Hi @Cathoderay, thanks for starting this. I have almost exactly the same experience as you with multiple tabs open and nowhere to go! I am currently using the information provided by the controller on my Mitsubishi Ecodan 8.5kW - which I understand can be inaccurate. I suppose the problem is that I don't know whether I can trust the COP that I am calculating and so I don't know if my system is therefore set up as efficiently as possible (this is my motivation - I am interested in data, but in reality once I know it is working well I can leave it alone but for the occasional check in!).

The people on here with good data seem to have installed a heat-meter on their systems - I don't have one, and so anything that provides a good alternative would be a good start. Beyond that it seems that there would be ways of doing something straight forward using CT clamps etc, but it would be nice to have something fairly simply laid out - as you say, easy to install and use by someone with average competence.

Cheers

Gavin

From the point of view of optimising your system, accuracy would be nice, but repeatability is more important. Take the LWT measurement for example, if say on your system the indicated temperature is actually 1C higher than the true value. Whether you know this to be a fact or not, provided that the difference is a consistent 1C over the operating range, it will still allow any actual change to be quantified.

Obviously measurement of COP involves several parameters, but if the response is consistent, even if it may not be accurate, it will still make it possible to quantify any improvements in COP that may occur by improving system operation.

@Derek-M - good point, but I get COP envy....

More seriously though, I suppose, having a forum like this and enabling 'peer to peer' learning with tech that is rapidly increasing in use, we ought to be able to discuss differences in performance. Especially given experiences of quality varying with different installers. Without the existence of 'smart' heat pumps, I want to make sure that I am maximising my investment. I am tweaking and checking based on previous COPs of my system, but I am still wondering why with low flow temps (29 degrees when I checked earlier today, against an outside temp of 7 degrees) my COP is around 3.2 vs others reporting up to 4.  

Sounds like you need a Data Acquisition System or even a Supervisory Control And Data Acquisition (SCADA) System.

Below is a link to a data acquisition system though there are many others, just Google Data Acquisition.

https://cpc.farnell.com/extech-instruments/407001-pro/data-acquisition-software-cable/dp/IN08841?mckv=sOxCjknCt_dc|pcrid|490691434298|kword||match||plid||slid||product|IN08841|pgrid|46440760989|ptaid|pla-369673148082|&CMP=KNC-GUK-CPC-SHOPPING-945672631-46440760989-IN08841&s_kwcid=AL!5616!3!490691434298!!!network}!369673148082!&gclid=CjwKCAiA8OmdBhAgEiwAShr40__p0nB2TsIzKWbkJy-Blzuz2U5Zy4rbi6a8VCl0sT6T1Vjs_65NZxoCeSQQAvD_BwE

I don't have any real details on this systems capabilities, but the software and interface seem cheap enough. Even using cheap sensors I think that you will be talking several hundreds pounds, if not more, to measure and display all the parameters that you have specified.

When it comes to SCADA systems this is more in the realm of building management or industrial system, and hence will no doubt be even more expensive.

One possible way to create a DIY SCADA system could be to use a Micro PLC (Programmable Logic Controller), which may be possible to read and write data via Modbus or Ethernet, from various sensors to a Data Acquisition system and possibly the heat pump controller.

Posted by: @gfletcher

↑

@Derek-M - good point

Derek, I think you are talking about what I would call reliability and validity. Reliable means repeatable, repeated measurements on the same thing in the same state give the same result, but the result may not be valid, ie accurate, eg I measure my height repeatedly using the same tape measure and get the same reliable answer, but what I didn't notice is the tape measure got stretched, and so the result, while reliable, is not valid. If your readings are reliable (but not valid ie not accurate) they become in effect an ordinal scale (place things in order, but don't know absolute value) and so are potentially of use for monitoring the same system (as long as nothing that alters the measurements happens), but you cannot compare with other systems, because you don't actually know what your COP is. All you know is your COP this week is better or worse than last week, but you don't know whether it is better or worse than other systems, or even an arbitrary benchmark like <2 is bad, >4 is good.

