Re: [EnergyPlus_Support] RE: Re: Cubic Curve in variable speed pump [5 Attachments]

Comnet has some here: http://www.comnet.org/mgp/content/685-pumps?purpose=0

Sorry, still no clear numbers, but reading from fig 48:

VSD Efficiency, Fraction 100% Design Speed
0.88, 0.0
0.92, 0.20
0.95, 0.40
0.97, 0.60
0.975, 0.80
0.98, 1.00

Assuing Pump Impeller Speed is directly proportional to Motor Speed (fixed gear ratio)

VSD Efficiency, Fraction 100% Impeller Speed
0.88, 0.0
0.92, 0.20
0.95, 0.40
0.97, 0.60
0.975, 0.80
0.98, 1.00

This allows us to use Pump Affinity Laws to relate, Fraction 100% Impeller Speed to Power, Flowrate and Pressure. Flowrate is linearly proportional to Impellar Speed.

VSD Efficiency, Fraction 100% Flowrate
0.88, 0.0
0.92, 0.20
0.95, 0.40
0.97, 0.60
0.975, 0.80
0.98, 1.00

So if we can get Pump Efficiency, Fraction 100% Flowrate at a constant pressure (as Jim says, most systems operate like this), then we can multiply the efficiencies to get the total efficiency into the equation and set motor efficiency to 1.

Total Efficiency, Fraction 100% Flowrate @ constant pressure (rated max pressure)

0.88*npump_0, 0.0

0.92*npump_20, 0.20
0.95*npump_40, 0.40
0.97*npump_60, 0.60
0.975*npump_80, 0.80
0.98*npump_100, 1.00

Power_in = Flowrate*dP/neff_tot...and dP changes quadratically with flowrate, therefore Power is propotional to Flow^3, therefore

P_in*neff_tot <=> Flowrate^3

FractionFullLoadPower_elec, Fraction of 100% Flowrate @ constant pressure (rated max pressure)
0.88*npump_0, 0.0^3
0.92*npump_20, 0.20^3
0.95*npump_40, 0.40^3
0.97*npump_60, 0.60^3
0.975*npump_80, 0.80^3
0.98*npump_100, 1.00^3

FractionFullLoadPower_elec = C1 + C2*PLR + C3*PLR^2 + C4*PLR^3
where C1,C2,C3,and C4 are Coefficients 1 ? 4 (below) and PLR is thee Part Load Ratio of the Flowrate.

For example:

PLR == Total Efficiency, FractionFullLoadPower @ constant pressure (rated max pressure)
0.88*0.10, 0
0.92*0.40, 0.008
0.95*0.62, 0.064
0.97*0.80, 0.216
0.975*0.88, 0.512
0.98*0.89, 1

PLR	FractionFullLoadPower_elec
0.176	0
0.368	0.008
0.589	0.064
0.776	0.216
0.858	0.512
0.8722	1

A regression analysis in EXCEL gives y = 13.072x3 - 17.091x2 + 6.7231x - 0.7416

i.e.
- 0.7416
+6.7231
- 17.091
+13.072

2012 ASHRAE Handbook--HVAC Systems and Equipment (SI), section 44 fig. 48;
2012 ASHRAE Handbook--HVAC Systems and Equipment (SI), section 45;

This is way too much thinking for me. I'm off to lunch...happy error finding.

2014/1/9 jeannieboef@xxxxxxxxx <jeannieboef@xxxxxxxxx>

Â

There was an ASHRAE Journal article recently about VSD partload performance...maybe it mentions something...I'll have a look.

Mit freundlichen GrÃ¼Ã?en- Sent from my iPhoneÂ (excuse the brevity)

i. A.
Jean Marais
b.i.g. bechtold
Tel. Â +49 30 6706662-23

On 09.01.2014, at 00:45, Jim Dirkes <jim@xxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:

Â

I apologize!Â I looked through my reference materials and cannot determine where I got those coefficients, or why I thought they were from ASHRAE 90.1.Â If I figure it out, Iâ??ll post the information.Â Otherwise, try the calculation I have attached. It assumes:

Â·Â Â Â Â Â Â Â Â The design delta P varies (shown as 100 ft H2O in the example)

Â·Â Â Â Â Â Â Â Â Pumps normally operate above a minimum speed (shown as 30% in the example)

Â·Â Â Â Â Â Â Â Â They normally maintain a minimum loop pressure setpoint (shown as 8 psig in the example)

Â
Â

James V Dirkes II, PE, BEMP, LEED AP
www.buildingperformanceteam.com

Energy Analysis, Commissioning & Training Services
1631 Acacia Drive, Grand Rapids, MI 49504 USA
616 450 8653
Â

From: EnergyPlus_Support@xxxxxxxxxxxxxxx [mailto:EnergyPlus_Support@xxxxxxxxxxxxxxx] On Behalf Of Adam Barker
Sent: Wednesday, January 08, 2014 4:28 PM
To: 'EnergyPlus_Support@xxxxxxxxxxxxxxx'
Subject: [EnergyPlus_Support] Re: Cubic Curve in variable speed pump

Â
Â
Hi James,

Â
Maybe Iâ??m not looking in the right place, but where exactly in ASHRAE 90.1 do you find these coefficients?

Â
Thanks,
Â

Â
Â
Â

Dear Adrian,
Lacking actual data from the pump and VSD supplier, use the ASHRAE 90.1-recommended coefficients:
E+ Format

90.1 VSD

C1

0.203671

C2

-0.827077

C3

1.181191

C4

0.442055

James V Dirkes II, PE, BEMP, LEED AP
www.buildingperformanceteam.com<http://www.buildingperformanceteam.com/>
Energy Analysis, Commissioning & Training Services
1631 Acacia Drive, Grand Rapids, MI 49504 USA
616 450 8653

From: EnergyPlus_Support@xxxxxxxxxxxxxxx [mailto:EnergyPlus_Support@xxxxxxxxxxxxxxx] On Behalf Of adrianchong85@xxxxxxxxx
Sent: Wednesday, January 08, 2014 2:46 PM
To: EnergyPlus_Support@xxxxxxxxxxxxxxx
Subject: [EnergyPlus_Support] Cubic Curve in variable speed pump

Dear EnergyPlus Users,

I'm trying to model a variable speed pump in eplus and one of the inputs required are coefficients to the cubic curve

FractionFullLoadPower =C1 + C2PLR + C3PLR2 + C4PLR3

where PLR = VolFlowRate / NomVolFlowRate

My question is how do we go about obtaining these coefficients? Is it correct to assume a constant head (maybe assume nameplate head) and using nameplate Power and Flow Rate as nominal data, obtain FractionFullLoadPower and PLR by reading off the Pump's performance curves (at different points with the same head) and then use least squares regression to get the coefficients.

Â
Adam Barker, C.E.T., LEED AP BD+C

SustainabilityÂ Project Manager
Provident Energy Management Inc.

T: 416-736-0630 x 1874 | abarker@xxxxxxxx

Â

<Pump Power Curve.xlsb>