Help overcoming view limitations in app vs. web

Data Analysis
1

It would appear there are some large gaps in parity between the data views available in the mobile app VS. the web app. Specifically, there is no ‘power meter’ view per device in the Web App.

This has created a specific challenge for me that I’m hoping someone else has already figured out how to overcome. I was first trying to visualize analyze the device power meter to find and quantify patterns of usage, but doing this on the mobile app is arduously inefficient (scrolling to find large blocks of usage, zooming in enough that the hour:minute start/end times are showing are visible at the bottom, taking multiple screenshots, and then repeating this for every shower and bath taken for the last 20+ months.), and the same view isn’t available on the web.

Given the lack of this view on the web, I’m hoping there is another way to export the data by scraping the web app or API.

I have a hot water heater (on a dedicated clamp CT) that was installed by my landlord in 2023. I know anecdotally that it has been progressively taking longer and longer to re-heat (from around 2hours to around 5 hours in the last 2 years). This topic isn’t about that issue or the cause. Its about my attempt to extrapolate evidence in the form of time series data from SENSE.

Since my primary evidence is linked to the re-heat time after heavier usage (a bath and/or longer shower), I would need to extrapolate from the ‘device’ power meter draw, find the re-heats that were long enough to represent a shower or bath, and show that this time has progressively increased over 2 years for similar usage.

The specific issue I believe I’m observing has resulted in a negligible to non-existent difference in electrical usage when I compare similar months across years. As such, I would need to use the higher resolution power meter data which isn’t available via conventional hourly/daily exports.

What I’m wondering is if someone has found a way to export the higher resolution data per device via the SENSE API or by scraping the API.

I found @kevin1 ‘s 2019 post to be really interesting, but I’m not sure it would apply to my use case, and working with R is far from familiar territory. I also dont have solar. Scraping the web app Power Meter for data analysis - Data Analysis - Sense

There’s also this 2018 post by @frankwin.hooglander, but I suspect its out of date: GitHub - Frankwin/SenseApiPostman: Postman collection for the Sense (Energy Monitor) API · GitHub

My end ‘hopeful’ goal would be finding a way to pull higher resolution exports (by minute, or 5 minute blocks) for all the devices for at least the last year.

Has anyone successfully done this already? I know it falls outside of what SENSE is willing to official support in terms of data export, but I suspect its possible. A strongly suspect that one or more of the many industries people in this community have figured it out.

2

Hi, Surias! Thank you for clearly explaining what you want. If you wanted to scrape data going forward, I would suggest Home Assistant. Since you are trying to scrape historical data, I have nothing to suggest. I don’t know of any way to do that. Sorry.

Hourly data export is the best you can do for historical data. And that might work for what you want, since you mention it takes several hours to re-heat after a shower. Create an hourly export from your CT clamp on the hot water heater. Open that .CSV file in a spreadsheet. Larger kWh values will represent the hot water heater working for more of the hour. There should be a clear maximum when the hot water heater has worked for the full hour.

You should be able to tell when there has been a shower because there will be several successive hours with a large kWh value. You can make such events pop visually via conditional formatting, or use an equation with some kind of IF statement to count groups of these high values. If your hypothesis is correct, you will be able to tell if recent showers are taking 5 hours to reheat versus older showers reheating in 2 hours.

Here is a sample of such a spreadsheet, with an equation in a new column beside the exported data. When there is a shower, the equation reports the word “Shower!”, if not, it says nothing. You would have to modify the criteria in this equation to fit the power of your heater.

shower
shower1688×699 79.2 KB

3

I haven’t tried using it recently, but the informal API might be the best way to go - it’s what Home Assistant uses. But does use a different endpoint for data than the current phone/tablet app, so some of the aggregated data seems to be update far later then the app. It also produces power, not energy measurements so you have to integrate those power samples if you want to see energy.