PV: Utility Rates and Demand Charge

[Skelufteay]

I have been encountering many facilities lately that are utilizing a demand charge based method to determine cost of electricity. I am trying to find the best way to lower our customer's utility bills by installing solar and/or additional technologies to trim the demand. This makes for an interesting calculation for ROI, since you cannot just take the production (kWh) times their rate and get a rough idea of monthly savings. So I am looking for some assistance and ideas... let's take an example for those example minded:

Company A is billed on a flat rate (yearly average numbers):
Billed $0.11/kWh. Uses 50,000 kWh per month. Utility bill therefore costs them $5,500/mo (not considering all those other fees they add in there)
Say Solar generates 20,000 kWh per month. They would effectively be saving $2,200/mo for a new monthly bill of $3,300.

Company B is billed on a demand based (yearly average numbers):
Billed $0.024/kWh and $20.58/kW. They too use 50,000 kWh and their peak demand for the month is 151 kW (measured for a higher peak every 15 minutes). That would mean their utility bill is $1,200 + $3,107.58 = $4,307.58.
Solar generates the same 20,000 kWh but only saves Company B $480/mo in usage (kWh).

So how can Company B benefit more from solar? or is Solar not an attractive option for companies on this billing rate?

So far I have gathered that if Company B only sees this 151 kW during a single day in the month, for 1 minute, they are still charged this rate even if their normal usage is down at 50 kW. I also know that PV can reduce the draw from the utilities and trim this peak, but if that peak happens to be during a cloudy day then they don't 'save' any on the demand rate because that one minute sets it for the rest of the month. I have been suggested looking into batteries that are charged by the grid during the night (when the peak is low) or using soft-starts on every motor we can find, but I am sure there are more ... perhaps better ideas ... out there.

 

[Hv&Lv]

Company A is billed on a flat rate (yearly average numbers):
Billed $0.11/kWh. Uses 50,000 kWh per month. Utility bill therefore costs them $5,500/mo (not considering all those other fees they add in there)
Say Solar generates 20,000 kWh per month. They would effectively be saving $2,200/mo for a new monthly bill of $3,300.

Only with net metering, which may not be available for a system that large. The savings would be less if they are paid avoided wholesale cost.

Company B is billed on a demand based (yearly average numbers):
Billed $0.024/kWh and $20.58/kW. They too use 50,000 kWh and their peak demand for the month is 151 kW (measured for a higher peak every 15 minutes). That would mean their utility bill is $1,200 + $3,107.58 = $4,307.58.
Solar generates the same 20,000 kWh but only saves Company B $480/mo in usage (kWh).

So how can Company B benefit more from solar? or is Solar not an attractive option for companies on this billing rate?

So far I have gathered that if Company B only sees this 151 kW during a single day in the month, for 1 minute, they are still charged this rate even if their normal usage is down at 50 kW. I also know that PV can reduce the draw from the utilities and trim this peak, but if that peak happens to be during a cloudy day then they don't 'save' any on the demand rate because that one minute sets it for the rest of the month. I have been suggested looking into batteries that are charged by the grid during the night (when the peak is low) or using soft-starts on every motor we can find, but I am sure there are more ... perhaps better ideas ... out there.

Staging starts may be an option. Get the real numbers, make sure the business is on the correct rate.

 

[jaggedben]

So how can Company B benefit more from solar? or is Solar not an attractive option for companies on this billing rate?

Unfortunately for those of us in the solar industry, solar is just not an attractive option for 'Company B'.

So far I have gathered that if Company B only sees this 151 kW during a single day in the month, for 1 minute, they are still charged this rate even if their normal usage is down at 50 kW.

If the highest demand will definitely be during the day, then there's a decent chance it will be lowered by the solar, especially if the high demand points don't occur everyday. I calculate that your talking about a solar system in the range of 180kW to cover usage. It's quite likely that even on the cloudiest day of the month the system will be producing a handful of kW.

But with that said, the problem is that you can't sell a system on this, because you can't guarantee there will be any effect. You have to sell it as 'icing on the cake'. And with the example you posted, that will probably kill the sale.

I have been suggested looking into batteries that are charged by the grid during the night (when the peak is low) or using soft-starts on every motor we can find, but I am sure there are more ... perhaps better ideas ... out there.

Probably not a viable option if sold as an adjunct to the solar. It is an interesting thought that batteries might in and of themselves be attractive for a demand charge customer. But I think you are talking 'grid scale storage' here, and from what I've read the available technologies are not particularly far along.

 

[Skelufteay]

Only with net metering, which may not be available for a system that large. The savings would be less if they are paid avoided wholesale cost. Staging starts may be an option. Get the real numbers, make sure the business is on the correct rate.

