I just want to check that I’m not doing something dumb. My COP isn’t very good and before I tackle the installers, I need to be sure that I’ve not got something wrong myself.
This feed calculation is starts with the flow temp input.
In case anyone is wondering, the x 70 is because we need to multiply by 4200 (specific heat capacity of water) and divide by 60 (to convert litres/min into litres/second).
The min value statement is because there must be some noise/inaccuracy in the sensors and occasionally the flow can be lower than the return (!).
The formula itself looks fine, though you may have small discrepancies depending on the order the inputs get read by emoncms. How is the data being fed in?
(Specific heat capacity at 45° is actually 4180, and closer to 4000 if it contains glycol, but that won’t make much difference)
A good way to check is to plot flow and return temperatures over time. While the pumps are on with the compressor off, you should see both temperatures fall together. Flow temperature should always be above return, except for defrosts or cooling.
Another sanity check is to plot flow vs. tank temperature during a DHW cycle.
Also check the physical installation of temperature sensors - they should ideally be fixed securely to the copper pipes with thermal paste and well insulated.
The orientation of the flow meter can also be a factor, so check that too.
I’m actually just picking up the sensors built into the LG Therma V heatpump via Modbus. I read the modbus in a Perl script and push into emoncms via the API in one shot, so they ought to be “moment in time”.
I’ve seen other systems that have been under-reporting the heat produced (and therefore COP) because the flow temperature sensor was not fitted correctly by the manufacturer. Doesn’t affect the running of the heatpump, but the calculated performance looks worse that it really is.
The other common cause of an under-performing heatpump is how the controls have been set. For example, setting the weather compensation too low can result in cycling, preventing the heatpump from settling into stable running.
Additional complexity in the design of the central heating system, such as zones, TRVs, by-pass valves, buffers, etc will make it harder to perform well.
If you’re able to share your heatpump data, some seasoned eyeballs may be able to diagnose some potential issues with it…
Ok, with a little fiddling I can now make my emoncms data available publicly. (cloudflare tunnel to a second instance of emoncms which is mirrored from the live system every 10 minutes).
I shall post something in the Applications category shortly.