Without getting too complicated, another potential problem is your measuring scale, unknown to you, may be a log scale, rather than a linear scale, or some other oddities of measurement may creep in. We really should i suggest aim to get reliable and valid numbers.   

Posted by: @derek-m

↑

Even using cheap sensors I think that you will be talking several hundreds pounds, if not more, to measure and display all the parameters that you have specified.

This I want to avoid, all the more so as the project may (I hope it won't) fail. If i just want to burn several hundred quid, a large bunch of tenners and a lighter are a much quicker simpler solution, and have the added bonus of providing some extra warmth, though the COP may be rather poor.

My current thinking is leaning towards tapping into what the Midea unit already records, via modbus. Something similar is probably available on Samsung units. But I've also got to think about the hardware to do this, I want to avoid spending £££, getting yet more clutter in the house, all for what at the end of the day is a monitoring system. I first have to decide which comes first, the hardware (which means I can measure these parameters) or the parameters (which means I need this hardware). I think it is the latter, ie start with the parameters, then work out what hardware/software I need.  

Posted by: @gfletcher

↑

@Derek-M - good point, but I get COP envy....

More seriously though, I suppose, having a forum like this and enabling 'peer to peer' learning with tech that is rapidly increasing in use, we ought to be able to discuss differences in performance. Especially given experiences of quality varying with different installers. Without the existence of 'smart' heat pumps, I want to make sure that I am maximising my investment. I am tweaking and checking based on previous COPs of my system, but I am still wondering why with low flow temps (29 degrees when I checked earlier today, against an outside temp of 7 degrees) my COP is around 3.2 vs others reporting up to 4.  

Here is a free tip, don't mention COP, just boast about your 29C LWT, then all the envy will be the other way around. 😎 

My field of expertise is Instrumentation and Control Systems, and I quickly realised that for a quiet life, don't provide the Operator with too much information. In industry there are some systems which are so important, that there would be three individual measuring systems all looking at the same parameter, but you could almost guarantee that all three readings would be slightly different. Give the Operator just one of the measurements on his display screen and he is happy, give him all three and he will be continually questioning which is correct.

If it is important to you then I would start by checking the LWT. Where is the sensor located? Is it installed correctly? Is it making good thermal contact with the pipework? Probably the simplest thing to check would be to measure the actual temperature on the pipework coming from the heat pump and comparing this against your reading.

How are you calculating your COP?

Do you have a dedicated electricity meter for your heat pump? Does it supply data to the heat pump controller? Is it correctly configured?

Post back with any further questions.

I shall be following this topic with interest - I had no idea it might be possible to see what a system's real world COP is! I'd be happy if I could just see exactly what y ASHP's electricity consumption is...

(Hopefully - if I can install it! - being remedied by adding a meter with a CT clamp, thanks to help from a friendly electrician forum. But my lack of data on actual performance and value of my system has always been frustrating!) 

Posted by: @derek-m

↑

If it is important to you then I would start by checking the LWT. Where is the sensor located? Is it installed correctly? Is it making good thermal contact with the pipework? Probably the simplest thing to check would be to measure the actual temperature on the pipework coming from the heat pump and comparing this against your reading.

Even 'simple' things like this are far from simple in practice. Who knows where Midea put the LWT sensor, and how it is calibrated. What I can do is, as you say, take temperature measurements from the pipework I can get at, but that too is not problem free. What sensor/measuring device should I use? Where do I put it? At the point where the pipe emerges from the external unit is not promising, awkward access and on a cold wet and windy day like today is almost certainly not going to measure the LWT, just some arbitrary temp somewhere in between the LWT and the ambient temp.

One possibility I have considered is using an RC-4 type temperature data logger (around £25 on ebay), which is what I currently use to log hourly room temps in the kitchen, with an external probe attached to the primary flow in at my plate heat exchanger, and then compare that to the LWT reported on the Midea controller. If there is a consistent relationship, eg the temp at the PHE is always 95% of that showing on the Midea controller, then I can probably assume the Midea reported LWT is good enough. If there isn't a consistent relationship, then I might just shoot myself.    

