Baseline Chilled Water Loop

16 posts / 0 new
Last post

Everyone,

I have a question regarding the baseline chilled water loop.

ASHRAE 90.1 2007 Section G3.1.3.10 states that for large chilled water systems serving baseline systems 7 and 8 that are greater than 300 tons the loop ?shall be modeled as primary/secondary with variable speed drives on the secondary pumping loop.?

The following section, G3.1.3.11 further goes on to say that ?each chiller shall be modeled with separate?chilled water pumps interlocked to operate with the associated chiller.?

As clarification, does this mean three sets of pumps are required, or just two? Are the ?primary? loop pumps specified as the chiller pumps?

Two reasons why I am confused. Firstly, the Chapter 11 baseline chilled water loop is supposed to be simulated as primary-only variable flow, but still has the similar requirement for interlocked chiller pumps. That would be two sets of pumps. Adding a ?secondary? loop actually adds a third set of pumps.

The second reason why I am confused is that the Simulation Guidebook prepared for Pacific Gas and Electric (same affiliation as CoolTools) has a visual of how to simulate ?primary/secondary? loops on page 52, but they do not include chiller pumps in their image of the eQUEST component tree.

I can?t understand why three sets of pumps would be required. I supposed a follow up question may be if the 22 watts/gpm is inclusive of all three sets of pumps. Hopefully someone can clarify! I appreciate your help.

Michael J. Smith, EIT, LEED? AP

Smith, Michael's picture
Offline
Joined: 2011-09-30
Reputation: 0

Hi Michael,

I think I can answer your App G question.

First, in the context of 90.1, ?variable flow? does not necessarily mean variable speed. It just means riding the pump curve with a bypass to minimize pumping pressure drops through the loop during part-load operation.

App. G Chilled Water Modeling <=300 tons

1. Create a chilled water circulation loop and name it something like ?Pri CHW Loop.?

2. Create another chilled water circulation loop and name it something like ?Sec CHW Loop.?

a. Change the Loop Sub-type of this loop to Secondary and choose ?Pri CHW Loop? as the primary loop.

3. Create a circulation pump and name it something like ?Sec CHW Pump.?

a. Set this pump as variable speed. I believe you are allowed to set the minimum speed to 50% for App. G. (someone here on the list can correct me on this if I?m wrong).

b. By default, the valve on this loop is three-way and will circulate water directly back to the primary loop as long as there are two-way valves at your terminal units (AHU coils, FCUs, etc).

c. Attach this pump to the Sec CHW Loop.

d. Change the loop head sensor location to ?at coils.?

4. Create a condenser water loop and CW pump and attach the pump to the loop.

a. Also create an open tower heat rejection device and attach it to your condenser water loop.

5. Create 1 primary chilled water pump per chiller. They won?t be attached to anything yet and will be at the bottom of the component tree.

a. The default for a pump is ?One-Speed.? This is fine and the pump will act as a constant volume pump with the three-way valve on the secondary loop.

6. Create a water-cooled chiller and choose the ?Pri CHW Loop 1? and ?water-cooled.?

a. Since there is no pump attached to the primary loop, you will be required to select a pump for the chiller.

b. You will also be required to select the Condenser water loop

7. Now what you have to do is size your pumps so the total pump energy is 22 W/pm. You will want to do something like 4-5 W/gpm for each primary pump and use the remaining for the secondary pump.

This system is primary/secondary with constant flow through each chiller and variable speed/variable flow through the secondary loop. Since each chiller has a pump directly attached to it, each primary pump will automatically cycle on and off with the chillers.

John T. Forester, P.E., LEED AP

jforester's picture
Offline
Joined: 2010-12-21
Reputation: 0

Fellow Modelers,

I?ve received a comment regarding my response last week on how to model CHW pumps for App. G. Specifically, item 7 (see below) and how the 22 W/gpm is defined for a primary secondary configuration. Section G3.1.3.10 can be interpreted (at least) two different ways.

