June 2025 has become the most high return month I’ve seen so far in the #NationalElectricityMarket with #AmberElectric. Between yesterday and the 11,12 June we are up $325 in one month. I can’t see any news stories of failed generators online yet, but something must be up. I would never expect to see $16/kWh prices unless there was a supply issue. Looks like today is going the same way.
I’m doing a repeat of the #HomeAssistant and Amber talk I gave at the programmable conference earlier this year in a few days. These results will make it much more compelling.

Power generation issues continue to impact the Victorian market. Yesterday (12 June) and the day before.
The automation sold into the peak period in the evening and averaged close to $10/kWh. #AmberElectric #HomeAssistant #GreenEnergy

Yesterday, and it looks like again today, the NEM wholesale prices are extremely high in #AmberElectric. The #HomeAssistant automation and #pyscript code I wrote performed well, picking the peak time to start exporting in a timely manner. As a result of yesterday’s export we sold $65 worth of leccy. There must be a generation problem for this to happen. Such events are rare, but they do expedite the ROI on solarPV and battery investment. #ElectricityPrice #GreenEnergy

I wrote some #code to manage my #Sungrow inverter and battery #AmberElectric https://github.com/spmatich/hass-sungrowmodbus-amberelectric

Continuing thread using #AmberElectric now with #homeassistant. The #Modbus integration works well with my #Sungrow SH10RT inverter and Sungrow home battery. The home assistant #AmberElectric integration works well for the most part, but I have noticed it just stops occasionally, which is fixed by a restart of HA.
I have had to create some bespoke code to choose the best sell time for the inverter based on the feed-in forecast from Amber. The code is triggered from automation using pyscript integration. Fortunately the version of HomeAssistant I’m on includes numpy and pandas, so doing some definite integration (trapz) over the leccy forecast price does not require any python module installs. I have also included some additional features not available in the Amber automation.
1. Overnight reserve level, stop exporting power when battery is at a predetermined level to allow for overnight use, (but above minimum reserve level as per Amber app for grid outages). This means I never charge from the grid.
2. Floor price. when calculating sell time in the evening do not sell if the feed in price is below a certain level. This preserves battery cycles for better prices in the future. Floor price is also active as a trigger when exporting from the battery if the price drops while selling
3. Stop exporting power from the battery if a better price is found in the near future. Configurable threshold for the stop trigger.
4. Delay charging the battery In the morning when the feed in price is still positive, and instead export power to the grid. Later if the feed in price is still positive, opt to charge the battery in the middle of the day
I will be giving a talk on this at the Programmable.tech conference in Melbourne and Sydney in March. I will also share the python and automation yaml on gh at some point.