System Efficiency at Partial Load

[IBechir]

Greetings

I encountered this chilled water system that I believe is inefficient. I have attached a schematic. Your thoughts are highly appreciated.

Brief System Description:
The 40 storey building chilled water supply comes from a district cooling provider. The lower and upper halves of the building are separated into two closed loop circuits. Each circuit has a heat exchanger and variable pumping stations. The heat exchangers and pumps for the first circuit are in the 1st basement while the ones for the second circuit are in the 17th floor serving the upper half only.

The problem is that the upper circuit is served by the lower circuit; i.e., the lower closed loop circuit has two functions: 1) to distribute chilled the lower half of the building and 2) to serve as a 'primary' loop for the second circuit (the upper half of the building).

Say if the building cooling demand is in the upper floors, both circuit pumps will be loaded even if the lower floors are unoccupied. Say, in another case, the cooling demand is in the lower floors, water will be circuilated through the heat exchanger of the upper circuit which is connected in series and without a 3-way valve to bypass it (also an energy waster).

I think it will be much more efficient if the pumps and heat exchangers of the upper half are also moved to the 1st basement so there will be two separate circuits. There will be an additional cost for piping and a larger capacity pump set but there will also be savings in the form of a smaller heat exchanger (that used to serve both circuits/loops) and better energy usage.

Much obliged for any information.

 

[Drazen]

separation is made to reduce overly high static pressure.

issues of inefficiency are more general issues of fixed flow system than something completely related to this separation.

upper fan coils are seemingly designed to somewhat different temperature regime.

i can imagine that this solution was made to somewhat simplify construction issues for risers. changes you propose would require major refurbishment, and taking into account the issues mentioned - reducing static pressure, achieving constructability of risers.

 

[ChasBean1]

I agree with you in that I might have piped the district CHW directly to each HX.

However if the lower HX is sized to accommodate the lower loads as well as the upper HX, you can look at the upper HX as another load off the lower system. If the lower system varies pump speed to maintain a loop DP, a reduced lower building load will have the pumps ramp down and only produce enough to keep the upper HX happy; if the lower building load is low and the upper is high, the lower pumps will modulate as needed to keep the upper system satisfied…

In general, I don’t see a problem with this configuration as long as the lower HX is sized to accommodate the full building load.

 

[IBechir]

Great. Last question: would you use a three-way valve to bypass the upper heat exchanger, when the load is low, or would you use a modulating two-way?

Thank you

 

[317069]

What about the flow in district cooling provider connection. is it constant or variable ?

 

[IBechir]

Variable.

 

[Drazen]

chas, on 40-floor building connecting upper floors to the bottom means additional 20 x say 3 m floor height = 60 m = appraoximately 6 bar. That together with 6 bar for upper floor makes 12 bar of static pressure, something that vast majority of HVAC equipment cannot handle.


The other issue is 120 m long riser, which makes outrageous demands on construction. Halving riser length halves thermal expansion and all construction loads related to it. 60 m is already long enough, requires special solution for fixed points and hanging.

 

[ChasBean1]

By my earlier post, the district would supply HXs on Floors B1 and 17. 125# would be the max rating needed for each system. I would suggest the 250# rating option for all devices in this high-rise.

IBechir, two way. Have an end-of-loop minimum flow device (controlled or by circuit setter).

 

[IBechir]

I just want to thank all those that took the time to reply to this post. You're great!

 

[317069]

but still now answer about if we move the upper exchanger to the lower exchanger level would this will save energy or not?
and you said the provider connection is variable, what control it?

 

[IBechir]

Heat exchanger levels remain as they are (one on B1 and one on floor 17) for reasons outlined in the posts above. There is no clear advantage in separating the upper-floors and lower-floors loops in terms of energy usage for distribution. Plus separating the loops will cost more piping and floor space. Plus, project is past design phase big changes are difficult to incorporate. Flow is controlled by a two-way valve that operates depending on temperature difference between supply and return.

 

[irfansiddiqui]

the design is fine. if you change the system to ground level (for whole building) it will cost a-lot more for the client. go for the same design. I would recommend 3 way by-pas valve.

 

[foreigner]

I agree with Drazen. I think that the upper HX was installed in the first place only to separate two circuits hydraulically. I also agree with ChasBean1 that a variable flow system will be quite feasible. I'd be surprised if the lower HX didn't have the capacity for the whole building.