Last week I was contacted by a Queenslander named Terry. He was chuffed about the electricity bill savings he was getting by using a demand tariff with his home battery.
Demand tariffs offer cheaper grid electricity but include a demand charge based on your peak power draw1 from 4-9 pm. As Terry used zero grid electricity during this period, he was happy with his $0 in demand charges thanks to his battery.
He contacted me because he wanted to know if using a demand charge was enough to make home batteries pay for themselves within their 10-year warranty period.
I was intrigued. I’d heard people claiming demand tariffs can allow a home battery to pay for itself inside 10 years, but because they were selling batteries, I had assumed these people were full of hot air. To find out how much a demand tariff was likely to save a battery household, I decided to make the following comparison:
- The best flat tariff vs. the best demand tariff.
If Terry’s annual electricity bill with a demand tariff reduced his annual electricity much more than a flat tariff, then home batteries at their current prices could pay for themselves, so long as…
- The battery can almost always reduce demand charges to zero.
- The home, like Terry’s, consumes more electricity than average.
But, to my surprise, even with the assumption that demand charges would always be $0, using a flat tariff turned out to be cheaper.
The difference wasn’t much, but I was expecting the demand tariff to come out significantly ahead. In practice, the demand tariffs will probably cost even more because the home is likely to pay a demand charge occasionally.
So it looks like claims that demand tariffs are enough to make a home battery pay are wishful thinking. But I still have an open mind – if you know of a particular demand tariff that can make batteries pay, let me know in the comments.
Household Details
Terry lives in South East Queensland. Not in the sprawling blob of Brisbane or in the cities that snake out of Brisbane to its north and south along the coasts. But he does live close enough to still be in the Energex area and have plenty of choices for electricity retailers.
I’ve put details of his electricity production and consumption below but rounded some figures to make comparisons easier. The rounded figures are what I’ve I’ve used to determine his electricity bills:
- His solar system is 13.3 kilowatts and generates 20,000 kilowatt-hours a year
- Solar energy self-consumption is 7,000 kilowatt-hours per year
- Grid electricity consumption is the same at 7,000 kilowatt-hours per year.
- Solar exports are 13,000 kilowatt-hours a year.
- His battery allows him to avoid all demand charges.
To make things even simpler, I will assume his battery is magical and operates with 100% efficiency. The actual round trip efficiency of his DC-coupled battery is likely about 90%. If it were 90% then 10 kilowatt-hours of energy would have to be put in for every 9 kilowatt-hours taken out.
His home consumes 7,000 kilowatt-hours of the 20,000 his solar power system produces, making his solar self-consumption 35%. This is high for a 13.3 kilowatt solar system but includes his battery storing solar energy during the day for use at night.
Flat Tariff Total Bill: $241
I’m going to use the best electricity plan I can find to work out the annual cost of a flat tariff. Using the SolarQuotes Electricity Retailer Comparison tool, the best I can find is Alinta Energy’s Priority Plus Plan:
- Daily supply charge: 99.47 cents
- Per kilowatt-hour charge: 18.16 cents
- Solar feed-in tariff: 11 cents
There are electricity plans from AGL and Momentum that have higher feed-in tariffs, but they are only available for solar systems that are 10 kilowatts or smaller.
After crunching the numbers and adding $37 to allow for Queensland’s metering charges, the annual electricity bill with this tariff comes to:
- $241
That’s pretty good. A 13.3 kilowatt solar system is normally big enough to give households a hefty credit on their electricity bills, but because Terry’s total electricity consumption is nearly three times what’s typical for a home in the area it’s necessary to shell out some money, but not much.
Note this result includes the benefit of storing solar energy in the battery for use at night.
Demand Tariff Total Bill: $256
I tried to find the best demand tariff plan available. It wasn’t easy because the SolarQuotes Retailer Comparison tool is still under development and can’t handle demand tariffs. (I told it not to feel bad about this because no comparison tool I know of handles them well.)
The best one I found was Alinta Energy’s Home Deal — Demand Single Rate & Solar. While there may be a better plan out there, I’m confident this is among the best available.
Once again, there are demand tariff plans with higher solar feed-in tariffs, but only for systems of 10 kilowatts or less.
