emonCMS has options to convert cumulative to Diff in a Graph(see under the graph for the button).
Or the Virtual Feeds options are powerful, all sorts of calculations -eg
Thanks @JustPlaying but I understood that won’t work on emoncms.org when I’m needing to pipe the data into the MyHeatPump App - that requires a ‘proper’ Feed (so not a Virtual Feed either).
I solved the problem by looking back at previous meter readings - I’ve got those logged at one minute intervals so I’m using the average usage over the past 2 minutes rather than the actual instantaneous reading as the "power’ and the latest meter reading as the ‘use’. Some minor niggles with the resolution of the readings (it only reports power to the nearest 30W) which can make the graph a bit ‘choppy’ when the heat pump is on standby but no issue when it’s running and drawing ~2kW.
I’ve put an old 240V coil relay I had lying about (who doesn’t 
) on the control to the 3 way valve and I’m reading that on a digital input on the ESP8266 I’m using to monitor the Sika Flow meter output voltage, so I have a digital input in EMONCMS that tells me wether the ehat pump is on heating or DHW.
3 Likes
I’ve been looking at the feeds on heatpumpmonitor.org and notice that my ‘COP 30 mins’ seems to do slightly random things and mostly shows 0 or a very low number. That doesn;t seem to be happening to other heatpumps. The COP in window seems to give the right figure. Any thoughts what might be causing this?
I’ve noticed too that my room temp trace disappears when I zoom out to a week or beyond in the detailed chart screen - it doesn’t seem to happen for other users, again, any thoughts welcome.
Also, I’m wondering if it would be possible to but the average daily ambient temperature (or better still a calculated degree day figure) as a point on the main History chart? It would make daily comparisons more meaningful.
https://emoncms.org/app/view?name=MyHeatpump&readkey=91f8fe2e4f55c147fa5f51afd4b6b876
Rachel
4 Likes
I’ve seen the same behaviour for my system (Derby, GSHP) - typically shows ‘COP 30 mins’ as 0.00 but occasionally an implausibly high figure. I’d wondered if this was because I only post updates to the Inputs / Feeds every 2 minutes but looks like your data is feeding much more frequently than that Rachel, so must be something else going on.
Looks like the code that drives this view is app/apps/OpenEnergyMonitor/myheatpump/myheatpump.js at master · emoncms/app · GitHub
I’m just gonna drop this link here…
8 Likes
1 Like
Wow! we have truly made it now 
nice work all involved! a nice bump in the number of people visiting the site:
or requests in the last 24h:
7 Likes
2 Likes
Nice work you wonderful people.
My installer is even linking to the article on “the socials” generating even more clicks.
2 Likes
Having come across this data on BBC very interesting to see actual COP being monitored.
However, the “Annual Heat” in the summary table is very wrong and confirmed by looking at the details, and therefore the summary gives the wrong impression of annual heat required by these properties.
I speculate the monitoring period isn’t being correctly converted to annual (and around half the properties have less than 100 days monitoring).
E.g. No. 18 claims just 246 kWh annual heat - but in actuality looking at the details that’s just the heat used in the last 5 days since monitoring began.
Of course anything with just a brief monitoring period will end up with an inaccurate average until a whole year of data is gathered (and with winter giving a poor/high use average figure), but with time should give an impression of how these houses are also performing from a heat loss point of view.
Once the annual averages are properly calculated it would also then be interesting to see kWh/m2/year for each property, even if this is only really known in reliable detail once a whole year of data is acquired. But it would also provide a good glance to check the calcs are generally right and nothing going wrong. E.g. the Passivhaus should end up 15kWh/m2/yr after 1 year, and expect values then ranging up to ~200 for old housing stock.
1 Like
Hi Matt, thank you for your comments.
Yes, it is true that many of systems do not yet have a full year of data, and so the annual kWh values for these are shown in grey. Hovering a mouse over one of these will show you when the data starts from.
I would be interested to hear any ideas of how to make this clearer. Maybe it just needs to be a paler shade of grey? Perhaps a checkbox to only show systems with a whole year of data? There’s currently just 3 systems will full years (mine has just passed its 1-year anniversary!)
I’m not sure if there’s a sensible way to extrapolate partial data over an entire year, without sophisticated modelling for each system. Simply multiplying one month of data by 12 would be wildly out.
The 30 Day COP is intended to allow for comparisons of different systems for the current heating season, which is likely to become less relevant during warmer months. Perhaps I’ll change this to be “heating season” instead of 30 days. (Hover over COP to see elec/heat kWh for this period).
I am planning to add a page that show running costs for each system (using capped tariffs, compared to gas). This will need to show either only systems with full years, or cost during heating season. Maybe average cost per unit of heat will be good? I shall have to think about this some more.
Are there other analyses that non-heatpump owners would like to see?
2 Likes
I’d noticed this. Perhaps a column with YYMM since start?
Is it really Annual or total (if over a year)?
Yes I’d thought that too, as it is the figure quoted on an EPC.
Maybe two columns for the “Annual” use - projected use where there’s less than a year’s data, and actual where there is.
As for cost - with rapidly fluctuating prices, surely this is largely meaningless, whereas ‘cost’ by kWh can at least be converted into a price given today’s (or yesterday’s, or next week’s best guess) tariff.
The heat demand column could be the kWh figure as detailed on the latest EPC for the property, this would highlight how inaccurate the EPC actually is.
Another variable, will be other heating sources (Log Burner, Immersion Heater) which impacts the kWh/m² figure.
Another variable will be DHW strategy (is it all off the heat pump).
Thinking further, other than demonstrating it is possible for a Hp to do the job, the Property information is not overly useful, I’d suggest. The setup (direct, tank, heat exchanger) and how it is controlled might be more relevant to the performance.
[edit]
and perhaps an explanation page is now required with a greater audiance, how to collect data etc. Perhaps some design hints to enable data to be collected?
It is the total kWh for the last 365 days, or less if the system hasn’t been running a full year yet.
The “Heat Demand” is usually from the EPC. My system was predicted to need 22,000 kWh, but I’ve only actually generated 10,500 kWh for the past year.
2 Likes
Perhaps Column heads could be “Electric” / “Heat” and then another column “Period” - e.g. “last 365 days”, with those being less than this greyed out more.
Energy use averaged out over a year anecdotally follows a sine wave peaking mid winter and dipping mid summer to zero (somewhat logical).
If a sine wave is assumed then I think in theory a sine curve and with in annual use could be deduced from a single data point (even 1 day, but 1 week or 1 month would be better) - as frequency is known as annual, and the dip at mid year can be assumed 0 for no heat required. Although won’t be possible to forecast if you only have a summer period with no use.
1 Like
Here’s a quick analysis of running costs for 18 heatpumps listed, ordered by descending SCOP, vs the equivalent cost for a 90% efficient gas boiler (red line), using current the Energy Price Guarantee (as set by UK Government). Analysis may not translate to other countries.
The three darker bars are systems with a complete set of data for the whole year, others are not.
Four paler bars have less than 30 days worth of data, and may change over the year.
Majority of these heatpumps are proving to be cheaper to run than an equivalent gas system.
System #18 is clearly an outlier, and does not perform as well as we’d expect for a heatpump.
Caveat: this is a very small sample, and is heavily skewed towards heatpump owners who have invested in getting good performance from their systems.
The actual running costs will depend on the particular tariff that each homeowner is using, and may be offset by solar PV or other renewable technologies.
5 Likes