes. There WILL be discharge isolation valves in any case to allow the check valve andpump to be isolated for maintenance, but I don't think that's the type valve we're talking about here.

Again, not trying to start and argument, but in my 30 years in the business, I've just not seen that type of series valve arrangement much.

 

[jproj]

Art:

I don't think tgreif was referring to you when he said, "everybody is saying to not start the pump against a closed discharge valve". Rather, I believe he was referring to uncleal52 and DLANDISSR.

Milkboy:

The pump startup procedure is really dependent on your process and piping (as noted by other posters). For EVERY installation we supply, we tell our customer to start the pump with the downstream valve closed. All of our systems have a recirculation line though, and the pump is never technically in a deadhead condition. If you don't have a recirculation line (which diverts the pumped liquid back to the source, or suction), then you want to open the downstream valve a fraction. Starting the pump with the valve wide open is just asking for all kinds of trouble! Besides, even if you do start the pump in a deadhead condition, the pump is not going to burn up in a matter of seconds. You will have plenty of time to slowly open the downstream valve before you damage your pump.

I suggest you contact your pump supplier instead of taking the word of strangers (who may or may not be qualified pump engineers). They will be able to discuss the recommended pump start-up procedures with you.

Good Luck

jproj

 

[stanier]

I suggest that not only contacting the pump manufacturer but also the system designer. The main criteria for starting centrifugal pumps against closed valves is to avoid waterhammer.

The system designer is the one with all the necessary information in respect of how a pump start will impact on the remainder of the system. That not only includes the hydraulics but as others have mentioned the electrical system as well.

If the system designer is not available then one has to get all the information that a competent system designer would have had and has generated before a decision can be made to alter an operating procedure. If you want to change a procedure then you have to demonstrate that there will be no ill effects to the system.

If you dont get yourself a good lawyer!

 

[longeron]

We've really beaten the pump part of the question to death. There is a second part of the question.... "how do I ensure the seal will be O.K.?"

Part of the answer is in the Start-up instructions you've quoted- Vent the pump. This is really important. Vent the pump AND the seal chamber. If you're using a tandem or double seal- be sure the outboard seal chamber (the volume between the two seals) is liquid full as well.

Often the old lantern ring connection is plugged when a mechanical seal is installed. If the process fluid isn't nasty, have your operator/millwright crack that connection open as the pump is inventoried. If the fluid is nasty, it might be possible to pipe the lantern ring connection to a drain.

If that sounds like too much work, or if you're worried that someone will leave the valve from the seal chamber open so you can't keep prime, there is another option. API Plan 01. (Yes- yet another flush plan to memorize) Plan 01 is Internal Recirculation From Discharge. Or better put, 01 is a Plan 11 performed with an 1/8" hole drilled at an angle upwards from the seal chamber to the volute. You'll want to do this at the top of the seal chamber, possibly through the throat bushing. (Pumps and Systems Magazine has an article on this in their archive section see

http://www.pump-zone.com)

Now for those tandem and double seals. The easiest way to be sure that the secondary seal is properly inventoried is to ask your seal vendor to put the Buffer/Barrier Out connection at the top of the Gland Plate. This promotes the same "automatic" venting as the Plan 01. The air in the seal is easily displaced upwards and out of the seal, through your flush tubing and into your reservoir. There are those who believe that tangential seal flush connections for secondary seals are essential. They have a point. The ammount of windage generated by most pumping rings is minimal, the secondary seal relies upon thermal cycling/siphoning to circulate the buffer/barrier fluid. Tangential connections decrease the head losses of standard connections. With these, I suggest that you leave the return tubing disconnected from the reservoir as it't filled. That way you can see air bubbles escaping, when they stop- connect the tubing and continue filling the reservoir.

There are other Flush Plans that deserve consideration here too. Plan 13 Recirculation to Suction: Best used on vertical pumps- automatically vents the seal chamber. Plan 14 Combination of 11 and 13: Again, automatically vents. Plan 23 Circulation through Heat Exchanger: This plan requires more thought. Here we circulate the fluid in the seal chamber again and again through a heat exchanger. The idea is to cool the same fluid instead of new hot fluid from discharge. You absolutely need a High Point Bleed with this kind of flush plan. Vapor in the tubing will kill the circulation, in turn killing your seal.

 

[cotes7]

You should never dead head start a pump.