Can someone point me to an somewhat detailed explanation of electric power demand metering? I understand that high peak demands require more capital equipment expense for the utility, and that's the reason demand metering is used to more fairly distribute that expense anong the users.
I'd like to learn more specifics about just how an electric utility's demand metering system works and how it affects the user's bill.
I sort of remember that the "demand parameter" results from averaging power draw over some relatively short time period like 15 or 30 minutes, but I don't know how often that average gets taken and if it ever gets "reset".
Okay, he says, (shifting into lecture mode -- and don't worry, this is leading up to some practical advice.)
While the principles of demand metering are fairly straightforward, the details of implementation aren't and tend to vary from regulated jurisdiction to regulated jurisdiction (or in the case of large customers, from utility to utility and for very large customers from contract to contract.)
The whole idea turns on the simple fact that the cost of producing electricity breaks down into three components: The cost of building and maintaining the generating capacity, the cost of the fuel to run the generating plant and everything else. Usually the first two are considerably larger than the third.
Fuel costs track electricity usage pretty closely. The relationship isn't proportional because most utilities use a mix of generating sources with fuel costs ranging from practically zero (hydro) to very high (natrual gas fired turbines). But the more electricity you use, the more you cost the utility in fuel.
Capital costs are another whole can of worms. They don't necessarily track electricity use at all because of the capital cost of supplying a customer depends on the total load on the system at the moment the customer wants the power. If your total system capacity is, say 2000 MW, and the system is running at full capacity when one of your customers wants an extra 10 KW, you are theoretically screwed because you don't have the capacity to meet the customer's demand. In practice you buy the extra 10 KW from one of your neighboring utilities at a high price. If it happens to be a peak demand period for the entire region the price will be extremely high and you're screwed economically rather than practically.
Of course you pay a cost penalty at lesser demand levels as well. It's a truism in the utility business that the last MW of power is the most expensive. If that added 10 KW demand tips you over from running steam plants to firing up a gas turbine, then your costs go up.
But what if the customer only needs that 10KW for a half hour or so? Then, under the conventional cents-per-kilowatt hour method of pricing electricity you're well and truly screwed because the demand doesn't stay high enough long enough to let you recover the capital costs of the extra plant needed to meet peak demands.
Which brings us to demand pricing.
Demand pricing isn't new. Utilities have been using it for large industrial customers ever since the 1930s at least and for commercial customers since the widespread advent of air conditioning in the
1960s. In the 1980s some utilities began offering demand pricing to residential customers. Part of your bill is based on the traditional cents-per-hour model and part of it is capacity cost based on your peak use of electricity during peak periods. Often the demand part will have a rachet, which means that no matter how low your demand drops in subsequent periods, the demand charge will remain high for months.
That's the simple, logical, part of demand pricing. The un-simple, illogical, part comes when it comes to setting the demand rate, including deciding just how costs will be apportioned and how the prices will be set. This is one part economics, one part back room politics, one part pressure groups and one part smoke and mirrors.
The details of the demand pricing scheme for residential and commerical customers is worked out in a deal between the utility and the state's public utility commission. Since there are dozens of ways of figuring the 'cost' of generating capacity, not to mention dozens of methods for deciding what constitutes cost recovery, and still more sehemes for deciding on what is a fair rate of return, the reasons for what gets hammered out in those rate hearings is pretty much beyond human ken. In other words, the details probably won't make sense to anyone who wasn't in those meetings, even though there are elaborately logical explanations for every single piece of the process.
What this means is that there is no one authoratitive source that will explain why your demand charges came out the way they did. However if you want to know how the charges are figured, that information is included in the rates filed with your local public utility commission. You can get that information either from the utility or from the public utility commission and if you ask real nice someone at the PUC will probably help you decipher it all.
--RC (who used to cover electric rate hearings as a reporter)
Sleep? Isn't that a totally inadequate substitute for caffine?
A very nice description of utility problems in a region such as Vermont where vacation condos have electric heat and the hot tubs and saunas are turned on during ski season. The seasonal variations mean that delivery capacity has to be sized for those peak loads but is way oversized the rest of the year.
Another problem is in an industrial plant where a very large motor sucks great amount of power when starting up. The demand meter is set for a short time interval for that load. I remember someone, perhaps on this newsgroup, told the story of a maintenance crew servicing a large motor, bumping the contactor to turn the rotor several times on a weekend. The cost for doing that was pretty bad.