Posted by: @cathoderay

↑

Posted by: @derek-m

↑

Even using cheap sensors I think that you will be talking several hundreds pounds, if not more, to measure and display all the parameters that you have specified.

This I want to avoid, all the more so as the project may (I hope it won't) fail. If i just want to burn several hundred quid, a large bunch of tenners and a lighter are a much quicker simpler solution, and have the added bonus of providing some extra warmth, though the COP may be rather poor.

My current thinking is leaning towards tapping into what the Midea unit already records, via modbus. Something similar is probably available on Samsung units. But I've also got to think about the hardware to do this, I want to avoid spending £££, getting yet more clutter in the house, all for what at the end of the day is a monitoring system. I first have to decide which comes first, the hardware (which means I can measure these parameters) or the parameters (which means I need this hardware). I think it is the latter, ie start with the parameters, then work out what hardware/software I need.  

You are correct, one of the first questions is what parameters do I need to measure, but even prior to that you need to decide what it is that you are trying to achieve, since that can dictate which parameters need to be measured.

If you want to measure COP with reasonable accuracy, then you would require an electricity power meter of reasonable accuracy, and the means of extracting the data in a suitable format. It may be possible to obtain the Heat Energy Output from the Midea controller, assuming that it is in fact reasonably accurate, and can be obtained in a suitable format. The question then is how to integrate this data to produce a COP calculation, every hour, every day, every week, every month, every year or a combination of some or all of these time periods?

@derek-m - you are on the right trail, it is COP I am after, but it also means I will get some other data I would like, because they are needed for COP calculations, eg energy used (and so bills to be paid). Lastly, I would also like ambient temp, room temp and flow temp, so I can see how COP varies as these parameters change. If I change my room set temp by one degree, what happens to my energy use/COP? In a cold spell, can I justify bumping things up (or letting Homely do that) to maintain set room temps, or does that put COPs into free fall and push energy use through the roof (the two are related, but not the same)? These are very practical day to day things that I would find very useful to know.

Hourly COPS will do very well, I think, and the data with then allow longer periods up to a SCOP, but there is a potential problem when measurements are instantaneous, eg a LWT on the hour may not be a reflection of say the average LWT during the hour. I may have to have a way of averaging LWT.

For energy in, I am lucky enough to have a dedicated accurate meter for the ASHP with a flashing LED. I believe it can provide, through an optical sensor, a good enough stream of energy use data, but I have yet to work out the practicalities, as what to plug in to what. For those without such a meter, I understand a clamp type meter round the cable can also provide the same data.

For energy out, I have two choices: either trust the Midea calculated figure (even though I have no idea how it calculates them) visible in Operational Paramaters on the wired controller (allegedly, what it calls Heat Pump Capacity is Heat Pump Output, and the figures are generally in a credible range eg a moment ago it was hovering around 7.4, but the units used are kW not kWh, and I may also hit the instantaneous value vs average value problem again) and retrieve them using modbus (there's a possible address for this in the registers, but the wording is confusing), or secondly calculate the output from the LWT/RWT/flow rate, all of which do appear to be in the modbus registers. Applying this more generally to other brands (or even my situation if I don't trust the Midea figures), the LWT and RWT can be measured using temperature sensors (but see earlier comments about where and when), but the flow rate is a problem. It almost certainly means an expensive (around £500?) setup, and it is invasive: the measuring device has to be physically inserted into the pipe work, so the flow actually passes through it. I did wonder about ultrasound, which is sometimes used in other situations for similar purposes eg boat speed through the water, blood flow through a blood vessel, but no obvious domestic heating US flow sensor popped out on google.   

With energy in/energy out, I can calculate COP. For the other temps, I can use simple temperature sensors, avoiding the Midea ambient one as it gets distorted by the operation of the pump.

I still have very little idea how i am going to collect all this data and process it, but I am working on it. All contributions very welcome!