Heywood Interconnector Failure Hammers SA Battery Owners On Tesla VPP

Tesla VPP battery cycling

Screen captures from a Tesla VPP customer’s monitoring app. Look at that battery go! Green is battery, grey is grid. Above the line is using battery/grid. Below the line is charging battery/exporting to the grid.

Tesla Powerwall owners taking part in South Australia’s Virtual Power Plant (VPP) project have a continuing exciting ride for a few more days yet.

With the Victoria-South Australia interconnector out of action following storm damage, their batteries will probably see ongoing heavy charge/discharge cycling as the VPP plays its role in keeping South Australia’s grid stable.

As South Australia’s InDaily reported, storms in late January damaged transmission towers near the Heywood interconnector, and on February 3 it was stated repairs could take up to two weeks. At the time of publishing, it’s estimated the fix will be complete on February 14.

A VPP participant contacted SolarQuotes early on in the event to tell us he noticed a big upswing in the number of charge/discharge cycles his battery was experiencing.

He told us his battery was being:

“cycled every day unnecessarily” (from his point of view),

and he is concerned this could be adding to his power bill, because it’s happening:

“even out of solar hours”

(hence his concern about costs).

He said Tesla had advised him that:

“anyone who is part the Tesla VPP plan will have their system used or abused in a similar manner”.

Our informant received the following explanation in an e-mail from Tesla:

“Your battery was charging and discharging in order to support the SA grid during what might have otherwise been an outage across the network. This is normal aggregation behavior and shouldn’t be considered unnecessary – it’s necessary as part of the virtual power plant operation.  Any grid charging the Powerwall experiences as part of the optimization in the VPP will be a credit line item on your power bill (meaning, this is not an additional expense or kW you will pay for).”


“The warranty guarantees 70% energy capacity at the end of 10 years. The activity you’re seeing is still covered by this warranty.”

SolarQuotes asked Tesla Australia to comment, but we haven’t yet had a response.

Our friend’s concern is reasonable, but we’d hope customers on the VPP won’t be hung out to dry.

As we commented in 2016 when the Powerwall 2 was released, the company’s “unlimited cycles” warranty (PDF) is not quite unlimited, but it’s pretty darn good. The warranty is only “unlimited” if the installation is only used to provide either daily usage or backup power for the customer.

Use in a VPP is obviously outside that spec, so the warranty is more limited (but still good), and we worked out that the “37,800 kWh delivered” warranty is 3,200 cycles – about eight years’ worth if cycled once a day.

What’s happening in South Australia at the moment could add up to be more than a “daily cycle”, so it’s fair for owners to be concerned. As our informant noted, this could well be the prevailing condition for users for the whole of the interconnector repair.

“Expect continual  aggressive battery power cycling  if you are a VPP participant for the next two weeks,” he wrote.


“Power generation, particularly solar generated power will have nowhere to  go, so the batteries will be sharing that power around.”

Tesla’s silence isn’t surprising – the company loves being completely in control of its media narrative1 – but it’s odd that with what looks like a good news story in the offing, it’s got nothing to say. At the cost of extra battery cycling (and the company promises customers will still get their warranted 70% capacity at the end of 10 years), the VPP is working as expected. It’s contributing to grid stability in the face of challenging conditions.

It would be interesting, for example, to chat to an engineer about what the challenge looks like from the VPP operations room – what are they learning about power flows between batteries? Is there anything to learn from what is, in practice, an experiment in accelerated aging? We’d love to hear from you, Tesla.


  1. And we’ve been on Tesla’s naughty list for years now
About Richard Chirgwin

Richard Chirgwin is a journalist with more than 30 years' experience covering a wide range of technology topics, including electronics, telecommunications, computing and science.


  1. I think Tesla might be too canny to participate in the day to day musings of the internet.

    • In my experience – as long as you don’t criticise Tesla – they are happy to feed you lots of news and responses. It is all about trying to control the narrative. Just look at how RenewEconomny or OneStepOffTheGrid report Tesla news without ever questioning the claims or figures Tesla present. Their reward: information from Tesla. It’s PR 101. We refuse to play.

