primary/secondary - Secondary pump location with respect to the bridge.

[BronYrAur]

I normally put my secondary pump, air separator, and expansion tank downstream of the primary/secondary de-coupler bridge. For reasons of significant piping simplification in the field, I want to put them upstream on a job. I know that will put the bridge (and chillers) on the discharge side of the secondary pump and raise the pressure in that part of the piping, but is that a problem? The closely spaced tees with both have the elevated pressure.

Any reason why I should not do it that way? See attached. Thank you.

 

[RandomUserName]

Secondary pumps should always discharge into secondary circuits away from common piping. Doing so ensures an increase in secondary circuit pressure over that established in the common piping by the primary pump. The common piping can be considered as the compression tank “No pressure change point” for the secondary circuit. It is consequently generally wrong to pump into the common piping from the secondary circuit because of a decrease in secondary circuit static pressure.

Looks like your primary system is setup with coupled pumps. Coupled systems should only be used when all the equipment in the system is the same capacity (chillers, cooling towers, pumps, etc.) If this is not the case, consider a headered system with control valves.

Also, air separators and expansion tanks are typically placed in the primary loop on the suction side of the primary pumps in a primary/secondary system. An alternate location in primary/secondary systems with a single secondary circuit may be the suction side of the secondary pumps; as you have indicated "where you normally put it".

EDIT: Attaching URL for BELL & Gossetts Primary / Secondary Pumping Applications Guide:

http://documentlibrary.xylemappliedwater.com/wp-content/blogs.dir/22/files/2013/02/TEH-775A.pdf

See page 3.

 

[BronYrAur]

RandomUserName, can you expound on your first paragraph? Why would the primary pumps dictate the pressure in the common pipe over the secondary if I am pumping those secondary pumps towards the common pipe?

 

[Drazen]

I read the statement in original document, and it is plainly wrong to say that position of pump determines static pressure.

Static pressure by definition is the one not dependent on pump. General point, though, stays - if you put pump as you showed, total pressure in most of secondary will be lower than total pressure in common point. as secondary has much more joints than primary, and they are spread all over the building, and those joints are generally not well-suited for underpressure - you will have many points of possible air suction.

 

[EnergyProfessional]

I usually use a hydraulic separator with air-dirt removal built in. It simplifies piping and fixes the pump locations.

 

[BronYrAur]

Again, to reiterate my question, I have always piped systems with the secondary pumps downstream of the common piping. That way, the suction sides of primary pumps, secondary pumps, and the common bridge are all approximately the same pressure where the expansion tank is located.

I am certainly not advocating that putting the secondary pumps upstream is better, but I just want to rule out problems.

In either situation for this situation, my expansion tank will be on the suction side of the secondary pumps. So pressure in the secondary system will be almost the same either way, won't it? The B&G article does not takes this into account when they indicate that it is wrong to pump toward the common piping. In the B&G article, the expansion tank is on the primary side of the common pipe, so you would be pumping toward it. So I agree that it would be wrong in that case.

So back to my question, provided the expansion tank is on the suction side of the secondary pumps (wherever they a placed), is there any reason why I should not pump toward the common pipe? It will substantially raise the pressure on the common pipe and the overall pressure in the chiller loop, but why is that a problem? Or will the primary pumps add head that must be overcome by the secondary?

I'm confusing myself the more I think about it.

 

[LittleInch]

As you've drawn it I can't see any particular reason why the location of the pump will make the overall system not work.

As you say, the key issue is that it raises the pressure in the common pipework and the chillers compared to your normal scheme. That becomes an issue if the pressure rating of the chiller loop is not as high as the secondary loop discharge pressure. In general the lower the pressure the reduction in stress and less chance of a leak or failure in the chillers.

The primary pumps shouldn't add any head, The end connections essentially see the same pressure, with the purpose of the pump simply to overcome frictional losses in the pipework from the common header and losses in the chiller.

Depending on flows from primary or secondary pumps, the flow in the connecting bridge could be in either direction or no flow if the flows are exactly matched, but the pressure comes from the secondary pump

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Compositepro]

Your suggested change will cause increase in pressure in your primary loop when the secondary loop pump is running. The pressure increase will equal the pressure drop in the secondary flow loop. In most cases I do not see that this would be a problem.

 

[urgross]

A backflow in the bridge is generally recommended.

 

[lukaiENG]

The main reason I would not do this is controlability. If you think about it your 2 sets of pumps are in series with very little pipework/fixtures in-between. You're going to get all sorts of pressure dynamics which will cause hunting of the pumps as they try to maintain their (I assume) pressure differential set points.
The typical pumping arrangement resolves this issue by placing the load and chillers (read: inertia) between the pumps which dampens down the pressure fluctuations seen at the sensors as the system adjusts it's operating point.

 

[Drazen]

you did not mention expansion tank issue before, but subject still remains relevant.

pressure level is determined by expansion vessel only, not by pumps in such system, so if you put it in the mentioned position, you have to raise expansion tank setting by calculated loop pressure drop, plus contingency, to make sure underpressure is avoided.