District Cooling - criteria for a unit pressure drop (Pa/m) and speed

[BIPVguy]

Dear all,

I would kindly ask for your assistance please regarding sizing pipelines for district cooling system, please.

I have been looking for criteria that have to be met in order to size district cooling systems with parameters supply 7C and return 12C. Terrain configuration allows both downstream and upstream flow with total lenght of 4 km.

From ASHRAE 2005 Habdbook, I understand that this question is not that straightforward. While for larger diameters criterion is water velocity, for smaller pipe diamteres is unit pressure drop. Generally, water velocities above 3.0 m/s are undesirable.

My concerns. ....

1) Generally, what unit pressure drop should be selected for calculations (Pa/m)?
2) What is the range for small pipe diamaters and that of big pipeline diameters.
3) In urban environment it makes sense to consider noise in pipes due to high velocities but is it of big importance if pipelines are installed on "virgin" land.

Some extracts from my calculations and what would you select taken into consideration the cost of pipes and above criteria.
Parameters: 7/12C, NP6 bar, Absolute roughness 0.02mm
Results:
38 320kW....DN800...w=3.57m/s....R=42Pa/m
33 20kW...DN700...w=4.68m/s....R=80.5Pa/m
6 000kW...DN350...w=3.0m/s....R=88.1Pa/m
500kW...DN125...w=1.7m/s....R=106.3Pa/m

 

[TBP]

You might check out the International District Energy Association. Many of their members run district chilled water systems, as well as steam and/or hot water.

I doubt very much that noise will be a factor. The pressure drop in city water mains will typically be much higher than you'd design for for a chilled water system, and I'm not aware of anybody being able to detect any noise from the water mains, let alone at level that would trigger complaints. Inside buildings of course, velocity noise is another matter.

Velocity, pressure drop, and flow are all linked. If you run a higher pressure drop, the velocity increases, but you can use smaller lines for the same flow. Now you're into pumping cost considerations. The higher delta-P, the more horsepower you'll need. This means both a higher first cost, and an ongoing operational cost.

Seriously consider checking out the IDEA people. I haven't worked in that business for a number of years, but I'd bet that the design info and actual running plant data available will be well worth the price of joining.

 

[marcoh]

First of all I've never designed a district cooling system before but 7/12 degrees seems to be a small dT? Assuming there will be heat exchangers etc to individual buildings I would have assumed a supply temperature of 4 or 5 degrees to enable a supply temperature to the AHU's of around 6 to enable good coil selections for dehumidification (might depends somewhat on where in the world the project is)?

 

[HVAC68]

You are right marcoh. The district cooling systems that I have come across, specify a supply water temperature of 4 to 5 deg C and a return water temperature of around 15 deg C (around 10 deg C delta T), with plate type heat exchangers for individual load groups/buildings.



HVAC68

 

[quark]

There seems to be a mismatch with the data. The biggest pipesize corresponds to the lowest power consumption, velocity and pressure drop. Further, for a constant flowrate, the increase in power consumption doesn't seem to be inline with increase in pressure drop (assuming pump efficiency constant).

3m/s is not hard and fast. You can go for higher velocities if the pipesize is big. I like the suggestion by TBP. In the absence of any help, I would just select the break even point of pipe cost vs power consumption.

You should consider heat gained by chilled water in the 4km piping. Otherwise, if the dT at chilled water generation system is 5C, you may end up with some trouble.

 

[ChrisConley]

Another thought to add is expansion. I've yet to work on a district cooling system that did not add several hundred more tons of cooling than was originally planned.

Upsizing the pipes to allow for future flow will reduce pumping costs as well.

 

[BIPVguy]

Thanks for the answers.

quark.....

In case of pre-insulated pipelines, heat transfer through casing and insulation is very low (cca. 3%) and thus; almost negligible.

Thanks Sasa