      • John Mitchell says

        Well that’s not unique to Tesla when it comes to those two sites. I like the content but I have seen lots of sloppy, unchecked reporting there and in The Driven. They seem to adopt the same attitude with Nissan. OTOH they have some great content like their interview with the Qld Renault Zoe owner.

        I don’t think it’s as bad as say “Wheels” accepting advertising and junkets from Mercedes and then naming the Mercedes EQC “Car of the Year” (when it’s a 5 year old GLS with an electric drive train). Nice vehicle, but a quick fix for Daimler to help with the new Euro emission standards.

      • hmmm –> ” It is all about trying to control the narrative. ”
        There’s a lot of it going around.

  2. John Mitchell says

    The whole idea of a VPP is exactly this. When energy storage is most needed, the grid will utilise the VPP it has at it’s disposal. If you participate in a VPP you know this up front.

    The only thing I’d question is this: “Any grid charging the Powerwall experiences as part of the optimization in the VPP will be a credit line item on your power bill (meaning, this is not an additional expense or kW you will pay for).”

    I assume Tesla must know how much power is being supplied to the PW as part of the VPP and when (whether it’s peak, shoulder or off peak). I guess I’d question whether this takes into account the inefficiency inherent in the charging cycle – ie are they measuring actual kWh @ the meter or charge received?

    One final thing – why is this a Tesla problem in particular? If you bought a Sonnen under the VPP scheme wouldn’t the same thing be happening?

    • hi john, It’s South Australia so tariff period is irrelevant pretty much averages around a 36c-/kWh flat tariff. I agree with your statement why”why is this a Tesla problem” it is a DNSP problem and Tesla are assisting in smoothing out the frequency and voltage that the DNSP can’t!

  3. Hmmm – punter gets a discount in exchange for participating in the VPP

    Then whinges when it’s used as part of the VPP

    We all have choices …..

  4. I’m in this program . For the month of February so far I have grid services out at only 0.1 kWh – so no idea why this person is experiencing this heavy grid service

  5. I have a Power wall 2. No complaints. But could you explain to me why it takes 0.1 KWh off the grid every day,except for the occasional day when it doesn’t? Why does it and why doesn’t it?

    • Is it because the 2 has to “see” the grid to work with it?

      • I sort of understand that, well, probably not really, but that still doesn’t explain why it sometimes DOESNT take anything from the grid.

        • It is not realistic to expect the PW2 to eliminate all grid imports, and in fact, it is very likely that even on the days you think it does not take anything from the grid, because monitoring does not report it, it probably does. Your 0.1kWh is actually a really good result, and totally expected for a range of technical reasons. But if you want to understand, read on.

          Firstly lets state the obvious reason (I am guessing a lot of these do not apply to you, but worth outlining anyway for completeness ) :-
          1. PW2 can only output according to the specs 5kW continuous or 7kW peak. This means that if you loads ever exceed these a bit of power will come from the grid. It is also worth noting, that even if you baseline power might be below these, it is not uncommon for both inductive or resistive loads to significantly exceed this. For example large resistive loads might for a brief period of time exceed the rated capacity, and for example heating elements warm up. But the big 1 comes from inductive motors (eg AC and fridge compressors). As anyone who has run these loads of a battery inverter knows, you might have to get a inverter that is 7 times the rated capacity. So your little 6kW split system (rated cooling capacity), who normally draws a maximum of 2kW of power, might routinely drawn up to 14kW of power for a very short period when the compressor starts. There is no way the PW2 can cover that. But it is only for a very short period of time, so often does not add up to much, but it is there.

          2. Obviously if your battery goes flat, then you will be importing (but I assume you will have eliminated this obvious thing from your consideration).

          3. I am not sure how you measure your 0.1kWh gap. But it is important to note, that both the grid meter and meter in PW2 have accepted accuracy deviation. Any mismatch here (which there is bound to be, on at least some level), is going to result in a small grid import or export. Lets look at PW2 specs. If you have Gateway 1, they claim their meter is accurate to +-1%. Which means if you are using 20kWh a day, that might be 0.2kWh a day alone (hopefully less). Gateway 2 is a little better at +-0.2%, but even that is about half the error you are seeing without considering other things. Plus you need to also add on the error of the grid meter which might be similar and if that aligns the other way, potentially doubles the error.

