How to design hydraulic system and size cooler

[Nathanwest58]

Hi all,

I've started working recently in a new job as an engineer in a lab with a bunch of physicists. A job they've given me is to size a water/glycol chiller (and basically design the hydraulic system i.e. fittings, tubing etc) to cool a bunch of equipment they've got in a few insulated boxes. The physicists have given me heat loads and acceptable temperature ranges for all of the individual things they want kept cool. All of the equipment and boxes are to be used in a small shed that will get pretty warm in direct sun, and they want all the equipment kept cool, well below ambient. I think I know what to do (haven't really done this sort of thing before), but would like to check with others before I go too far.

Questions I have:
A. Are the steps that I'm outlining below sort of on the right track? (Points 1 and 2 below)
B. How do I check if the chiller can actually pump the fluid through the system, overcoming the friction of all the tubes and fittings? (Point 3 below)

These are the steps as I see them:

1. calculate flow rates required to remove heat generated by equipment in individual boxes Q=ṁcp∆T,
2. sum all of these heat removal values, apply a safety factor and check that this sum is lower than the cooling power of the chiller,
3. specify all the tubing, fittings etc. Find the pressure drop/head loss across the entire system and then compare to the chiller's spec'd flow rate?

Any other tips or tricks for this sort of work?

Thanks in advance,

Nathan

 

[jrmichler]

The chiller will deliver water at a temperature.
Given that temperature, calculate the flow rate AND pressure drop for each piece of equipment.
The pressure drop includes line loss for the tubing and fittings that supply each piece of equipment.
The chiller pump must meet the total flow and worst case pressure drop.
Each piece of equipment needs its own flow control valve so that equipment with less pressure drop does not steal all the water. These can probably be hand valves.

All of the equipment and boxes are to be used in a small shed that will get pretty warm in direct sun, and they want all the equipment kept cool, well below ambient.

Insulate that shed. If not, the cooling requirement for the equipment will be increased by the temperature difference between the air in the shed and the desired equipment temperature. Or did the physicists already include that? You need to check this very carefully. The cost of insulating the shed is part of the job, not a separate project.

Depending on how many pieces of equipment, a job like this can be an exercise in keeping track of all the flow rates and pressure drops. Documentation is critical because you will be coming back to make changes in a year or two, long after you forgot what you did. I like to use a word document with heat loads, flow rates, and pressure drops summarized in tables. You do not need to include all calculations, just the results. Especially document the need for, and recommended amount of, insulation for the shed.

 

[erobz]

So you are running an evaporative cooler? A basic sketch of the system with relevant data would be helpful. You don't have to share the values, but some sense of the working model would be good.

 

[Nathanwest58]

The chiller pump must meet the total flow and worst case pressure drop.

Yes the pump meeting the total flow needs makes sense to me. How do I know if the pump is going to overcome the system pressure drop? I've seen things like 'supply pressure' or 'forward operating pressure' specified on chiller datasheets, do I need to make sure that supply pressure is greater than the pressure drop induced by the friction of the system?

Each piece of equipment needs its own flow control valve so that equipment with less pressure drop does not steal all the water.

Good tip, thanks. I'm guessing the equipment with less pressure drops steals all the water because the resistance to flow is lower than other equipment with higher pressure drop?

Insulate that shed.

Not an option unfortunately. This system is serving as a test bed for hot external environments, so it's part of the operation profile. The individual boxes will be insulated. One thing I am curious about, do I need to cool the air inside each box as well, or is it enough to simply cool the equipment inside each box? That part is confusing me.

Depending on how many pieces of equipment, a job like this can be an exercise in keeping track of all the flow rates and pressure drops. Documentation is critical because you will be coming back to make changes in a year or two, long after you forgot what you did. I like to use a word document with heat loads, flow rates, and pressure drops summarized in tables. You do not need to include all calculations, just the results. Especially document the need for, and recommended amount of, insulation for the shed.

I will absolutely do this, thank you for the suggestion.

So you are running an evaporative cooler? A basic sketch of the system with relevant data would be helpful. You don't have to share the values, but some sense of the working model would be good.

Most likely a recirculating chiller. Quick sketch of system below. Three boxes each containing a piece of equipment. I think I've drawn the symbol for a flow controller before each box correctly. I've just drawn a single line from the manifold to each box, instead of supply and return.

Thank you both for your help so far.

Nathan

 

[Lnewqban]

Please, see:
https://ctashrae.org/images/meeting/101112/hydronic_systems___dinner_session_presentation.pdf

https://www.literature.mestek.com/dms/SpacePak/WWNL-3-15.pdf

https://www.col-int.com/

 

[jrmichler]

How do I know if the pump is going to overcome the system pressure drop? I've seen things like 'supply pressure' or 'forward operating pressure' specified on chiller datasheets, do I need to make sure that supply pressure is greater than the pressure drop induced by the friction of the system?

Simple answer: Yes. Better answer: The chiller pump has a pump curve (search the term) that shows the relationship between flow rate and pressure. The pressure and flow listed on the datasheet is one point on that curve. Your system has a hydraulic system curve (search that term) that shows the relationship between flow rate through the system and the pressure to make it flow at that rate. The actual flow rate is the intersection between those two curves.

I'm guessing the equipment with less pressure drops steals all the water because the resistance to flow is lower than other equipment with higher pressure drop?

Yes, or more correctly: The system with less pressure drop takes more water than it needs, the total flow rate increases, which causes the total pressure to decrease (see above about system and pump curves), which causes insufficient flow to the equipment with high pressure drop.

This system is serving as a test bed for hot external environments, so it's part of the operation profile. The individual boxes will be insulated. One thing I am curious about, do I need to cool the air inside each box as well, or is it enough to simply cool the equipment inside each box?

If the purpose is to test equipment cooling in hot ambient temperatures, then I THINK you just cool the equipment. But that's just my opinion. This is a question for the physicists because it's their equipment.

 

[erobz]

Yes the pump meeting the total flow needs makes sense to me. How do I know if the pump is going to overcome the system pressure drop? I've seen things like 'supply pressure' or 'forward operating pressure' specified on chiller datasheets, do I need to make sure that supply pressure is greater than the pressure drop induced by the friction of the system?

The chiller pump can supply refrigerant at some differential pressure that is a function of the volumetric flow rate of the system. You would plot the system curve as a function of volumetric flowrate along with that pump curve and the point of intersection is the theoretical flow rate in the system. So, if you need each load to have flowrates A, B,C then your pump in the unit needs to supply A+B+C at whatever the differential pressure of the system is at that flowrate. Also, optimally it should do so in the vicinity of its peak efficiency.

Good tip, thanks. I'm guessing the equipment with less pressure drops steals all the water because the resistance to flow is lower than other equipment with higher pressure drop?

Is it water being circulated or refrigerant? Are you designing the heat exchangers from scratch or purchasing?

Not an option unfortunately. This system is serving as a test bed for hot external environments, so it's part of the operation profile. The individual boxes will be insulated. One thing I am curious about, do I need to cool the air inside each box as well, or is it enough to simply cool the equipment inside each box? That part is confusing me.

I think you want to cool the equipment inside directly. Going through the air surrounding the equipment just adds another thermal insulating layer for heat to pass through before leaving the system. But there could be reasons for indirect cooling to consider...