These are some real numbers from a facility. This one qualified to be on either pay scale so I was able to see the difference in cost. However some facilities can only be on the demand based, as they use too much energy (25kW+) and do not qualify for usage based.

Some additional stuff I learned today .. this specific utility company only allows for a maximum of 25 kW PV system to be installed. Any larger and they will not allow you to connect to the grid. This seems wrong to me, but I guess we don't get a whole lot of choice in that they are essentially a monopoly on energy to the building.

The way I see generation (I could be wrong) is that they wouldn't necessarily have to 'be paid' to see savings. If they are using less, then they are spending less which means more money they can spend else ware.

But with that said, the problem is that you can't sell a system on this, because you can't guarantee there will be any effect. You have to sell it as 'icing on the cake'. And with the example you posted, that will probably kill the sale.

Unfortunately that is what it is seeming like, unless creative solutions can be thought of. From all of the utility bills I have viewed, that basically kills the sale for more than 80% of commercial businesses. Kind of makes me wonder how other companies in the area sold PV systems if the whole area is Demand based.

Thank you for your input thus far!

 

[mivey]

I have been encountering many facilities lately that are utilizing a demand charge based method to determine cost of electricity. ... I also know that PV can reduce the draw from the utilities and trim this peak, but if that peak happens to be during a cloudy day then they don't 'save' any on the demand rate because that one minute sets it for the rest of the month.

It is interesting that you have understood the dilemma. A lot of times I see the "scandal" type response from PV installers who refuse to understand demand. The delimma you see is the same one the power company faces: getting rid of a lot of energy need does not mean we can sell the generator asset because there is still that "cloudy day" scenario when it will have to be used. FWIW, a system may experience peaks before it is bright outside (early morning loads, especially in the Winter).

You have touched on one of the basic problems: energy storage. If we could store energy for the PV ystem, then it could be a 24x7 source like the fossil resources (or some pump-back hydro dams or other energy storage technology). I have heard solar cell manufacturers state they are ready to go "head-to-head" with the fossil & nuke units as the costs have come down so much for the cells. Presentation slides will focus on daylight loads. But when asked asked about the dark load and the cost of energy storage to provide resources that really do compare with fossil & nukes type units they will waffle. The response at a recent meeting I attended was "well we really haven't studied those costs but they are still very costly and do change the numbers". Well duh.

But not all PV folks are like that and they do realize the problem. Here is an article: http://www.rechargenews.com/energy/solar/article324145.ece

The belief that cheap energy storage will revolutionise the PV industry, allowing it to compete head-on with dispatchable fossil fuels, is increasingly being challenged by experts who say the technologies are developing too slowly.
...
"But the reality is that electricity is very difficult to store ? it?s the first thing you learn as an electrical engineer. And I have to tell you that unfortunately it?s still true.?

Therond says the cost of storage would have to decrease ?by a factor of four? before it becomes useful to PV in any commercial context apart from niche markets.

Focus on the load seems to be a better use of resources for some sites. As for storage, there is a wealth of info and you could start here: http://en.wikipedia.org/wiki/Energy_storage and also add additional search terms like "industrial" and "PV" and "demand management" you should find more info than you want. You should look at not only improving the load factor (trimming the peak demands) but also see if you can take advantage of time-of-use or seasonal rates which provide a means to shift production into cheaper POCO hours/days/months.



Energy rate note: even for the customers billed on energy rates, the costs are allocated using load profiles that include demand. That is because the costs really do have a demand component. The reason residential and small commercial have traditionally had energy rates is because the cost of demand meters made it impractical to put a demand meter on a load with a relatively small bill. With the age of smart meters, we should see a move to rates that better reflect the costs and that means a demand component in the rate.

 

[mivey]

this specific utility company only allows for a maximum of 25 kW PV system to be installed. Any larger and they will not allow you to connect to the grid. This seems wrong to me, but I guess we don't get a whole lot of choice in that they are essentially a monopoly on energy to the building.

They have a responsibility to provide quality power as well. They also limit the size motor you can start across the line for similar reasons.

Unfortunately that is what it is seeming like, unless creative solutions can be thought of. From all of the utility bills I have viewed, that basically kills the sale for more than 80% of commercial businesses. Kind of makes me wonder how other companies in the area sold PV systems if the whole area is Demand based.

There may be several reasons but the main ones I see are:

1) The business really does believe that installing solar will help the environment and they are willing to pay extra to do their part.

2) The PV sales company sold them a bill of goods and they think they will be saving a bunch of money. These are the ones I have to deal with the most because when the savings don't turn up they start blaming the POCO and look to them for answers as to why they are "being ripped off".

3) The business is going to use solar in addition to changing the way they use energy (i.e. change their load profile).

4) The powers-that-be are giving the business money to help offset the capital costs of the solar system.

Even with #1, #3, and #4, I see many that do not realize the actual costs. The ones that do are usually the ones that work with the POCO on the front end to get realistic numbers and to explore rate options.

 

[JoeStillman]

I have used the PVWATTS program to estimate solar production (watthours) for a given location. It's a free tool at http://www.nrel.gov/rredc/pvwatts/

This might help you with your payback analysis.

Good luck with that. Here at 40 degrees north latitude, $0.14/kWH, I have yet to evaluate a system that pays for itself in less than about 25 years. The only way to make it worthwhile financially is if someone else lays out the first-cost. I have never had a client that was so gung-ho for clean energy that they were willing to take the financial hit without some help from the government.

 

[mivey]

The delimma you see is the same one the power company faces: getting rid of a lot of energy need does not mean we can sell the generator asset because there is still that "cloudy day" scenario when it will have to be used.

I forgot to mention the demand costs for the assets used to tie to the grid (meters, transformers, lines, poles, staff, protection, regulators, etc).

 

[mivey]

I have never had a client that was so gung-ho for clean energy that they were willing to take the financial hit without some help from the government.

Interestingly enough, we have had customers pay extra for a block of "green" energy. I suspect it comes from the PR budget.

 

[ghostbuster]

In our area the government has a very high level of subsidy and pays solar producers $.60 to $.80/KW.produced

Paybacks are 8-10 years

 

[petersonra]

solar PV is just not viable economically without huge subsidies somewhere in the mix. that is just the way it is.

having said that, there are other solar technologies that are economically viable that are just ignored entirely, likely as they are not sexy enough.

take some of the space used up by solar PV arrays and use it for solar heated domestic hot water. usually the payback is a few years.

as someone mentioned, solar is not an especially reliable electricity producer. like wind, you can never depend on it being there at any particular point in time.

 

[Skelufteay]

Ah! figured out how to multi-quote...

solar PV is just not viable economically without huge subsidies somewhere in the mix. that is just the way it is. having said that, there are other solar technologies that are economically viable that are just ignored entirely, likely as they are not sexy enough. take some of the space used up by solar PV arrays and use it for solar heated domestic hot water. usually the payback is a few years. as someone mentioned, solar is not an especially reliable electricity producer. like wind, you can never depend on it being there at any particular point in time.

You should look at Sundrum Solar's technology ... it makes thermal sexy and shares space with PV.

It is interesting that you have understood the dilemma. A lot of times I see the "scandal" type response from PV installers who refuse to understand demand. The delimma you see is the same one the power company faces: getting rid of a lot of energy need does not mean we can sell the generator asset because there is still that "cloudy day" scenario when it will have to be used. FWIW, a system may experience peaks before it is bright outside (early morning loads, especially in the Winter).

Sometimes I think I understand it too well ... and volunteer in giving up sales. I'd never sell a system if it wasn't a good deal for the new owner. The way I have come to understand the Demand charge is an opportunity cost of energy. You are paying for the ability to cover spikes instead of browning out.

You have touched on one of the basic problems: energy storage. If we could store energy for the PV ystem, then it could be a 24x7 source like the fossil resources (or some pump-back hydro dams or other energy storage technology). I have heard solar cell manufacturers state they are ready to go "head-to-head" with the fossil & nuke units as the costs have come down so much for the cells. Presentation slides will focus on daylight loads. But when asked asked about the dark load and the cost of energy storage to provide resources that really do compare with fossil & nukes type units they will waffle. The response at a recent meeting I attended was "well we really haven't studied those costs but they are still very costly and do change the numbers". Well duh.

We get asked that a lot, and have actually included it in our slides. I've spoken to many owners that had previous bids on solar, who were promised 100% coverage of the load (but on lights only!) and then ask us why we can only cover 15% of their total load. It's all about how you word it and how much you educate your customer. We always give full information to new clients ... it makes the sale much harder but makes the final install and outcome so much better. Batteries, we have found, sometimes exceed the cost of the panels / wiring themselves. It makes payback not even worth mentioning, unless it is an off grid application. I am greatly looking forward to their work in carbon nanotube batteries, but that is years away. I've also heard (All political comments aside - please don't make a political reply to this haha) Obama has plans on spending more money in Research instead of installation like he did for the past 4 years. Hopefully some new storage, net meters, and even panels come out of it.

Focus on the load seems to be a better use of resources for some sites. As for storage, there is a wealth of info and you could start here: http://en.wikipedia.org/wiki/Energy_storage and also add additional search terms like "industrial" and "PV" and "demand management" you should find more info than you want. You should look at not only improving the load factor (trimming the peak demands) but also see if you can take advantage of time-of-use or seasonal rates which provide a means to shift production into cheaper POCO hours/days/months.

