Picking a dry day, I drained the central heating system and fitted the new PWM controlled pump. At the same time I reversed the heat meter direction. Everything went as planned and both seemed to perform as they should. The flow rate recorded from the heat meter matched that reported by the Samsung controller. Noticeably the flow had reduced, presumably modulated by the PWM control, this also reduced the electricty demand of the pump. The difference between flow and return temperatures, was still about 3 degrees C, which might be something to look at in the future as a delta T of 5 degrees c is commonly quoted as usual.
Having changed the heat meter I could now see reliably how my heat pump was performing. Although the outside temperature wasnt cold, I was happy with an average COP of more than 4 over a couple of days.
0 Comments
Having made some enquires online with people who know more about the issues than I, I came down on the side of replacing the pump I had bought, with a replacement that did have PWM control. In fairness I did get differing views about how useful this would be and the decision wasn't clear cut, but at least now I would be able to measure if the PWM pump was more efficient than the pump I already had. I managed to find a Grundfos UPM3 FLEX AS 25-75 130 AZA on eBay at a very good price, this would replace my Grundfos UPM3 AUTO 25-75 130. You can see how easy it is to make a mistake. :-) Another thing the learned online population managed to help me out with was sorting out the issue with my heat meter. In the end it was surprisingly easy to diagnose. I had put the meter on upside down! While I did feel a bit of an idiot, at least it was an easy fix.
I decided to wait until my new pump arrived and then drain down the system and do both jobs at once. Having installed a Sontex 789 heat meter in my pipework, now was the time to use it to monitor how my heat pump was performing. The meter has an MBUS output and by using an MBUS to USB adaptor it could be connected to my EmonPi. The Sontex 789 is a little fiddly to get installed on the EmonPi and it was useful to have had a bit of experience working with Raspberry Pis'. Once I had managed it though, I had my first chart of how things were going. To my surprise it indicated that my heat pump was performing very badly, but that was contrary to fact that it was heating the house using only 700W of power. After some troubleshooting it appeared that the heat meter wasn't measuring flow rate correctly and might need to be taken off and examined. For now however I just created a scaling factor based on the flow rate reported by the Samsung controller, to give a rough idea of what was going on. The red and green lines show the flow and return temperatures of the heat pump, showing a difference of about 3 degrees C once it reaches steady state. The orange line shows the amount of heat produced in Watts and the blue line the amount of electrical power in Watts needed to create it. Dividing the Heat output by the electrical input gives the Coefficient of Performance (COP) of the heat pump at that moment. The purple line at the bottom shows the temperature outside.
My graph shows that with a flow temperature of about 39 C, my COP is about 4. In other words its about 400% efficient when the temperature outside is 9 C This might sound crazy, but its possible because the heat pump isnt using electricity to create heat like an electric heater does. Its using the electricity to move heat, from the outside to the inside of your house. As a preliminary figure, I was very happy with this and it would indicate that if I ran the heating all day at this outdoor temperature my electricity use would be about 24 hours x 0.7 kW = 16.8 kWh. In reality I might not run it all day and some or all of the electrical energy would come from my solar panels. Following the palaver I had getting the heat pump to work, I contacted The Heat Pump Warehouse by email, regarding the problem of the unconnected fan and compensating me for my time diagnosing and correcting it.
On the Tuesday after the weekend, I received a phone call from an Adrian Thompson, Regional Sales Manager for Joule UK. Although I was expecting a call from The Heat Pump Warehouse, I had noticed a Joule sticker on the heat pump so guessed they had supplied it. Instead, as I expected, opening the conversation with an apology for the trouble it had caused me, he instead launched into an aggressive attack, criticising me for not registering the product, not being a qualified heat pump installer, not having a G3 qualification, using antifrost valves instead of antifreeze solution (as per there instructions), not needing to put the central heating on as it was a balmy 10 degrees up where he was and he didnt need it and finally insinuated that I might have removed the fan connector myself! I was a bit taken back by all this, but pointed out that: 1. I didnt need a G3 qualification as I hadn't installed a cylinder 2. that antifrost valves were accepted as a viable alternative to antifreeze by most heat pump installers and manufacturers and anyway there were no Joule instructions supplied with it 3. that I wasnt really bothered about the Joule warranty (which judging by their reviews, isn't worth a great deal anyway-I didnt say that bit) 4. that I was probably a soft Southern B*st who needed excessive warmth. 5. I didnt like being accused of sabotage and fraud He took on my point that I hadnt been supplied with a Joule manual, and I took his that the standard thing to do would be to call them and get an engineer down to fix it. However I did stress it was the weekend and I needed the heating and that I had saved them the cost of an engineers visit. The conclusion to the call was that he said he would get back to me. A couple of days later I had an email from Joule, suggesting that I get in touch with Samsung. Moral of this story: Dont touch Joule UK with a barge pole! It was pointed out to me by a couple of people, that I might need to get access to my pump and other plumbing once the tank was installed. This would be difficult because of the lack of room in my cupboard. The solution was to raise the tank off the floor so I could access things from below it.
