We have a Vaillant Arotherm ASHP, one of the first out of the factory as it happens, and I’m keen to get some insight into COP and power usage etc using OEM.
I’ve worked out that I’ll need an emonPi and would like an emonTX for additional room temperature monitoring, potentially also with an external temp probe. What I’m trying to work out is the most cost effective way of monitoring flow and return temps to get acceptable COP calculations.
I’ve read the page about heat meters, however at the moment there’s not one plumbed into the system, and they all seem pretty costly. So I’m trying to work out if I could (for example) mount a couple of thermal probes (DS18B20?) on the flow and return with suitable insulation, or whether the results would be so far off as to be useless.
Basically I’m a complete newb at this so any advice would be very much welcomed!
Hello @gmason and welcome! Great to hear that you are planning to monitor an Arotherm I’d be interested to see what your graphs look like as I’ve been trying to diagnose an issue with one and haven’t got to the bottom of it yet.
It doesn’t give you accurate COP, but you can calculate an indication of what the COP should be based on the carnot equation (this is integrated in the emoncms heat pump app but should be taken with a bit of a grain of salt as it’s just a theoretical estimate and doesnt take into account things like defrosts etc). You cant really beat a good heat meter for accurate COP calculation.
An emonPi with an RJ45 terminal breakout board with the DS18B20 temperature sensors connected to this should do what you need, and emonTH room temperature nodes.
Thanks @TrystanLea - happy to be a guinea pig! How much error do you think having DIY heat sensing could introduce? I’ve seen some heat meters more agreeably priced than the first one I looked at so I could be persuaded, but not in winter. I don’t think the rest of the family would appreciate my thirst for data when the heating was off at this time of year
Also is it worth monitoring flow/return on the heating circuit or just flow/return to the AHSP?
Without doing some manual stuff I understand you can have:
EmonPi - 6x DS18B20
EmonTx - 3x DS18B20
I moved my emonPi to the “boiler room” (it used to house 2x LPG boilers) where the ASHP flow/return enter the building and the DHW cylinder and buffer tank are also installed.
My cylinder has 3 “pockets” into which I can pop the temperature sensor then wedge the wire in place with something appropriate.
The sensors on the pipes are double wire-tied to the bare pipes under insulation. There’s some instructions somewhere on here about doing it really properly with heat paste and metal tape but the simple method seems adequate for my needs.
For room & outside temperature you could get an EmonTH plus a DS18B20, wire it in and put the emonTH in a room with the wired sensor outside through a window? I have an EmonTH outside in a waterproof enclosure. I also have 3 others inside doing room temperatures, radiator temperatures and fish tank temperature.
The limitation is the time taken for the sensors to return their data, which interferes with continuously collecting the voltage and current data in the emonTx and emonPi. If you only want temperatures, a custom sketch in an emonTH - powered by 5 V USB because of battery life considerations - would give “unlimited” (not really, but a lot of) 1-wire temperature sensors.
Ah interesting so it seems I can do without the additional emonTX and just use the inputs on the emonPi, this is good news. I too would mount it in my “boiler room” - in my case custom made for the ASHP installation. So just one emonTX in the lounge with an external sensor through the window frame as suggested.
The approach with the sensors (cable ties, insulation, etc. etc.) sounds about right for me. I’ve got some thermal paste around here somewhere which could also help.
Interesting idea on the fish tank too, I have one of those also! Maybe I will get that second emonTX after all.
This would be amazing @Robert.Wall, are there any instructions around?
I could bring my outside emonTh back in from the cold and it could take over doing the temps the emonPi does plus a few more via the 2nd breakout board I already have. I would definitely want 8 temperature sensors, maybe a few more?
If you lose a sensor (it fails to respond on power-up), then the order that the sensors are detected, therefore the Inputs to which they are allocated, changes. I’d call that a big no-no. It will be far better to hard-code the sensor addresses, the order would then be preserved irrespective. But you’d need to edit and reload the sketch if you replaced a sensor. But that’s probably better than shuffling all the data back into the correct places.
The message sent by radio could exceed the maximum allowed by the library we use. The sketch sends 8 bytes plus 2 for each external DS18B20, and you’re allowed 253 total (though I’ve never tested that many). You’re clearly OK with 8 sensors.
“Star” wiring, with all sensors coming back to a central point - the breakout board, might be OK with short leads, but the recommended method is ‘daisy-chain’, with the bus wiring running from sensor to sensor, and only a short (a few inches) stub between the bus and the sensor itself.
I’d recommend starting again with a modern sketch for the emonTH V2 and expand it to take the required number of sensors.
I’d probably set that in my mind because the DIP switches allow setting from 23 to 26.
Reading more following your question I can now see “Up to 30 emonTHs can theoretically connect to a single emonBase / emonPi. A USB to UART cable and Arduino IDE can be used to set additional unique node IDs by changing the nodeID variable at the beginning of the sketch.”
So maybe I will treat myself to a V2 and include that change when modifying for multiple sensors!
The published Node IDs are purely a recommendation - they mean The Shop can despatch units “ready to go”. You can have anything you like in the range 1 - 30, provided you don’t have two nodes with the same ID, and you change emonhub.conf accordingly.
There are quite a few factors that go into how much error there can be, while you can get some surprisingly good results just doing a simple carnot equation calculation with flow and outside temperatures you really cant beat real heat metering with a heat meter. I know some are having good results with these Sontex units Sontex Superstatic Heat Meter | UK Distributor – Tagged "Superstatic 789"– Stockshed® I helped someone connect one of these up recently, but we are not sure how long it will last on battery power at 10s readings and the manufacturer and supplier dont seem to be forthcoming on this information either…