The details of the Alinta plan are:
- Daily supply charge: 139.07 cents
- Per kilowatt-hour charge: 16.31 cents
- Solar feed-in tariff: 11 cents
There’s also a monthly demand charge equal to the highest grid kilowatt-hour consumption in any half-hour period from 4-9 pm multiplied by $17.08. So if the most grid energy a household used in half an hour2 during the peak period was 2 kilowatt-hours, they would have a demand charge of $34.16 for that month added to their bill. If they averaged that amount every month it would come to $410 a year.
If we assume the battery allows the household to avoid demand charges entirely, then their annual electricity bill would come to:
- $256
The demand tariff had a lower per kilowatt-hour charge and an identical feed-in tariff, but it still came out more expensive overall due to its higher daily supply charge. So even if the home never has to spend a cent on demand charges, it still ends up worse off than on the best flat tariff I could find.
There are demand tariffs with a considerably lower per kilowatt-hour charge for grid electricity. The lowest I found was 12 cents. But as these also have lower feed-in tariffs, the home would be worse off with them.
Discover VPP Savings: Hard To Say
Terry also joined a VPP or Virtual Power Plant that helped improve the return from his battery. In an ideal situation, this would provide the same return whether his home used a flat tariff or a demand tariff. But in practice, he is likely to receive a lower return from joining a VPP and using a demand tariff than he would with a flat tariff.
This is because VPPs can take control of your battery and send stored energy into the grid whether you want them to or not. Some VPPs will never drain a battery below a certain point — for example, they may always leave at least 20% charge remaining — but Terry joined the Discover Energy VPP and they give themselves the right to drain all the usable energy from Terry’s battery if they want to. If this happens the only way to avoid being hit with a demand charge will be to sit in the dark without using any power – or start up a generator.
If you want the details of the Discover Energy VPP you can find them on our VPP comparison table. It’s impossible to figure out how much joining is likely to save, as it will depend on what happens in electricity markets. It can definitely pay to join a VPP — but only if you don’t mind losing control of your battery.
Why Demand Tariffs + Batteries Aren’t Great
I was surprised at how poorly a demand tariff + battery performed. I thought the zero demand charge assumption would put demand tariffs well ahead. I expected to have to spend a paragraph or two explaining that relying on a battery to always eliminate demand charges was a risky proposition, as bad weather, high consumption, battery breakdowns, or — worst of all — house guests can always result in unexpected and potentially hefty demand charges.
There are a few reasons why a demand tariff plus battery doesn’t result in great savings:
- The main benefit of demand tariffs is the lower per kilowatt-hour charge and solar households normally use much less grid electricity than non-solar homes. This is especially true if they have a battery.
- The benefit of a good feed-in tariff often outweighs the benefit of cheaper grid electricity.
- Unless you have a reliable auto-start generator, it’s not realistic to expect your battery to always allow you to avoid demand charges. Especially if you join a VPP.
It is possible some households with batteries and small solar systems could be better off with demand tariff, but — provided there is room on the roof — they would be better off installing more solar. Once batteries become cheap enough, a battery plus a demand tariff could work well for homes without roofs.
For now, a time-of-use tariff, one that charges different amounts for grid energy depending on the time of day, is likely to be the most cost-effective tariff for homes with a battery. Once home batteries that I consider reliable fall significantly in price, I’ll write about how much a battery plus a time-of-use tariff is likely to save a solar household.
But just because home batteries don’t pay for themselves yet, it doesn’t mean it never makes sense to get one. Terry bought his battery to deal with frequent blackouts in his rural location. Even if the electricity supply is very reliable where you are, you may simply want a battery. After all, it’s not as if all those people driving BMWs bought them to save money.
But if saving money is important to you, don’t believe salespeople who claim demand tariffs are enough to make batteries pay. Carefully go over any calculations they provide. If they manage to convince you, send them to me before you buy anything because I may be able to show you what they got wrong.
