Making sense of TOU modes

[Dave101]

Hi all, a newbie here!

I have been using solar since the first UK FIT scheme in 2011. I have 2 arrays and typically generate just over 8,000KWh per annum. My export tariff is based on 50% deemed export at 3.82p per KWh (ie about £150 per annum).

I had a Powerwall with the new Gateway installed last month and took the opportunity to update the gas system at the same time to include an immersion heater in a more efficient cylinder.

I am really pleased with everything so far and have not used any gas or electricity since the installation. I suspect that will be the case for several months to come. I am still exporting a lot as well and I like the look of the new Outgoing Octopus rate.

However, I am looking ahead to Winter, when there will be a different need. We have a conservatory with 2 Fischer storage heaters which draw up to 2.7KW total.

I am considering switching to the Octopus Go EV rate – 5p per hour from 00:30 to 04:30 each day.

In my ideal winter world (I think!) I would use this 4 hour window to run the storage heaters and get them hot for the next day and also to top up the battery ready for the next day. I recognise that this may mean that I over fill the battery some days, but Outgoing Octupus means they would buy surplus energy back at 5.5p. I am aware that I would forego my FiT export component which is at a lower rate anyway. I think I can make sense of this calculation through trial and error.

However, my question is not really about my logic in trying to achieve this. I have been trying to make sense of how I would use the Tesla App to make this happen.

I think that I would have to use Cost Saving mode and, in the late evening, set the Reserve to a percentage close to the State of Charge at that time. I would set the strorage heater timers to come on at 00:30 and off before 04:30 and anticipate that they would then draw from the grid as the battery is below the reserve figure, so that I am using the 5p rate to maximise their heat capacity. They can then trickle charge through the day, probably with a boost in the afternoon.

I am also assuming that, because I am “cost saving” mode, the battery would take the opportunity to get back above the reserve level by charging from the grid. Is this a reasonable assumption? I have not convinced myself yet!

I would not be surprised if my thinking is flawed, as I have found it difficult to understand the way the different modes are intended to work.

Any views, comments would be very welcome

Many thanks

 

[JohnRatsey]

Welcome. Here's an explanation of the ToU different modes.

I'm also in the UK and have been using a PW for about 15 months so that I get more benefit from a nominal 4kW array. I'm already with Octopus and aware of the Octopus Go tariff (I'm planning to enquire further) but wasn't aware of the Outgoing tariff. At the moment I'm on an Economy 7 tariff and last winter the PW would charge using off-peak electricity when it thought appropriate. However, Tesla's charging algorithm is based purely on hindsight as, evidently, the programmers live somewhere which doesn't have changeable weather. Consequently on some nights, after a couple of days of poor solar generation, the PW would fully charge itself even though a sunny day was expected. And sometimes the opposite would happen and it wouldn't charge in anticipation of the weather turning murky and there was no way I could tell it to do so. Tesla UK support claimed to have passed on my request for better user control over the charging and other people here have also had info that something is in the pipeline.

Based on my observations last winter, four hours is just about long enough for the battery to charge up from the mains (the charge rate seems to be about 3.2kW). You can see my PWs behavour here if you click on the right-most blob below the date (to reveal the extended data). It rarely got close to empty and there was clearly some thinking about how much to charge by.

If Tesla let us program our PWs to export then the Agile version of the Outgoing tariff would be if considerable interest, particularly during the summer when it would be possible to export energy profitably during the evening peak in the knowledge that it can be bought back at lower cost early the next day. There would be some, but less, potential to do this during the winter depending on how much charge is remaining in the battery. The country needs the extra storage on the grid but we need Tesla to give us the control functionality. BTW, Telsa's 10 year warranty is equivalent to one charge-discharge cycle per day and the most profitable operation could exceed this.

By the way, my solution for conservatory heating was an air-source heat pump which uses around 500W of power to push out about 2kW of heat. And, given that it's an air conditioner running backwards, it can also be used for cooling when the conservatory gets too tropical.

 

[mikemillar]

From what I have read on Powerwall forums, I'm rather hoping that my Powerwall 2/Gateway 2 combination, when it is installed early in August, will have the ability to exercise some manual control over night charging, presumably by adjusting the backup reserve level. I'll be rather dissapointed if it doesn't have that facility. I was advised by the installer's surveyor that, at a Tesla seminar he attended in February, he was told they either had or were about to introduce forward solar radiation forecasting into the control system, something I can do manually if I can exercise some control over night charging. However, recent posts suggest that Tesla haven't yet delivered on that. Could it be that most of you can't control it manually because you have the non-backup Gateway 1?

 

[dandrewk]

In the USA at least, it's easy to adjust the reserve level. Only caveat - it can take up to an hour for the change to take effect.

 

[NuShrike]

In the USA at least, it's easy to adjust the reserve level. Only caveat - it can take up to an hour for the change to take effect.

Mode changes take 30 minutes.

From my recent measurements, reserve changes should take between 15 to 10 minutes, or faster. From the Australian videos, practically immediately.

 

[dandrewk]

I posted about this (I think) in the PW FB group. I changed from backup to self-powered mode, and it took about an hour. Most who posted said it takes anywhere from zero to 60 minutes. Others said it updates at the same time every hour, so that could explain the variance. I've not tested this myself.

 

[cwied]

In my experience the Powerwall updates from the network backend at the same time every hour. If you make the request directly from the Powerwall over the LAN, though, the changes are instantaneous. Unfortunately the setup wizard doesn't support configuring TBC, so I don't know of a way to do TBC changes locally.

