4 ms·
i would question whether that is in fact happening. MAYBE scop, for the recent winters where it has not been cold, but unless you have a vastly overdimensioned
by redeeman 4y ago
i would question whether that is in fact happening. MAYBE scop, for the recent winters where it has not been cold, but unless you have a vastly overdimensioned setup for what is done now then, you'll be screwed if a real winter happens again
- brnt 4y agoOverdimensioned: that can be a cause for low COP, and not always a surefire solution. But yes, properly dimensioning to your particular situation is more important than with other heating methods, agree. The best point I think is to target not the coldest winter you'd ever expect, because that means suboptimal runtime the rest of the year. Better is to target optimal runtime for most of the year, and accept horrible performance for a few days every 5 years or so. Modern heatpumps work perfectly fine down to -20C or so, which covers nearly all of Europe. If your winters are around 0C, it really pays to get units with better defrost management. You can use power output ratings; cheaper brands might have a third less output at 0deg, while others barely dip. In colder countries the cheaper units are funnily enough a great option, because once it's cold enough to continuously flow, you don't have to defrost at that often anymore.
- redeeman 4y agothey might work at -20C, but you cannot get ANYWHERE near the output out of them as you can when its hotter. Several of my family members have heatpumps, myself included, and I can tell you that the COP reduces significantly as soon as we hit just 0C outside. my panasonic has a supposed SCOP of 5.1(or 5.2, cant recall), when it hits 0C here, its down to COP of ~2.5 at my best estimates. when it hits -10, the COP is at best 2. I know this because I think around 1300W into it, and it is not able to heat my house more than a 2.5kW resistive heater. this is their 3 year old flagship "nordic heat pump" that they will market as the sole heating solution for houses twice the size of mine, yet can barely do mine when it hits -5-10C :) edit: im not saying heatpumps are bad, they are great, but I dont think they are a great ONLY solution for places that are below 0C for more than tiny peak times. when its 5-10C outside, and you want 22C inside, they are really amazing
- brnt 4y agoLike I said, around 0 you're gonna get the worst performance, as outside temp goes down most models will see they COP actually increase (compared to ~0 C). Of course, delta-T is still important, so it helps hugely if you can have your water temp at 30 C (and still radiate effectively). It takes a carefully planned systems to reach that though (i.e. active ventilation on your radiators, or floor heating), so many will need to get water out at 40 C or even higher, which is going to bring you down from sCOP 4. At -20 COP is going to be 2-2.5. But, for many Europeans, that's not a frequent event. Even so, 200% efficiency is better than <100%. I think many don't put much effort in the radiator side of things. E.g. here gas heating is usually ran at 60-70 degrees, while for that sCOP of 4+ it should be 30-35 C. Roughly, you're going to need a factor 4 more radiating surface. People don't always install that, and then run their heat pump at 50 C and get a COP of 2. Well, no shit :) So, you gotta take a reasonable low point, understand how hot your circuit temp is going to be (the formula for radiators is pretty much the same over radiator brands from those I've used, floor heating idem), account for buffer losses (with floor heating, you can cut out your buffer and gain 0.5 COP just like that), and, if you have an outdoor temp of ~0 C for much of the year, look for a model that comes with a good defrosting strategy and keeps an acceptable efficiency around that temp. sCOP 4 is for the Netherlands, which has mild winters thanks to the Gulf Stream. However, family is in Poland, and there heat pumps are now also being installed everywhere (where below -20 C is expected for weeks each winter) and while you're right that this means an sCOP reduction, it can still be near 3 if you configure the rest of the loop right. And 250-300% efficiency is going to beat resistive heatings 100% or combustive methods each and every time. 2.5 kW sounds severely underdimensioned for a house at -20 C though. That would not cut it for my smallish house in NL, I've calculated ~5 kW at -15 C needed (and don't forget, make sure your radiators can emit this at the T_out!). Typically people would install a 7-10 kW unit for my sort of house.
- redeeman 4y ago> Like I said, around 0 you're gonna get the worst performance, as outside temp goes down most models will see they COP actually increase (compared to ~0 C) This simply makes no sense, and is not to be found in ANY datasheets from any of the big manufacturers. How do you propose this will work? a given volume of air, say 1 cubic meter, has a certain amount of heat at 0C, if you take it down to -5C or -10C, that amount of heat is reduced. The heatpump can only put so much air over the heatsink, and thus the absolute amount of heat visible to the heatpump reduces as temperature goes down. > 2.5 kW sounds severely underdimensioned for a house at -20 C though well yeah, I have of course not installed such a heatpump (with a max draw of ~1500W in plus degrees) as my only heat source, im just trying to convey the output capabilities