          But the real thing that most people are not aware of that I suspect might be the biggest contributor is the PW2 can’t magically just offset any loads in the house by discharging. With a grid connected battery, there is no way to do this. All it can do is monitor the power draw to and from the grid, and based on that measurement, set the PW2 to charge or discharge to try and match any grid imports or exports. ie it can’t actually do anything till there has actually been a grid import, or export that you are trying to avoid!!! But this process is never going to be perfect for a whole range of reasons, and thus you will never complete match the actual grid demand and supply. You will only ever approximate it constantly playing a game of catch up to every change in loads or actually supply (ie think solar which is always changing output through the day) within the house. One of the main problems is the process of measuring and then telling the PW2 to charge its output does not happen instantly. So there is always a delay while any charge is measured and PW2 outputs adjusted and in this time there will always be a little grid export/import. Now with PW2 these delays are relatively short, and because power is measured in kWh, and they are relatively small, but they do add up over the day. In fact you reporting 0.1kWh is very good result, and it is not uncommon for typical gaps to be closer to 0.5kWh a day. Individual differences will vary for all sorts of reason, but 1 is different usage patterns. eg, if you have a combination of loads that tend to be small and/or very consistent output with no great fluctuations in power, I would expect your number to be closer to zero, but if you have large loads that are regularly cycling on and off (ie non inverter AC cycling on and off regularly), it would not be surprising to be further from zero. Same if it is day with speckled clud cover where 1 minute solar might be running 100% and next second substantially less. This might also be the reason sometimes you see 0.1kWh and sometimes think it is 0kWh.

          Also note, that there will almost certainly be the exact opposite happening at exactly the same time. Ie while you can see you are importing 0.1kWh you don’t think you need in theory, at practically the same time, it is almost certain that at times you battery will also be discharging to the grid when you don’t need it, and wasting valuable battery power for a messily solar FiT. This all happens for exactly the same reason outline above. eg you battery is running offseting you running your kettle and producing say 2kW to cover that. But then the kettle clicks off, and for a fraction of a second, the battery is still producing 2kW, that now must be exported, till the PW2 sees this and adjusts output. Most people don’t notice this, because they have solar, and the solar is often exporting, so this is very much hidden in the solar FiT numbers, and they assume it is the same.

          In any case, 0.1kWh is a very good result. I would be very happy with that once you understand the technical limitations.

          • Thanks. I take on board that 0.1 is a good result. The rest is way above my understanding. On my own level of cerebration, I would like to say that I would like to have my panels solaring away for me if the world comes to an end and the grid fails. And I understand that if the grid fails, so does my solar. (For some reason)

          • If you have single phase solar, you should be able to keep that running in a blackout with a PW2 if the installer wired it correctly. If you have a 3 phase solar inverter, that will never be able to work in a blackout. If you have micro inverters, you should be able to keep any solar that is on the same phase as the PW2, but you would have to get a special profile that might be hard to get.

  6. John, spot on – One final thing – why is this a Tesla problem in particular? If you bought a Sonnen under the VPP scheme wouldn’t the same thing be happening?
    It is the retailer not the battery manufacturer,
    and for $5000 from Simply Energy to access the battery over 2 years, No Problem,
    They can take all the battery power they need, For the amount they are paying ,, I’ll go book into a motel for the week and watch reruns of Fin & Ronald.

    • Why is is Telsa problem in particular?
      I realise I’m late to the conversation, but I understood this article to be specifically a Telsa/EnergyLocals VPP customer aka “Telsa Energy Plan, ” not the Simply Energy VPP plan.
      The Tesla website doesn’t promote the Simply Energy retailer VPP deal on it’s webpage (probably because there is a choice of competitor batteries with the Simply Energy deal).

