Short duration loads with Demand Meters
Short duration loads with Demand Meters
(OP)
Does a 15 minute demand meter take a 15 minute RMS value? Or does it take 15 minutes of MWH and divide by 4?
Assuming it is the latter, is there a common way to account for loads that vary minute by minute? For example, a crane load might have a duty cycle of:
2 MW for 10 seconds
0 MW for 20 seconds
-0.5 MW for 10 seconds
0 MW for 20 seconds.
The RMS value would be 0.84 MW
The MWH/0.25hours would be 0.25 MW for a net meter
The MWH/0.25 hours would be .33 MW for a 4 quadrant meter
If I size cables for a new installation based on the 15 minute demand meter reading from a similar installation, I am concerned that the losses (and heating) might be higher than expected. Although diversity helps at the substation level, there are individual feeders serving each load.
Assuming it is the latter, is there a common way to account for loads that vary minute by minute? For example, a crane load might have a duty cycle of:
2 MW for 10 seconds
0 MW for 20 seconds
-0.5 MW for 10 seconds
0 MW for 20 seconds.
The RMS value would be 0.84 MW
The MWH/0.25hours would be 0.25 MW for a net meter
The MWH/0.25 hours would be .33 MW for a 4 quadrant meter
If I size cables for a new installation based on the 15 minute demand meter reading from a similar installation, I am concerned that the losses (and heating) might be higher than expected. Although diversity helps at the substation level, there are individual feeders serving each load.






RE: Short duration loads with Demand Meters
When the math is done, a step increase in heat input or load will result in a temperature increase to 63% of the total temperature increase at the end of the first time constant. In the second time constant,the heat will rise 63% of the remainder. After 5 time constants the temperature has stabilized for all practical purposes.
Example:
A change in load will result in a final change in temperature of +100 Degrees.
At the end of the first time constant the temperature will have increased by +63 degrees. There are 37 degrees left to the final temperature.
At the end of the second time constant the temperature will have risen by 63% of 37 degrees or 23.3 degrees for a temperature rise of 63 deg + 23.3 deg = 86.3 deg. There are 13.7 degrees left to the final temperature.
At the end of the third time constant the temperature will have risen by 63% of 13.7 degrees or 8.6 degrees for a temperature of 86.3 deg + 8.6 deg = 95.5 degrees.
As you can see this is closing on the limit of 100 degrees quite rapidly.
The constant of .63 is based on natural logarithm base which is 2.718
One minus the reciprocal of the base of natural logs =~ .63
1 - (1/ 2.718) = .63
The curve is the same one you see charging a capacitor or applying current to an inductor.
If a step load is applied for one time constant and rises by 63 degrees, and is then removed, the temperature will drop on a similar curve, and will take five time constants to stabilize.
However, the decay is now based on 63 degrees rather than 100 degrees.
So, a three minute load application will result in a demand charge of 63% and when this load is removed it will take 15 minutes to reset to zero.
But that's old school when thermal elements were used in electro-mechanical meters.
Newer electronic meters tend to take the average of a rolling 15 minute window. The maximum 15 minute average in a month is the demand for that month. Are there any meter men out there who can fill in any details here??
Mr. Edward Cowern P.E. explains RMS HP loading in the Cowern Papers. Available on the Baldor Motors web site.
http://www.baldor.com/pdf/manuals/PR2525.pdf
See RMS Horse Power Loading (note the pdf pages do not align with the paper pages. Jump to PP 49 to get to paper page 43)
Cable sizing; In the Canadian CEC there is a table for sizing conductors feeding variable duty motors. There are probably similar rules in the NEC
Bill
--------------------
"Why not the best?"
Jimmy Carter
RE: Short duration loads with Demand Meters
RE: Short duration loads with Demand Meters
RE: Short duration loads with Demand Meters
Since I work for a utility, we usually size cables significantly smaller than NEC allows. NEC 430.22(E) lists 85% of nameplate amps for intermittent duty motors. In my case, sizing based on the demand meter would be around 20% of nameplate and the RMS calculation would require about 30% of nameplate.