That last sentence probably gave me the answer I was looking for RC.
A few months ago my son greatly reduced the peak power drain of the electric heaters used on a power hungry degreasing tank system in his shop. He rewired the nine heating elements in the tank so that they drew one third the power they had been drawing and thus stayed powered up three times as long to keep the tank temperature at its set point.
He'd made some measurements prior to changing things over and found that the original heater setup pulled nearly as much current when it was on as as everything else in the place was drawing, pretty much doubling the power drain when those tank heaters were on.
He was suprised to find that after waiting out FOUR MONTHS of "estimated" meter readings, when an actual reading was finally taken his demand number hadn't dropped much at all.
The goodfaddles he's allowed to speak to in the utility's "customer service" department sound like they know less about the subject than we do.
That's one more example of why that act which a farmer brings his cows to a bull for is referred to as..."service".
In practice it's worse than that because many of those condos are only occupied on the weekends. If you don't have enough hours of use to spread the demand over, the cents-per-kilowatt-hour model produces some real unfortunate results.
When they were first settind demand rates here in Arizona there was a similar phenomenon called the 'pool party effect'. Your teenage kid has a bunch of friends over, they've got the patio doors standing mostly open, the AC cranked up and the pool pump running full blast. It's only a for a few hours but what it does to a demand rate is truly ugly.
Now the good news is that if you understand the details of your demand rate you can often game it. One of the experimental demand rates down here used cents-per-kilowatt-hour that varied steeply with the time of use (since that tracks residential demand). Power consumed between about 10 p.m. and 6 a.m. cost practically nothing, power used in the morning and mid-day cost a lot more and power used in the peak late afternoon and early evening hours was very expensive. So they had one homeowner who turned his air conditioning on full blast at 10 p.m. and ran it that way until 6 a.m. Then he used almost no electricity for the rest of the day. The house was ice cold when he woke up and pretty warm by 10 p.m. but he cut his electric bill to something like $17 a month.
Needless to say, that variation of demand rate didn't make it past the experimental period.
--RC
Sleep? Isn't that a totally inadequate substitute for caffine?
A lot of demand rachet rates for commercial customers set the rachet annually, which means you don't see a drop in your bill for months.
However by law in most jurisdictions, the utility company must actually read the meter if the customer demands it.
The utility company's customer service department is about the worst place to go looking for information on the details of rates. Not only don't the driods on the phones know, the company has no incentive to tell you.
My strong suggestion is that you ask, in writing, for a copy of the rate you are on, along with an explanation of all the details and technical terms. Some utilities even have this on their web sites.
The next thing to do is to contact your state's public utility commission and get a copy of the rate from them, plus and explanation of how it works. Generally PUCs love to see consumers taking an interest in their utility bills -- as long as they don't get abusive about it.
Once you understand the rate it may be possible to modify your behavior in some way to minimize your utility costs under that rate. In some cases you can 'game' the rate very nicely (although the PUC isn't going to point out loopholes for you) if you understand what's going on.
The other thing to look at is the possibility of switching rates. It is not unusual for utilities to have several possible rates for each class of customers and it may be you can save money on some other plan.
Oh yeah. Utility companies also frequently have a service to help customers, including businesses, save energy. It might may to have your place checked over.
No problem. Good luck.
--RC
Sleep? Isn't that a totally inadequate substitute for caffine?
They have actually encouraged that practice in the past at some utilities, though I'll bet they don't discount it as deep as they increase the day rates - LA DWP even used one dodge at one of their own buildings...
They build a huge insulated water tank in the "back 40" of the building and put in some huge refrigeration, and turn that tank of water into a big block of ice overnight using cheaper electricity. And during the day, they circulate water over that big block of ice, and pump the chilled water inside to air handler coils to cool the building.
Result being the only loads needed during the day are the fans and pumps, which is nothing compared to running big chiller compressors.
If we didn't do just fine with swamp coolers most of the time, I'd be tempted to build a system like that myself. That, or use a more efficient water cooled condenser on the AC and dump the "waste" heat into the swimming pool.
There are many ways for utilities to figure the appropriate charge for electricity. They can range from charging only for peak demand (nowadays I don't think any utilities use this approach) to changing the price per KWH in near real time. Today's metering technology has made real time pricing possible although the equipment may cost more than any additional income it would generate. Large users will be the first to adopt or possibly be forced into real-time pricing. Some large users have accepted this type of metering because it can save them a lot of money and it also helps the utility avoid the cost of adding capacity. The savings is possible only if the user is willing to shed load when the cost becomes too high.