      I think the Simply Energy VPP is a better deal financially…. although it’s actually 3 years, they say to allow for installation and set up ;/ (scratch head, yes 1 whole extra year does sound like a long installation and set up time).
      Simply Energy charge a whopping $360 p/a for supply, and the grid charging cost is a maximum of $144 p/a ($432 for 3years). They then give you the 5100 in credit (over three years). Part of that compensates the extra grid use (by the VPP…$432), and the rest helps reduce the cost of installing the battery in the first place. (I haven’t gotten quotes in SA under 8k for installing this). So the Simply Energy VPP does work out a cheaper way (in the long run) to buy a Telsa PW2, working at about 4k (end of day after installation & all credits), even after you factor the different FiT /General Usage Tarrifs offered by each VPP deal, Simply Energy seems the winner.

      Telsa have their own VPP with Energy Locals called the “Tesla Energy plan” and it’s only for 12 months and doesn’t have a supply charge (vs Simply Energy’s is 360p/a), but you don’t get 5100 back in credit either. Tesla Energy Plan pays a total of 29.8c/kWh compensation for discharging the battery, but the consumption general usage rate is charged at 31.13ckWh …. so you’re still out of pocket. I emailed Tesla about this and the response was:
      “you have to consider GST. That figure of 29.8c/kWh is excluding GST with GST it equals 33.1c, meaning you get more that your import tariff credited back to you”
      I didn’t have the energy to reply… “by that logic, you have to consider GST on the general usage charge which would then 34.243c, which still means you get less back than what you are charged.”

      I calculated for the Telsa PW2 battery, at the end of the day after installation financially you are better off with the Simply Energy VPP deal, rather than the Telsa/Energy Locals VPP “Telsa Energy Plan.”
      And the Tesla website – wow, what an clunky experience – woeful website. Just lucky they aren’t relying on the website to sell cars I guess.
      Their page on the “Tesla Energy Plan” aka Telsa/EnergyLocals VPP deal on the Telsa website is even worse. It’s pitiful. The write up is amateur at best. See for yourself https://www.tesla.com/en_AU/tesla-energy-plan

      Simply Energy’s T&C’s are slightly more professional. https://www.simplyenergy.com.au/vpp-terms-and-conditions/

      I wouldn’t enter a VPP deal without expecting there to be some kind of carrot for retailers, so it’s whether or not your think having a battery on a VPP is worth the consequences (and knowing what those consequences are) vs not having a battery (vs having a battery that doesn’t pay for itself).

      I suspect Simply Energy VPP do the same heavy charge/discharge cycling as Tesla VPP but no one on the Simply Energy VPP plan has noticed the monitoring on this to report it yet. Maybe that’s just my pessimistic view, but it would be nice to be wrong on this one.

  7. Anyone who has been watching OpenNEM since the separation and AEMO took over the big batteries would not be surprised by this.

    I think everyone is on a steep learning curve

  8. It’s great that we’re getting this preview of the future and gathering indisputable, Australian (since that seems to matter somehow) proof that NO COAL is needed to keep the lights on, nor interconnectors, and how even a small amount of batteries is helping a renewable-heavy grid stay up in an environment way beyond its current design, even running a baseload-inventing Aluminium smelter. Proof that the new way works, and a small inkling of what it will be capable of once everything’s designed with the modern technology in mind.

  9. “Hey solar panels don’t work in the dark” scomofo

  10. When the Heywood interconnector failed did the Murrylink fail?

    • Ronald Brakels says

      My understanding is the Murrylink is still operating. It’s underground so it’s resistant to bad weather.

  11. paul sinclair-wadham says

    AGL are flogging a similar plan and will give $280 a year for free access to your battery .WHOOPEEE . We know who’s making the MONEY here and wearing out YOUR battery.

  12. David Thrum says

    I think the words “Heywood Interconnector Failure” are somewhat misleading. The Heywood Interconnector did not fail, rather a high voltage transmission link between Heywood and most of the east of Victoria.

    During the outage, the Heywood interconnector has continued to operate, and in fact the area in western Victoria between the damaged transmission link and the SA border has continued to operate using power transmitted from SA across the interconnector. So this part of Victoria (including the Portland aluminium smelter) effectively became a part of the largely islanded SA grid, with power supply continuing across Heywood.

Speak Your Mind


%d bloggers like this: