Andre, helped me loads on that thread. My return temp sensor was out by 0.5 and ended up under reporting the COP by 33%. The one thing I did learn is that despite everyone focusing on COP, if you have a low flow temp you have an efficient COP no matter what the Vaillant app report. I was running my Vaillant with a 0.35 Heat Curve but still getting a COP between 3.0-3.5. Andre worked the maths for me and gave me the confidence that the return sensor was incorrect. After installing the OEM kit we proved his calcs. Just a small temp sensor difference was underreporting my COP by 33%.
I can try the same thing again and shut down the unit totally and see if they match, as when the circulation pump is running they donāt but I didnāt leave it for more than 10 mins or so after the compressor stopped. What I do notice is every time the return drops 0.1c the flow will follow it within about 10 secs with the circulation pump on.
Whatās the reasoning behind doing the test with the circulation pump on?
With the pump on and the compressor off, both sensors see water of the same temperature. If the pump were off as well, temperatures would drift apart due to the slighly different thermal environment both sensors see (different positions in the device, one might be better insulated from the external environment than the other). Of course if the heatpump were off for days and you had constant temperatures outside, you would read the same on both sensors as well.
So swapped the sensors today and some interesting results. The gap is now 0.8-0.9c with the pump running the flow now being the hotter reading, a jump of 0.3c over the 0.5-0.6c discrepancy I was getting when in the other position. Why? If the sensors are mismatched by 0.5c why when swapped does it increase by 0.3? Could this increase be due to wire lengths or bad pcb connections?
I took them off again and cleaned the copper pipes and sensor tip in case some conduction issue. Didnāt make any difference.
could you put all your numbers next to each other so I can easily see them together. Specifically:
Flow temperature and return temperature when pump is running and compressor off - both before and after the change you did. Not just the offset between them, but the full numbers.
Then the same numbers but with compressor on since at least 20 minutes.
Weāll take it from there.
It could be that on top of bad sensors there was some insulation trapped below them. Did you observe that?
Ok,
Compressor after about 30 mins run. Flow 31.7c and return 28.8c.
Earlier compressor off while I swapped sensors and cleaned filters. Around 25.81c flow and 25c return.
What Iāve noticed is the flow always seems to be a 0.xx figure and the return a 0.x.
The installer who took them off and let them settle in the plant room which was about 19-20c measured one at 1220 and the other at 1990. If I look at the chart someone on fb posted, I know itās not totally linear but between 20and 25 is 48.9 per degree resistance, that work out with the 30 difference between sensors of 0.6c.
I checked and rechecked the sensors when I took them off, twice today, that no insulation was in the way.
What I do wonder is if it would be possible to get 2 sensors that when settled read the same then fit a variable resistor in series with one to tune the last bit of error out.
Iāve also found todayās hot water cop has jumped to 4.87.
Think Iāve figured it. Iām guessing the return is 0.3c reading hotter than the flow sensor which is reading 0.5c cooler. So when they swap places the difference is 0.8c.
Did another test this weekend with compressor off and in original positions.
Flow 25c
Return 25.6c
So the gap definitely increases in one position against the other.
Getting crazy cops with them in the wide position though. Also seems to have fixed the hot water always beating the heating issue as I know have a decent gap. Iām pretty sure though these new cop figures are an overestimation.
What I need is a calibrated sensor and a pile of these to find a good pair otherwise itās always going to be running and reading wrong.
How much were they and where from? Also without a calibrated one how can you tell they are accurate? Cause if they both read low Iām guessing this will increase energy use? If they both read high Iām guessing it will just mean increasing the weather curve to compensate?
If they both read a bit too low/high this will just impact the weather compensation, nothing else. Only the difference between them matters for heat energy. And only flow temperature is used for the weather curve.
They came and after a bit of push back accepted my idea to put all 4 sensors in the same place indoors and wait for them to balance. Now have 2 that are within about 100ohms of each other. Couldnāt suggest a fix for the humming other than the installer should have used Vaillant flexis. Did say the installer could move the external pipe clamps which I think might also help.
What Iāve noticed is Iāve still got the 0.2c gap which Iām guessing is the wiring resistance possibly. Iām going to cover the sensors in some insulation to see if that helps. If not Iām assuming the sensors goto the main pcb box so might measure the resistance there. Why they donāt put some small trim pots or software adjustment on these I donāt know.
I did find out today that the sensors are exactly the same as the ones used on Vaillant boilers. As thatās what he had with him. You can remove them from the boiler clip and fit them to the heat pump clips.