I love Wikipedia! I totally forgot about looking at this article, thank you!

I have used the PVWATTS program to estimate solar production (watthours) for a given location. It's a free tool at http://www.nrel.gov/rredc/pvwatts/

Yep! we use this as a supplemental payback device. It is one of 5 things we use when presenting payback scenarios, as you also have to look at Federal, State, Local incentives as well as depreciation of the system (plus increasing energy rates and decrease in production of panels over lifetime). Doing so yields paybacks of less than 1-4 years in all of our projects.The Bonus Depreciation of 2010-2012 was a huge help in this ...hoping it comes back this next year. You can find more info on this stuff at www.DSIREUSA.org. I have found several locations that, with zero incentives, have 5 year paybacks, but that is my little secret

I forgot to mention the demand costs for the assets used to tie to the grid (meters, transformers, lines, poles, staff, protection, regulators, etc).

I should have elaborated a little more, but it was late at the time. I can totally understand their requirements to keep the system small as they are paying for this stuff yes. But! If we were to install a 50kW system that covered only 50% of the buildings load for example (in the daytime) then we would be saving their equipment from being used as much, and would not send more power on the lines than we could produce, so they shouldn't be worried. In that aspect, it makes it seem like they are artificially restricting competition, after all, that is what it would be if not for the requirement of green energy on 20% of their customers buildings by 2020 (I think that is the final goal date?). But having a sliding requirement causes too much work to be put into each case, so a firm number was set.

Thank you again for the replies ... makes me feel better that this is an actual issue and not something I "should have already known".

 

[mivey]

Sometimes I think I understand it too well ... and volunteer in giving up sales. I'd never sell a system if it wasn't a good deal for the new owner. The way I have come to understand the Demand charge is an opportunity cost of energy. You are paying for the ability to cover spikes instead of browning out.

That is one aspect. Also consider that, to use the dreaded water analogy, water is the energy but the demand tells you the fastest flow rate. With that, demand determines how big the equipment has to be. A big pipe needed for a high peak flow but that spends most of its life with relatively low water flow raises the cost of water.

If we were to install a 50kW system that covered only 50% of the buildings load for example (in the daytime) then we would be saving their equipment from being used as much, and would not send more power on the lines than we could produce, so they shouldn't be worried.

That's true for the days you are running and if you don't use power at night. If it were really true that you have covered 50% of the load then the demand would go down 50%.

Let's consider that you have covered 50% of your load part of the time. During the run times you have made it where they can shut off one of the units. That saves variable costs like fuel. But they still have to have the unit standing by for you when you can't run. The wire capacity still has to be standing by for you when you can't run. Etc on the standby equipment.

Now let's consider what happens when your unit trips off line. Regulators have to step, units have to ramp up, etc. Your neighbors have a power quality issue. The utility suffers added costs for power imbalance penalties.

Suppose you have a unit of significant size. Now there are coordination and power flow issues because there are multiple sources on the feeder. It is expensive to control these problems.

That is just a few of the things that the POCO has to worry about so I'm not following what you mean by they shouldn't be worried.

 

[jaggedben]

In the vast majority of cases, a given solar system will have no noticeable effect on the demand costs for grid assets. That is to say, most individual systems are too small to produce a noticeable effect by themselves. And of course, they can't make demand go up, only down.

I understand (in general) the issues utilities face, but 25kW seems like a very, very low number for a utility to use as a general cut-off on installing a PV system. Seems like a business decision rather than a technical one. Either they don't want to spend the money on developing requirements for larger systems, or on the staff to review applications, or they have simply decided to discourage and limit PV (or all of the above).

I have read (can't remember where) that in Germany every PV system, large and small, is required to have box that reports its production status to the grid operator over the internet. That may go some way towards explaining how Germany has managed to integrate such large amounts of PV.

 

[G._S._Ohm]

I am trying to find the best way to lower our customer's utility bills

This is a pure Game Theory problem.
The customer's opponent in this game is the PoCo. In some ways it is a Zero Sum Game because Poco's loss is the HO's gain.

You can bet PoCo has anticipated every reasonable move by all HOs and has guarded against being in a losing position more than half the time, if that is possible.

The book by Barry Nalebuff, Dixit, Avinash called something like Thinking Strategically may help you see what you and the HO are up against.

 

[mivey]

This is a pure Game Theory problem.
The customer's opponent in this game is the PoCo. In some ways it is a Zero Sum Game because Poco's loss is the HO's gain.

Not always. There are ways for both to save and that is when the POCO should have no problem passing along the savings (assuming that the cost-accounting method does not eat up all the savings).

 

[G._S._Ohm]

passing along the savings

Trickle down? Maybe.