bypass decoupler pipe in primary/secondary chilled water system 1

[nabilbasha]

Hi everybody,
In the attached typical primary / secondary chilled water system, you will notice that there are two (2) bypass pipes. This is the first time I come across such arrangement. i.e. 2 bypass. Also the location is wrong. In addition, having visited the job site, one of the bypass has long and smaller pipe size.

Question: Is it necessary to have a bypass? Note: the system consist of 3 chillers, 3 primary constant speed pumps, and 3 secondary pumps with VFD.

If it is necessary, then where is the best location noting the strange current piping layout.

Thank you for your assistance.

 

[Drazen]

it's a bit strange, but it can work, can it?

decoupling lines present points of zero differential pressure and you must not alter it if you don't plan to resize pumps.

in the marked schematics, green, blue and red lines are covered by secondary pumps, while primary pumps fight resistance of yellow lines only.

the arrow on upper bypass can change direction depending on flow variations in secondary system.

so the answer is - it is absolutely necessary to instal bypass lines per design.

 

[tys90]

What's that in the lower left hand corner of the drawing? It appears that it's a tee in the primary loop going to the secondary return, making it the third bypass.

It is a bit odd, I can't tell why they would need all those bypasses without speaking to the designer (something you may want to do). The bypass in the lower right-center of the drawing appears to allow the primary pumps to bypass the chillers if needed. The one in the upper part of the drawing is probably the main bypass. Like Drazen said, it would work fine so I would install it per the drawings, you never know why they might need one of them.

 

[nabilbasha]

Thank you TYS90 and DRAZEN for your contribution and willing to support.
The color lines I add it to make it easier to identify the flow diagram of supply and return.
I have attached actual photo of the bypass from job site.

DRAZEN: I am not quit sure the bypass on the top-lift corner is at points of zero differential pressure!
I am interested about your statement that the primary pumps fight resistance of yellow lines only. Please elaborate. If true then I must recalculate both primary & secondary pumps heads. Finally, I agree about the necessity of bypass but not sure of necessity of TWO (2) or THREE (3) bypass and/or its location.

TYS90: I am not sure what is shown on the lower left hand corner on the drawings but I shall check it on next job site visit and report back.

I have been called on to balance/optimize the water/hydronic system.

The situation is a bit complected at the moment. The owner selected, provided and installed the Chillers. The contractor provided & installed the pumps and piping network. There seem to be a dispute between them over why the return HIGH water temperature. I have been asked to optimize the system irrespective whose fault it is. i.e. balance the system to make the most out of the current chillers.

The first thing I thought to do is to check the system for any wrong installation (piping, valves, vents, bypass, etc), then take temperature, pressure, and flow measurement. Then adjust the primary pumps valves to fulfill the chiller requirement. The secondary pumps are on VFD controlled by Pressure Differential sensor located next to the pumps on the supply & return pipes. Both pumps and chillers controlled by the BMS.

Following are main system data:-
Required (based on HAP software calculated load): 750 ton at design ambient 106 Deg. F
Delta-T: 44 / 60 F (16F).
Chillers: 3 x 387 TR at design ambient 115 Deg. F (580 GPM water flow rate).
Primary Pumps: 3 x 15hp 570 GPM @ 50 Feet - constant speed.
Secondary Pumps: 3 x 75hp 570 GPM @ 180 Feet - variable speed.
System is 2 way valve with preset dynamic balancing valve.

I would very much appreciate your views.

Thank you

 

[nabilbasha]

Sorry... I meant to load this photo instead of the other on previous post.

 

[KiwiMace]

I would say the bypass at the bottom of the drawing was intended for flushing, low flow or commissioning reasons and it would normally remain closed. If this valve is open, you will short cct the chillers and get a high return temp.

The bypass near the primary pumps is the decoupling bypass, but personally I think it is misplaced and undersized. It should be an 10 or 12" line connecting from the (red) Secondary CHWR before the primary pumps to the (green) primary CHWS downstream of all the chillers near the 45deg bend along the wall.

 

[nabilbasha]

KiwiMace: I agree with your comment but I think it is a long run bypass hence pressure drop may be too much.
The question then, can I get away without the bypass?
If I have to use it what would be the maximum acceptable PD ?

 

[Drazen]

nabilbasha, bypass is not at the point of zero differential pressure - it creates zero differential pressure.


pressure drop through bypass should be made equal to the pressure drop of the whole network being bypassed, so you would likely need to throttle it even if pipe diameter looks too small to you.

it is crucial element of your hydraulic scheme, so you cannot remove it (if you decide so nevertheless, let us know what happened).

rectifying your problem with new chillers is not an easy task.

you should first check supply temperature - did it rise as well, or only return temperature rose? that could be sign of too low flow in secondary network.

 

[KiwiMace]

The bypass should offer a low pressure direct return for the primary loop. The point is to decouple any influence of the primary loop from what is going on in the secondary loop and vice versa. If there is a similar pressure drop to your connected network as Drazen has suggested then this would transmit the pressure influence between the two networks. To establish your bypass flow and sizing, you need to look at your flow balance for staging chillers in your primary side and the operational aspects of load in the secondary. At a minimum this flow will be one chiller evap flow, but if you routinely stage a couple of chillers or your load is prone to large fluctuations then this may be more. Obtain your flow, size with a low pressure drop like 50 Pa/m or thereabouts.

 

[Noway2]

Nabilbasha, it sounds to me like is trying to operate a system designed as a primary-secondary as a variable primary. With some care, and VFD based pumps, this can be done, at least to some degree. The variable speed pumps can be set to track and maintain the system DP. The thing is that sequencing of and flow control to the constant speed chillers can be really tricky (see the following for a discussion of some of the issues:

http://www.trane.com/commercial/library/vol31_4/).

If you are getting high return temps, which is what I read in your OP, my initial guess would be that you have too much of an imbalance between your distribution pumps and your return pumps.

 

[DRWeig]

A check valve in the bypass may be the ticket. I have attached a couple of good articles.

Engineering Report Variable Pumping CHW

Chillers - Variable Flow Systems

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[ChasBean1]

nabilbasha, this system has the exact issues you identified in the two locations you circled. The pipes in these two locations need to be line sized, not the half diameter reduction indicated. Increase these pipe sizes first before any other actions...