 

[JohnRatsey]

I should add that I don't have the backup gateway which has only become available in the UK very recently. Given the very infrequent occurrence of power outages in my urban location it's difficult to justify changing the gateway although I would have probably opted for the backup gateway had it been available when my PW was installed.

 

[mikemillar]

Thanks Guys. Is the reserve level setting available in Advanced cost saving mode? That's the most appropriate for my situation. Given the very variable UK weather it means rather a lot of manual inputs, but I'm retired, so hey ho, something to stop me getting bored! Upping the reserve level for a bad next day forecast should prevent running out of battery during the peak rate time. It wouldn't give any control over the gateway ordering too much night cheap rate charge when it's going to be a good day. I'll report how it works out once installed.

I didn't have the option on the gateway and it bumped up the cost by £1000. As John says, most of us don't get too many power outages in UK so difficult to justify the extra cost on those grounds, but more control is a definite advantage in tariff cost saving terms.

 

[Dave101]

Thanks for all of the helpful replies.

I have the new Gateway 2 and it does have the reserve level setting available in Advanced cost saving mode.

It is much clearer to me now how the Cost Saving mode works, even if it is not optimised for UK weather. I suspect that I will need to see how it works this winter, see what results that I get and then experiment with the reserve level, including when changes are actioned!.

Hopefully, Tesla will change the App to allow the user to make their own decisions and to not have to rely on the “artificial intelligence” built into the system. Maybe by winter the user functionality will be available and/or the artificial intelligence will be more aware of UK weather.

John, your comments and the links that you sent were incredibly valuable. Thank you. I will also send a request for user control of charging times and see what response I get.

I am on a fixed rate plan for supply until late November, so I will aim to switch to Octopus Go then and get a smart meter installed. Outgoing Octopus looks interesting. I am not sure about the Agile settings as I have not seen any suggestions from Tesla that they will support timed export. As you say John, this would increase wear and tear on the battery.

Have added air-source heat pump to my research for next year. I will see how things go this winter.

Thanks again to you all

 

[mikemillar]

Thanks for that useful info on the reserve level setting, Dave.

Re. heat pumps, I looked into that as well recently for the currently oil fired central heating. Here's a back of envelope running cost calculation.

I use 2100 litres of oil a year, current cost about 46p/l = £966
1 litre of oil =10kWh energy.
Oil to air source heat pump energy ratio is about 3, i.e. 3kWh oil = 1kWh heat pump electricity demand, so 7000kWh annually for the same heat output.
Assume efficiency losses cancel out.
I'm on E7 but Most heating will be during the peak rate period. My peak rate is very high (23.75p/kWh) in order to get a very low night rate, but assume an average 18p/kWh = £1260/year, maybe a bit less with some at night rate use. So economically it currently doesn't work for me, if my assumptions are correct. Also as I have a fairly old house with radiators, they would all have to be changed for bigger ones to provide sufficient heat from the lower temperature water supplied by the heat pump.

I have read horror stories about air source heat pumps resulting in a big increase in electricity bills and my calculation seems to bear that out. I'd love to stop using the nasty climate changing black stuff, but economically and practically it doesn't really make sense for me. If you are building a new energy efficient house with under floor heating, heat pumps make good sense, but questionable for existing old fashioned CH with radiator systems. I have a friend with a lot of solar, Powerwalls and a heat pump for his heated swimming pool and his baseline house consumption is nearly 2 kW, so his heat pump is taking a lot of power to run (look for LC on the PVOutput website).

 

[JohnRatsey]

Continuing on this off-topic heat pump discussion, if most of the electricity used is either solar generated or off-peak power stored in the battery then the economics look somewhat better. In my case it's a single room unit and doesn't add much to the domestic consumption. I am however, planning to investigate a ground source heat pump as a potential replacement for the boiler when it starts showing its age. I have already noted the issue of lower temperature water - the current boiler has an outside temperature sensor and adjusts the water temperature accordingly such that the far end of the system in the house is somewhat tepid. I also wonder if a ground source heat pump can be run in reverse during hot weather when chilled water running through the radiators would provide some modest, but welcome, cooling.

 

[mikemillar]

John, you've made a good point about the solar and battery input. The problems I see are that firstly, unfortunately, solar generation is at it's lowest when we need the most heat, mid winter and in the evening plus early morning (the main time when it can use night rate electricity). Secondly, my estimated 7000kWh per year requirement translates into 40kWh per day if heating is on for 6 months of the year. That would require several Powerwalls to service, in addition to the other house demands. It needs to be assessed on a case by case basis but implies more solar panels and more Powerwalls, which is the reason why my friend with the heated swimming pool has 16kW of solar and 4 Powerwalls!

I also suspect that because heat pump output is less controllable and needs to be on for longer than a boiler, we would tend to maintain a more even temperature in the house, very nice, but it would probably cost more. I know recent heat pump controllers are quite intelligent and can offset time of use and outside air temperature etc. which helps efficiency.

 

[Dave101]

Tried an experiment yesterday evening with a reassuring outcome. When the SOC was around 75%, I set up the timing for 30 minutes of "off peak" to start 2 hours later with a reserve of 90% and in cost-saving mode. The battery started charging when the off peak period started, took the battery through the reserve amount and stopped charging after 30 minutes. I was not sure whether it would stop when it reached the reserve point or after 30 minutes. On this occasion, it ran for the 30 minutes.

All good!

 

[mikemillar]

That sounds promising. I'll try it when my PW2 has been installed!