Footnotes
- Note: this is Power (kW) – not to be confused with energy (kWh) ↩
- These half-hour periods start on the hour. So the first half-hour period during the peak period starts at 4:00 pm and ends at 4:29 and 59.9999… seconds. The second half-hour period starts at 4:30 and so on. This means it’s possible to reduce your demand charge by spreading a 20 minute period of high electricity consumption across two half-hour periods rather than having it occur entirely in one. ↩
About Ronald Brakels
RSS - Posts
Hey Ronald,
Regarding your comment about “electricity plans from AGL and Momentum that have higher feed-in tariffs, but they are only available for solar systems that are 10 kilowatts or smaller. ”
It’s actually the inverter size that they care about (and hence the max export it could provide) – at least for AGL in NSW with distribution from Endeavour.
I’ve got 13kw of solar panels too and I’m on the AGL Solar Savers plan which is limited to customers with 10kw. This information was told to me by my solar installer, and so I called AGL and provided the solar paper work and signed up without issues.
I have heard of people with over 10 kilowatts of panels getting these plans, but AGL themselves say the limit is 10 kilowatts of panels. It may depend on the area, or perhaps AGL have changed their stance. I will look into it.
UPDATE: I’ve confirmed that in South Australia it’s 10 kilowatts of panels. Of course, I suppose I suppose it’s still possible you might get lucky and be able to get a higher feed-in tariff out of them anyway.
I’m with AGL in NSW on Solar Savers. When I signed up it was 10Kw of inverters. My system 12 Kw.
Nice analysis. I expect that the best TOU tariff would be better still for Terry
I don’t understand how demand charges can be zero.
I was under the impression that a minimum of 1kW would apply per day (M-F and then it’s mulitplied by the number of days in the month regardless how little power was actually used during the demand period. So, if one had only used 500W maximum power during the 4pm-8pm M-F period, it would still be charged at 1kW minimum demand.
Most of the retailers I looked at have pretty wide margins in demand charges.
Some charge $0.00/kw/mth up to $4.99/kW/month
Some retailers charge out a daily demand rate or a monthly demand rate, so one has to be careful how they calculate their demand charge.
Then the actual usage charge (anywhere between 18c – 25c/kWh) on demand tariff plans depends on the retailer. This rate stays the same all day (unlike ToU).
As I understand it, if one were to use 1kW of power averaged out per hour (but derived from the 30min smart meter daa) in the demand period (4-8pm M-F in my distribution area. Then, it would cost anywhere between $0.00 to $4.99 per month for the demand charge. This does not include the actual kWh usage charge.
But if one uses 2kW on one of those days between 4-8pm M-F, then demand charges would be $9.98 in total plus the usage charge. Even though, it happens only on one day, the rest of the month is charged per day at the same power rate.
I know for a fact with my Powerwall, there would be some power coming from the grid 24/7 (it averages nearly 1kWh/day, so about 50W from the grid average all day, but it does reach 0W with excess solar exporting during the day, obviously) (the power fluctuates from 50W upwards when it comes from the grid). The Tesla Powerwall app doesn’t show it because it only shows in 0.1kW increments, it can’t detect 0.05W. So when the app says it’s 0W from the grid (it could be anything up to 50W being drawn from the grid). It is near impossible to get 0W grid import. Unless, you flip the main switch off and then the battery truly works on its own or only when solar is exporting excess back to the grid.
I contemplated going on a demand tariff plan for my house but nah, not worth it.
I’m better off on the ToU plan with Momentum with $0.21 peak rate (7am-11pm M-F) and $0.12c off peak for all other times and I have no controlled loads as I have a solar hot water system. FiT is 7c.
Correct me if I’m wrong how demand charges are worked out.
3 parts that would make up the demand tariff plan.
1. Demand charge (minimum of 1kW/day or month M-F 4pm-8pm), regardless how much actual power was used during that period if it was less the 1kW, or whatever the maximum power was recorded would be multiplied x 30/31days. It would only take one of those days to go over the minimum to be applied to all days.
Say, the battery supplies 4kW no problem for 29 days during the demand period. Then day 30 comes, the battery doesn’t work and now 4kW comes from the grid, that means 30 days will be charged at 4kW each day, even though that was not the case?
2. Usage charge – actual kWh usage
3. Daily supply charge
Minimum demand charges were applied in the past, but I didn’t find any mention of them in the current plans I looked at. But I do recommend checking to be sure no minimum demand charges apply to anyone thinking of getting a demand tariff.