A quick mental calculation guestimated that the shelf would need to support approaching 200kg and would need to be pretty meaty, so some timber, concrete screws and a couple of hours produced the following result! After some further research and fault finding, I discovered that my problem with the E912 error wasn't to do with a sticky relay. I had purchased a Grundfos UPM3 AUTO pump because the Samsung ASHP can use a communication signal called PWM (Pulse Width Modulation) to talk directly with the pump and modulate its output very accurately depending on conditions. The problem I had was that this pump doesnt actually support PWM. In my defence, it has the communications plug socket for it and other pumps in the same range do support it, but mine didn't! To get my pump to work properly, I had to change its connections in the Samsung control unit from 1 and 6 to 7 and 8, remove the PWM lead and change a setting in the controller's FSV values. This was value 4051 which should be set to 0 (Not Used). Once I had done this, the error disappeared and the pump turned off when it should :-)
The dilemma now was whether to just suck up the fact I had made a mistake and live with it, or shell out another £100, get the correct pump and put the old one on ebay! The heat pump worked perfectly after starting it up and all the rooms became lovely and warm. It was a sunny day and the 5 kWh battery was full so the heating was on till 11pm and cost nothing to run :-) I decided to turn the heating off when we went to bed. There is an argument that a heat pump should be left on 24/7 for maximum efficiency, but I knew my battery would be totally drained shortly and even if the heat pump was less efficient while starting up in the morning, I would likely have free solar to cover it. I woke at 6am in need of the toilet and to my horror saw an error message on the heat pump controller. Once I was awake I looked it up, E912 a flow switch error. Luckily there was a forum entry on buildhub.org (https://forum.buildhub.org.uk/topic/25596-samsung-gen-6-ashp-e912-flow-switch-off-and-water-pump-on-error-e912/). The error can be caused by a sticky relay on the control board that causes the pump to stay on even when there is no call for hot water. Lo and Behold my pump was on continually. Something else to contact the heat pump supplier about on Monday! I added some CT clamps to monitor the electricity through the heat pumps outdoor and indoor units to gain some insight into how much it was going to cost the run. The OpenEnergyMonitor system didn't let me down. This graph shows my solar output in blue, the heat pumps power usage in orange and the outside temperature in red. Sadly it was a very dull day.
While thinking about my problem overnight, I thought that I might be able to use a USB endoscope on my phone to have a better look at the circuit board. This proved not to be the case and I was just about to unscrew the casing of the circuit board when I realised the plastic top was just clipped on. Removing it revealed the back of the circuit board with labels! My suspect port was labelled CNP901 BLDC-FAN, Eureka!
Connect fan, re-attach covers and then test run again. Bingo. Now to set up the control unit using the wired display. Youtube proved very handy and with the exception of one issue of which way around to enter the required numbers, weather compensation was set and the heating switched on. After draining down the system, I cut into the combi boiler's central heating pipes and reconnected them to the heat pump flow and return pipes. This was not without its problems, with some residual water spoiling a few joints. However with the help of my grumpy mate Mark the pipework was finished in half a day. The rest of the day was spent connecting the wiring to the control unit. and then having a test run. Initially the test run caused an error E911. On looking it up as a low flow rate error, I realised that most of the radiators were turned off so I opened the valves and tried again. This time a E458 error came up. this was an outdoor unit fan error. After a bit of research on the www, I found out where to look for the fan circuitry, the lead from the fan would be connected to the socket labelled CN90 or CN901 and from there I could test the fan with a multimeter. I decided to remove the top cover of the outdoor unit to see what I could. On doing this this problem seemed to be self evident! Amazingly the fan had not been connected to the circuit board !!! The problem I had now was how to be sure where the correct socket was. I located a promising one of the right size but couldnt make out any labelling on it even with the help of a mirror. At this point I resigned myself to a night without central heating and called it a day.
I had now completed as much of the plumbing for the heat pump as I could and it was time to switch the central heating over from the combi boiler to the heat pump. At this point I wasnt intending to switch the hot water side over as because the cylinder I had chosen was pressurised, this could only be done by a 'Gas Safe' registered plumber.
The plan was to drain down the central heating and cut into the 22mm flow and return pipes from the combi boiler. These would then be connected instead to the flow and return pipes from the heat pump. The two heating pipes left attached to the combi would have to be joined and then re-pressurised so that the combi would then still supply hot water. After that the flow and return heating pipes would need filling and the heat pump commissioned and tested. Fingers crossed. Should I have problems doing this, at least I would still be able to use the wood stove and AC unit to heat the living room and kitchen in an emergency. |
AuthorEx Radiographer, Information Analyst, Teacher and Self-builder. Now retired Archives
December 2023
Categories
All
|