First, Section G3.1.3.10 Chilled-Water Pumps (Systems 7 and 8)
?The baseline design pump power shall be 22 W/gpm. Chilled-water systems with a cooling capacity of 300 tons or more shall be modeled as primary/secondary systems with variable-speed drives on the secondary pumping loop. Chilled-water pumps in systems serving less than 300 tons cooling capacity shall be modeled as primary/secondary systems with secondary pump riding the pump curve.?

The question:
Should each primary and secondary pump be sized for 22 W/gpm OR should this ?design pump power? be divided up (in some way) between the primary and secondary pumps?

Position A:
If the 22 W/gpm should be the TOTAL pumping power, there is no guidance on how to divvy up this requirement and the constant volume primary pumps could be sized for 20 W/gpm and the secondary (VSD or riding pump curve) can be allotted 2 W/gpm. This would maximize the pump energy in the baseline model.

Position B:
Sizing each primary pump and secondary pump to 22 W/gpm will over-estimate the baseline pump energy (VSD or riding pump curve) and this is not the intent of 90.1. The ?design pump power? of 22 W/gpm is the same requirement as Section 11 (ECB method) just with a different pumping configuration.

How do you interpret (and model) 90.1 App. G chilled-water pump power requirement of 22 W/gpm? (Your interpretation doesn?t necessarily have to be one of the positions above.)

Thanks,

John

John T. Forester, P.E., LEED AP

jforester's picture
Offline
Joined: 2010-12-21
Reputation: 0

I?ve seen different people provide different responses to this question, but I?ve always modeled EACH pump in a chilled water system as 22 W / gpm. Don?t prorate.

?The baseline design pump power shall be 22 W / gpm?. It can?t be any more clear than this. 90.1 tells you when you need a primary / secondary system, and when you can get away with primary only. In other words, they are telling you how many pumps you need to model. And they specifically list the pump power in terms of GPM, which is PER PUMP. Nowhere does it say to prorate.

James Hansen, PE, LEED AP

James Hansen's picture
Offline
Joined: 2011-09-30
Reputation: 200

I usually don?t respond to the information on this forum but I completely disagree with this latest e-mail.

The chilled water is supplied by a chiller not a pump. ?The baseline design pump power shall be 22 W / gpm? is for the GPM delivered by the chiller and the pumping system does not matter. The power is specified per GPM not per pump. If you use primary/secondary then the energy is divided between the pumps. We usually assume 6W/gpm for the primary pump and 16W/gpm for the secondary pump. If one uses 22W/gpm for each pump, then the total of 44W/gpm the pump energy is huge and the savings unrealistic.

R I C H A R D E L L I S O N
PE, CEM, CEA, CTAB, CBEP, LEED AP

R I C H A R D E L L I S O N
BEAP, BEP, BEMP, BESA, CBEP, CDSM, CEA,
CEM, CMVP, CTAB, DGCP, HBDP, LEED AP, PE
Manager of Energy and Simulation
Southland Industries
22340 Dresden Street, Suite 177
Dulles, VA 20166
Office: 703.834.5570
Direct: 703.

Richard-Ellison's picture
Offline
Joined: 2011-09-30
Reputation: 1

As I indicated earlier, everyone has their own opinion about this credit. But consider this:

The pressure drop across a cooling tower and condenser water piping system is roughly the same as the pressure drop thru the chiller evaporator and primary loop of a primary/secondary chilled water plant (on a per gpm basis once you correct for the higher flow rates of the condenser side). In other words, on a PER GPM basis, a typical water-cooled chilled water plant will have similar pump HP per GPM for the condenser water system and primary chilled water system. Not identical, but similar. I?ve got 10+ recent base-building designs that confirm this. 90.1 requires 19 W / gpm for a condenser water system, and you are saying that you only model 6 W / gpm for the primary chilled water pump? That is below realistic design values for a primary system, at least in my opinion.

Example: 1,000 ton chiller at 54/42 requiring 2,000 GPM. 6 W / gpm thru the primary loop would only use 12.0 KW, or a 15 hp motor frame. If you assume a chiller evaporator loss of 15 ft, and make only modest estimates of an additional 20 ft for piping losses, strainers, mixing valves, bypass valves, etc (35 ft total), you can try and select any pump you want but you?re going to need at least a 25 hp motor. And with most chilled water plants, even 35 ft seems low for the pressure drop across a primary loop. I don?t see how 6 W / gpm can accurately represent a chilled water primary loop even in the most efficient central plant designs.