I found it interesting to read that you do not think batteries pay for themselves yet. I bought a 5.3kW solar system with 6.5kW LGCHEM battery in mid 2018 for $13500. Before then we had been spending approx $1700 per year on grid power so I used that figure to calculate my payback period. My thinking at the time was that if I was able to always remain in credit due to a generous feed in tariff I should be able to payback the cost of the system in less than 10 yrs. At that time I was xpexting to be earning a 20c FiT through AGL. We have found that we have been able to achieve a credit for 3 of the 4 quarterly billing periods over the past 3 years since the system was installed so we are on track to payback our cost within 9yrs. However, The plans I can now source in 2021 in the Essential Energy region market have changed for the worst so my payback period is extending because it is hard to find a generous FiT with a low daily supply charge. I am hoping that community batteries and peer-to-peer trading of excess solar might offer me future savings.
A solar system and battery can pay for itself because the good return from the solar system masks the poor return from a battery. If you use our solar and battery calculator:
https://www.solarquotes.com.au/solar-calculator/
It shows a solar system by itself can pay for itself in only a few years in some cases but a battery system can take decades. Since batteries can’t be expected to last that long they are likely to never pay for themselves. Then said, there are reasons other than decreased electricity bills for getting a battery. For example, backup power during blackouts.
Very interesting investigation. I am on wholesale rates through “Powerclub” in Victoria, since middle of last year. Occasionally we get smashed with high prices in the evening peak, as the generators play games with bidding up prices. The price changes can be violent and random. Overall though, it seems we are better off in our use case, as there are long periods of low prices as well. Especially if there is plenty of wind in SA, like today. We can change usage pattern to exploit that. Regarding domestic batteries, certainly not viable yet, but I do wonder if another path to adoption is avoiding price spikes where domestic users are on wholesale price plans. Happy to provide screen shots of typical day usage and price graphs if interested.
Nice assessment. I went through and modelled how a battery would perform with a demand tariff.
As a comparison I also modelled how a demand tariff compares with best regular TOU tariff without a battery. In our case it turns out the demand tariff plan is not much different, about $20/year in it. Just the way it works out for us.
So when I add the battery (and model using our 5-min interval data), it turns out the battery makes very little difference to our peak period demand. Yes a month here and there it does reduce peak period consumption a bit but for most months it doesn’t change things much.
Why?
For exactly the reasons you cite – it only take a day or two of poor solar PV output for the battery to not be able to cover peak period production.
Re Terry;s situation – the VPP he is on offers a ridiculous high FIT (circa 25c/kWh), a FIT that is IMO unsustainable and once his plan’s FIT normalises with the market then the benefits of that plan with dissipate pretty quickly.
Ronald,
I have long been an ardent fan of Vegemite. In fact, I am convinced that Australian civilization as we know it will come to an end if Vegemite were to disappear forever from the shelves of our grocery stores
Have you ever considered adopting a ‘Vegemite with everything’ approach to your diet?
Putting vegemite on anything other than bread or toasted bread is heresy.
Unless, of course, you are treating a vitamin B deficiency in an animal and have to put it on their paw for them to lick off.
But anything else is right out. I don’t make these rules. I just follow them with idiotic zeal.
Put some in the gravy of your next slow cooker
Never!
Vegemite on Iced Finger Buns, YUM!!
fresh baked crusty bread works well.
Interesting comments on Queensland Demand pricing. I switched to AGL’s Solar Saver plan in Feb 2021 and without consultation AGL immediately switched me to Demand Pricing. They would not respond to my questions until I went to the Energy Ombudsman. AGL then stated it was mandated by Energex. Checked with Energex – Demand Pricing is not mandated until 1 July 2021. It is then mandated for ALL customers with smart meters from 1 July 2021.
After some argy bargy, AGL finally switched me back to Single Rate until 30 June. I then tried to analyse the cost of Demand Pricing after having had 24 days Demand Pricing data from 4pm to 8pm and found the following:
• Demand Pricing takes the HIGHEST power usage in ANY 30-minute period in the Demand Pricing time slot, doubles it and applies the Demand Pricing Tariff per kWh to determine the surcharge it will apply to EVERY day in the billing period.
