I have a local install of Emoncms monitoring my Heatpump which I brought preconfigured from OpenEnergymonitor. This is displaying total heat and electricity values which match my heat and electricity meters
Dave
There must be some differences in input processing and or feed selection in MyHeatPump, you can use the sync module to sync local Feeds to emoncms.org
@glyn.hudson
Hi Glyn,
I looked at the feeds and they were already synced
local
Emoncms.org
I also looked at the graphs for heatmeter power and energy and again these were identical
Local
Hello @dave_walker is enable daily pre-processor enabled on emoncms.org but not locally? If so this is likely to be the cause of the difference.
If so the lower SPF figure is to do with the way the energy registers on the heat meter do not subtract defrost heat loss and so present a slightly more optimistic picture than the result from integrating the power data which is the approach taken by the daily pre-processor. The daily pre-processor is more accurate measure of net heat delivered.
A difference of 0.1 SPF over the year is also a common result from the effect of defrost heat loss.
That been the case why does the app show a heat lose due to defrosts of
Good spot, that is again because of the limitations of the energy registers. These have a resolution of 1 kWh, whilst the power data integrated energy values have a resolution of 0.001 kWh.
Thanks for the update. I have turned off the prepocessor because I prefer the simplicity of every matching i,e heatmeter display/ Emoncms etc. I know this will give me a slightly optimistic view of days with defrosts but will now give me much more realistic data on days with no defrosts which is the majority as it can be seen that just today it is overstating the lose by over 1000%
@dave_walker you misunderstand me, the pre-processor is the accurate result, can you list the total heat values for the last 5 days from both methods? E.g
| Date | Pre-processor | Energy registers |
|---|---|---|
| Nov 2 | ? | ? |
| Nov 3 | ? | ? |
| Nov 4 | ? | ? |
| Nov 5 | ? | ? |
| Nov 6 | 25.9 kWh | 27.0 kWh |
| Date | Pre-processor | Energy registers |
|---|---|---|
| Nov 2 | 46.6 | 48.0 |
| Nov 3 | 34.1 | 35 |
| Nov 4 | 58.7 | 60 |
| Nov 5 | 54.9 | 56 |
| Nov 6 | 25.9 kWh | 27.0 kWh |
defrosts and other heat lost has varied between 0.149 and 0.197 on those days
Hmm that’s interesting, I was expecting one day to over-read and the other under-read due to rounding effects at 1 kWh resolution, but there’s a mostly consistent difference with the energy register reading higher than the power integral… I will look into this in more detail!
Looking at mine over the last 5 days, this is the result I was expecting, we can see the effect of the 1 kWh energy register resolution. But overall I expect these to track each other over time, e.g over last 7 days the difference is 0.05%.
| Date | From power | Energy registers | Difference |
|---|---|---|---|
| 2024-11-03 00:00:00 | 20.939 | 21 | 0.061 |
| 2024-11-04 00:00:00 | 23.727 | 23 | -0.727 |
| 2024-11-05 00:00:00 | 24.602 | 25 | 0.398 |
| 2024-11-06 00:00:00 | 20.187 | 20 | -0.187 |
| 2024-11-07 00:00:00 | 5.728 | 6 | 0.272 |
| Totals | 95.183 | 95 | -0.183 |
I will try and review how many systems show the extent of the errors that you are seeing @dave_walker because that is certainly a concerning under-read!
If it’s helps, there’s a very small discrepancy between my two instances:
private instance, live data, no daily stats: https://emon.cryosphere.co.uk/ashp
Electricity 7535 kWh, Heat 26969 kWh, SCOP, 3.58
public instance, synced at midnight, with daily stats: https://emoncms.org/cryosphere
Electricity 7548 kWh, Heat 27099 kWh, SCOP 3.59
All my energy feeds are calculated with Power to kWh rather than accumulated by the meter.