Energy use is measured in KWH and demand is measured in KW. Demand is usually averaged over short time (15- 60 minutes) and the largest average demand for the month or possibly the year is the peak demand. In the last paragraph I have briefly described how a typical demand meter works
There are different ways the utility can use the peak demand. They usually charge an amount based on the peak KW. They also often change the cost for each KWH based on the peak demand. If your peak demand is much larger than your average demand each KWH will cost more. If your peak demand was only slightly above your average demand your cost per KWH would be lower. To figure the cost per KWH they use something called hours of demand use.
As an example:
Demand charge is $8.00 per KW
For the first 100 Hours of demand use the charge is $.10 per KWH For the next 150 hours of demand use the charge is $.07 per KWH For the next 200 hours of demand use the charge is $.05 per KWH For the balance the charge is $.03 per KWH
Two worked out examples:
-----Customer number 1-----
Peak demand is 10,000 KW total energy is 6,300,000
Demand Charge 10,000 KW @ $8.00 $80,000
100 hrs X 10,000 KW = 1,000,000 KWH @ $.10 = $100,000 There are 5,300,000 KWH are left
150 hrs X 10,000 KW = 1,500,000 KWH @ $.07 =$105,000 There are 3,800,000 KWH left
200 hrs X 10,000 KW = 2,000,000 KWH @ $.05 $100,000 There are 1,800,000 KWH left
1,800,000 KWH @ $.03 = $54,000
Total energy charge $359,000 Total including demand charge$439,000
-----Customer number 2----- Peak demand is 20,000 KW total energy is 6,300,000
Demand Charge 20,000 KW @ $8.00 $1600,000
100 hrs X 20,000 KW = 2,000,000 KWH @ $.10 = $200,000 There are 4,300,000 KWH are left
150 hrs X 20,000 KW = 3,000,000 KWH @ $.07 =$210,000 There are 1,300,000 KWH left
200 hrs X 20,000 KW = 4,000,000 KWH But there are only 1,300,000 KWH left 1,300,000 @ $.05= $65,000
0 KWH @ $.03 = $0
Total Energy Charge $475,000 Total including demand charge $635,000
Notice that customer number 2 used the same amount of energy a customer 1 but paid much more for his energy as well a paying a larger demand charge.
Meter Description: The most common way to measure demand is to have a watt hour meter with two sets of dials that count the revolutions of the meter disk. The disk rotates based on the energy (KWH) used. One set of dials measures the total energy used. The other set measures the energy used for a short period of time. Maybe 15 minutes to an hour. The demand in KW is equal to Kilowatt hours divided by the time interval. The design of the demand counter is such that it is reset to zero at the end of the time interval but the dials don't reset. As the KWH is measured for the next time interval the counter advances but doesn't move the dials unless the total count exceeds the reading from a previous interval. After the meter is read, probably once a month, the demand counter is reset to zero. The peak demand used to calculate the charges can be based on the peak for each month or the peak demand may be the highest monthly reading for the preceding year. (Some newer system are electronic and can be read remotely but the idea is still the same.)
It looks like a built in rip off i have 2 places with those meters alls i have to do is live like a normal person turn on the air conditionr turn on the pool pump turn on the air compresser turnon the hot water i am screwed befor i get to take a hot shower fill up my floates jump in the pool and set in the air condition!!!!!!!!!!!!!!!!!!!!!
Thanks much Stephen. I'll forward your post to my son and we'll chew it over. We're both EEs, but neither of us got involved with the intricacies of this stuff before, and our curiosities are peaked. (Misspelled pun intentional.)
Load shedding might be something to look at. I don't know what the thermal storage capacity of his degreasing tank is, but it may be large enough to let him put in some sort of system that monitors the other loads in his shop (Pause for a proud papa plug here.)
formatting link
and prevents the tank heaters from turning on when the other (uninterruptable) loads are above some predetermined limit.
I'm also wondering whether or not he could fake out the measuring system by gating the power supplied to the tank heating system with period close to that of the demand metering period and a duty cycle just large enough to keep the tank temperature in a functional range during use. Based on what you said above about KWH used divided by the 15 to 30 minute averaging interval, that would seem like it could pay dividends by reducing the calculated "peak demand".
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.