• Often my solar eliminated or reduced any charge between 4pm & 5.30pm (that was summer, would be shorter period in winter)
• Over a 21-days (x 8 = 168 usage periods) the 30-minute Demand Charge intervals showed
o one period at 0.807 kW the next 0.551.
o average of the top 10 = 0.431.
o average of 104 usage periods over 0.1 kW = 0.171
o 40 usage periods under 0.1kW = 0.034
o 24 Periods had zero usage due to solar.
o average over the whole 168 usage periods 0.146
• So, the one 0.807 kw 30-minute usage = 1.614 kWh @ Demand Pricing tariff of $0.26444 = $0.43 applied each day in the 90-day billing period = $38.70 surcharge on the bill.
• I quickly worked out the best plan under Demand pricing as applied by Qld Energex is to adopt the shortest billing period possible (monthly with AGL Qld) as the Demand Pricing surcharge can only every increase in any billing period – it cannot decrease.
I’m retired so I can minimise my Demand Pricing usage by cooking my main meal at midday not evening. I would not like to be supporting a family with a working wife & 3 school age kids.
I understand the need for Demand Pricing, but I think it is too punitive to simply take the single highest usage 30-minute Demand Pricing period to determine the Daily Surcharge for the whole billing period. It may have occurred due to extreme events (eg extreme weather) or exigent/emergency circumstances (health etc). It would be more equitable to (say) take the average of the top 6 or top 5%.
Just a minor correction on your demand calculations John. The demand charge is calculated on the highest usage each calendar month (or part month if your bill date is during the month) not for the whole billing period. So for example if you are on quarterly billing and the bill covered March, April and May you would have one demand charge for March based on the highest usage in March multiplied by 31 days, another for April (so highest for that month multiplied by 30 days) and a final one for May.
As you note though soon everyone with a smart meter in QLD will be on demand charging (or another TOU plan) and flat rates will not be an option. Smart meters aren’t that common in QLD yet but since they are now compulsory for any new houses, meter changes and many solar installations it wont be too long before the numbers are substantial.
Terry, you may be right & I hope you are. I would like to know your source. My AGL Demand Pricing plan states
Demand price (2) c/KW/day 24.04 (ex GST) 26.444 (incl GST)
Your bills will show the GST exclusive rates and GST will be added
to the totals and appear as a separate line item.
(1). ……
(2). Maximum kilowatt demand measured as a single peak over a
30 minute period between 4pm and 8pm on workdays during
the billing period. Workdays are weekdays but exclude
government public holidays. For the first 12 months on this
tariff, customers with an annual usage less than 10 MWh will
have their demand capped at 5 kW. Customers must have a
Type 1-4 meter to access this tariff.
There is no mention of Monthly periods. In my discussions with the AGL Resolutions Manager, Energex & The Energy Ombudsman no one disputed my calculations of the whole of billing period Demand charge. The AGL Resolutions Manager agreed with me that monthly billing was best to minimise Demand Pricing.
Interesting when I asked for my solar refund to be auto paid monthly I was advised we only refund at the end of the 3 month billing cycle even if on monthly billing. Conversely when I asked whether monthly debit bills could be accumulated and paid at the end of the 3 month billing cycle – no way!!
John, my source for them monthly billing of the demand charges comes firstly from the Energex document that details the demand charges (“Energex Network Tariff Guide 2021-22”) which specifies that the charge is calculated as “$/kW/month”. I have also been on a demand plan for several months and that is exactly how it is calculated. For example I have a bill covering the 6 weeks from 18/4/21 to 29/5/21. On that bill I have one demand charge for the 13 days from 18/4/21 to 30/4/21 and then a second charge for the period from 1/5/21 to 29/5/21. Previous bill to this one had a charge for March then a charge for the first 17 days in April, etc.
Note that on the latest bill we only exported 927kWh and imported 1129 kWh but I still have a credit for the period due to the fact I am receiving more for the feed-in that what I pay for the usage and my demand charge was minimal.