Furthermore, if ASHRAE had wanted us to prorate the 22 W / gpm between the primary and secondary loops, wouldn?t they have instructed us to do that, especially since there is a huge difference in energy consumption depending on which of the two loops has the higher W / gpm? Have they instructed us to do that in an addendum?

I stand by my use of 22 W / gpm for each pump, and have never had a reviewer instruct me to do otherwise (which of course doesn?t mean that I?m not wrong).

James Hansen, PE, LEED AP

James Hansen's picture
Offline
Joined: 2011-09-30
Reputation: 200

James,

I see your argument that it is a challenge to design to 22 W/gpm. However, using your example of 2,000 gpm and 22 W/gpm, are you suggesting that a 44 kW (60-hp) motor is acceptable for a primary-only pump on a 1,000-ton chiller?

Section 11 states that the 22 W/gpm was chosen based on 75 ft of head and 65% combined pump/motor efficiency (the 90.1 User?s Guide actually says 60%) for a primary-only loop. This would include all system pressure drops. Again, this is not easy to design to. But at least the Section 11 example is clear on how much pump energy to include in the baseline design.

John

John T. Forester, P.E., LEED AP

jforester's picture
Offline
Joined: 2010-12-21
Reputation: 0

I raised the question initially with John offline. My interpretation of ?pump power shall be 22 W/gpm? is that each pump in a primary/secondary system should have this power. There is no wording that indicates that it should be otherwise, though I see Richard?s point that ?gpm? could be interpreted to mean CHW produced. My chilled water system design experience is limited, so I don?t have a sense of ?typical? CHW system pumping power. Using only the system that I am currently modeling for a project, the proposed primary pumps are 28 BHP and 953 gpm, or 21.6 W/gpm for the primary loop. The secondary CHW pump uses a 20 HP pump at 380 gpm. I don?t have the BHP, but using 16 BHP results in 31 W/gpm for the secondary loop. So, my sample size of one already uses more pump power than a baseline system with 22 W/gpm applied to each of the primary and secondary pumps.

William Bishop, PE, BEMP, LEED? AP

Bishop, Bill2's picture
Offline
Joined: 2011-09-30
Reputation: 0

In 90.1 Section 3.2 Definitions, "Pump System Power: the sum of the nominal
power demand of motors of all pumps that are required...."

This wording makes it clear that in a primary-secondary pumping system the sum
should be 22W/gpm for both pumps, not per pump.

Paul Diglio

Paul Diglio's picture
Offline
Joined: 2011-09-30
Reputation: 400

To be more precise, it says:

?Pump System Power: the sum of the nominal power demand (nameplate horsepower) of motors of all pumps that are required to operate at design conditions to supply fluid from the heating or cooling source to all heat transfer devices (e.g. coils, heat exchanger) and return it to the source?

The term is mainly used to determine the horsepower cutoff for certain requirements and exceptions in Chapter 6, although it has also been picked up in Chapter 11 and App G, even though the intent is slightly different.

This definition, in my opinion, is just as ambiguous as the other sections addressing this issue. Couldn?t you argue that with a secondary chilled water system, the cooling source is the primary loop? And thus there are two ?pump systems?, one to circulate primary chilled water (to a secondary loop heat exchanger or 3-way diverting valve) and one to deliver chilled water to the AC units?

I?m playing devil?s advocate to a certain point, but I still don?t see anything that convinces me 100% one way or the other.

James Hansen's picture
Offline
Joined: 2011-09-30
Reputation: 200

James,

I agree completely with Paul and John in their e-mails. The source of chilled water is the chiller (or chillers) itself. Not the primary loop. Really one must use common sense in interpreting these rules in that there is some ambiguity on how some are written.

The fan energy calculation in appendix G is the same things. If you read all the sections and past revs of 90.1 it is clear the fan energy given is the total of all fans. I had several, including large companies how write load software, that incorrectly attempted to apply that fan energy to each fan when the intent is clear to apply it to the total fan consumption in the baseline HVAC systems. The pump consumption appendix G is for the total pump consumption.

