General questions - Wiring / Access / COP

[ecylcje]

Hi,

Thanks for the great information. Just in the process of getting a grant-aerona3-r32-17kw-air-source-heat-pump installed and had a few, probably very, silly questions. Planning on using the USR-W610 Wi-Fi/Ethernet to Modbus adapter as my Homeassistant system is in the garage, far away from the ASHP:

• Your guide mentions connecting to ground on port 32, but there is no ground on the W610. I've been told ground isn't required for RS485, so no common ground is required between both units. Am I missing something?
• As I've not yet got the ASHP installed, or even seen it, is there an easy way to get the Cat 5 cable into the side panel without causing damage? Gromits etc? I was hoping to get the installer to do the connections for me? They probably won't like drilling holes!
• Is the information from the modbus interface enough to work out the COP? Or is something else required? (Apologies I'm very new to ASHP etc)

Thanks again,

Chris

[caliston]

Hi Chris,

1. I'm not familiar with the W610 myself, but I think it's galvanically isolated in which case you won't need a ground. If you have a shield on the cable you'd be advised to ground it at either end, but if it's unshielded it should still be ok.
2. Here's my side panel on the outdoor unit: As installed there were two 20mm conduits in there. One carries the 32A feed for the ASHP, the other carries two cat5 cables. My installer used one for the room controller (just one pair) and the other for the Grant relay box (3 wires). For Modbus I just borrowed a spare pair from the room controller cable. Latterly I added a third 25mm conduit in the middle with a 9 way mains control cable (to have the ASHP drive the zone valves, not an external heating controller) and another cat6a (for the tank thermistors). The ASHP only has mounts for two conduits, so this is securely cabletied to the other two. You likely wouldn't need this, but an option just in case.
3. The Modbus isn't quite enough to get COP. It'll tell you (among many other things):
   • The flow and return temperatures of water going out of/into the unit
   • Water pump speed in RPM
   • The instantaneous electrical power consumption in 100W steps Flow minus return gives you deltaT, how many degrees are going into the house, but to get the energy going in you also need to know the flow rate. The pump speed will theoretically tell you that, but that is RPM not litres/min. My install has an analogue flow gauge, so in theory I could calibrate RPM into litres/min, and then I know that X litres/sec at deltaT=Y degC of water or glycol mix with specific heat capacity Z joules/degC/litre gives me X*Y*Z watts of heat going in. And so: COP = (flowrate_lpm/60) (flowtemp_degC - returntemp_degC) SHC / electricpower_watts It won't be super accurate but gives a rough number. Using an external power meter with better resolution than 100W will help. I haven't tried this, but probably should at some point.

[ecylcje]

This is amazing. Thanks so much.

Apologies last question related to COP (might be a bit off topic). Would you recommend getting a flowrate meter installed (can I ask what your one is?) and using the modbus or would you recommend getting a heat meter installed (which I guess does everything the modbus does plus flowrate) ; something like a 'Kamstrup Multical 403 Heat Meter.' and then use m-bus instead?

I just worry adding a Heat Meter or Flowrate meter will add a pressure drop and increase pump energy?

[caliston]

I think this is the flow meter I have: https://www.caleffi.com/international/en-int/catalogue/balancing-valve-flow-meter-132402 You have to pull the pin to make the flow gauge operate - so it's not really amenable to automation.

I suppose it depends how accurate you want it to be. One thought occured to me though: the flow and return are only reported in whole degrees, so if you're measuring a difference it can be quite coarse - eg the difference might be 3C or 4C, not 3.1C or 3.8C. If you're using that to calculate COP when it tips from 3 to 4 it's quite a step.

So I suppose if you want to accurately get COP you either need accurate temperature probes and flow meters, or heat meters to do it all for you. Probably the heat meters would be more accurately calibrated, although they can be pricey.

I'm not a heating engineer so I couldn't say for certain about pressure calculations, but I note the Kamstrup ones are ultrasonic, so there's no impeller or anything that gets in the flow. Even if there was, I wouldn't expect the pump running slightly harder to make a big difference (the pump takes 100-200W, so a 10% increase would only be 10-20W, and presumably some of that is going into the water to become usable heat anyway)