My actual rates from Discover Energy are:
Daily Supply = 96.8c
Usage = 14.88c/kWh
Feed-in=25c/kWh
Demand charge = 16.5c/kW/day
Plugging those figures into your example i would end up with a credit of around $1818 for the year or roughly $2000 savings compared to flat tariff without the battery giving a payback period for battery of around 5 years. As Alex pointed out this is partly due to the big FIT that Discover Energy currently pays and that may not last but I am still confident that in MY case the combination of battery with demand tariff and VPP will pay the battery off (but it probably wouldn’t be the case if I didn’t have an EV and therefore used less power).
With regards to the comment that a battery cannot save you from a demand tariff fee if the weather is bad that is not correct (or at least not if you have control over the battery charging). I simply have my system set up to charge the battery from the grid between 2pm and 4pm if the battery is at less than 90% charge at 2pm. That way I always have at least 90% available when the demand period starts.
The note about the VPP and demand tariff possibly being an issue is somewhat true but so far it hasn’t cost me anything as most demand events either happen early morning (around 7am to 8am) or in the evenings and the maximum that can be drawn from my battery is around 2.5kWh in any 30 minute period (as battery can’t put out more than that). So even if Discover Energy took power from me for 90 minutes (usually they only draw for around 30 minutes) I should still have enough battery remaining to get me to 9pm (or if I don’t the demand charge should still be fairly low in comparison to what I have been paid for the power).
I use ground source heat pump heating. 2 Powerwall batteries and sufficient panels enable the house and heating to operate from battery through the peak period. Any imported power is charged at shoulder or off peak tariff.
So any answer to the question must be specific to the location and its demand.
It would be useful to make an analysis of the 250 odd “offers” for power from the multitude of suppliers. It’s almost impossible to competently compare the “offers”, since each “offer” includes special deals usually short term.. Moreover it is impossible to analyse the benefits or otherwise of time of day tariffs unless the user has access to nearly continuous demand data. In my case, the data available from Tesla’s App allowed me to analyse production, export, import and power consumption at 5 minute intervals for 365 days, facilitating a change from a single tariff to time of day metering, with savings.
A huge and of worms DONT OPEN
Ronald, maybe a little off topic but I cannot find a more relevant part of the blog in general so if you can move this to a more relevant place (not the bin please) that would be great!
I have read with interest a lot of material about recharging Powerwall 2’s from a generator (ok an alternator) and the main responses focus on advising that it is not possible and that the system operates by using the available power in the Powerwall 2 then switching over to an available suitable standby power unit.
I get all this and I can see that having a mains connection and a standby power source wired together could be problematic to say the least.
As we have an Enphase based PV system and a Powerwall 2 it follows that the Powerwall 2 and Enphase inverters see the mains and synchronise with it and the PV and mains and Powerwall 2 share duties all well and good.
However when we get to a period of poor solar output or if we have a prolonged mains outage the offered solution is as stated above is to run the Powerwall 2 to its set minimum capacity and then switch to the standby unit until mains is restored.
All well and good but operating a standby unit throughout the night for minimal loads is a pretty poor use of resources hence my desire to be able to charge the Powerwall 2 from a standby unit.
I am interested to know if in the event of a mains outage the mains connection could be disconnected and an alternative high quality suitable standby unit connected and the Powerwall 2 would then see the standby unit as mains for all practical purposes.
I can hear the wailing so lets be clear the standby unit would be an inverter unit producing clean high quality AC power large enough (6-8kW) to handle the charging current for the Powerwall 2 which is about 3 kW from my observations as I have switched to 100% backup and watched the Powerwall 2 fully charge itself from the mains.
A small load would be provided across the standby unit’s output so that it would be operating in a stable state prior to connection to the Powerwall 2.
I would envisage operating the standby unit on a timer to prevent issues which might arise if the Powerwall 2 became fully charged and would plan to use the charging system at night to prevent issues that might arise if the PV system came to life and produced a lot of power that the Powerwall may not be able to manage in concert with the standby units output.
I have seen elsewhere that Tesla car owners have been able to charge their vehicles from a relatively small “suitable standby unit” and suspect that much of the architecture in these systems is shared.
As I said earlier this question comes up often but I have not seen it articulated as above and hopefully wiser people may be in a position to comment?