R I C H A R D E L L I S O N
BEAP, BEP, BEMP, BESA, CBEP, CDSM, CEA,
CEM, CMVP, CTAB, DGCP, HBDP, LEED AP, PE
Manager of Energy and Simulation
Southland Industries
22340 Dresden Street, Suite 177
Dulles, VA 20166
Office: 703.834.5570
Direct: 703.

Richard-Ellison's picture
Offline
Joined: 2011-09-30
Reputation: 1

I don?t disagree with anyone?s opinion, and I agree that there is ambiguity with how it?s written. I?ll agree that 22 W / gpm might make sense for a primary only system. But if I?m modeling a 1 million square foot building with 100 ft of pressure loss in the secondary piping system alone, there?s no way I?ll model the baseline chilled water plant at 22 W / gpm TOTAL, especially knowing that # is based on a total system head of 65?. Anyone that?s designed a decent size building (5+ stories, 200,000+ sq ft) with a primary / secondary chilled water loop knows that the total system pressure loss is always > 65 feet. I imagine that if ASHRAE wants to properly address this, they?ll revise the calculations similar to what happened with the fan power calculations in 2007. The W / gpm # should not be the same for primary only and primary / secondary systems, and it should change with building area, especially since they don?t allow modeling of piping losses.

James Hansen's picture
Offline
Joined: 2011-09-30
Reputation: 200

Paul,

Thanks for adding to the conversation. John and I were hoping more people would contribute. Like James, I think the wording is ambiguous enough that clarification is required from ASHRAE. I?ll add a few more comments.

* If Appendix G. used the wording ?pump system power?, rather than ?pump power?, I would agree that 22 W/gpm should be the sum of the power of all pumps in a system.
* Richard Ellison commented that this should be compared to the Appendix G fan power calculation, which describes the method to calculate total fan system power, not power for each fan in the system. However, the heading for 90.1-2007, G3.1.2.9 is ?System Fan Power?, whereas G3.1.3.10 is ?Chilled-Water Pumps (Systems 7 and 8)?. Again, the wording ?pump system? is not used in Appendix G. The heading ?System Fan Power? was changed from the 90.1-2004 wording of ?Supply Fan Power?, making it less ambiguous.
* CHW systems are only modeled for the baseline for buildings with 6 or more floors, or >150,000 ft^2, so we are not modeling small systems which would have less head loss. In Section 11, where the 22W/gpm requirement first appeared in earlier versions of ASHRAE Standard 90.1, there is no such size constraint, and systems are modeled as primary only, which presumably would require less pump power.
* My CHW system design experience is limited, but I?ll note that systems in the north are typically designed with high percentages of glycol due to freezing concerns, which increases the pumping power required.

Bill

Bishop, Bill2's picture
Offline
Joined: 2011-09-30
Reputation: 0

If you argue that you can apply the 22 W/gpm to the secondary as a standalone loop power, then you are either arguing 1) the primary gets the same 22 W/gpm or 2) the primary is not addressed.
Both of those seem pretty unworkable.

So, I still think the 22 W/gpm is a system value not a per loop value and the missing direction is in the break out between the primary and secondary. If you have a primary/secondary system, I would state your case for keeping the primary power the same in both models. If your design does not have primary/secondary or chilled water at all, then you will have to use your judgment.

You do not have to design to 22 w/gpm, you are just compared against it under Appendix G. I am not saying that agree that the value of 22 W/gpm is the right reference for all building types, shapes, sizes, and locations but the intent is mostly clear.

Paul Riemer, PE, LEED AP

Paul Riemer's picture
Offline
Joined: 2011-09-30
Reputation: 0

FYI, I called the engineer at ASHRAE and he is researching this question and
promised to respond.

I will forward his response as soon as I receive it.

Paul Diglio

Paul Diglio's picture
Offline
Joined: 2011-09-30
Reputation: 400

To put my vote it, I agree with Paul. I interpret the 22W/gpm to be 22W per system gpm (not a per loop value) and have modeled my projects in the past this way using my own judgment on how to break out between primary and secondary.

Jeremy

Jeremy McClanathan's picture
Joined: 2011-09-30
Reputation: 200