Posted by Richard Bennett on July 26, 2005, 4:28 pm
My question concerns the insulated return pipe from a solar collector.
I intend to run copper pipe to the hot water tank/heat exchanger but can't
decide what size pipe to use (the anticipated hot water flow will be perhaps
Comparing a 5m run of 15mm against 22mm copper pipe, which is more
efficient (less heat loss)?
22mm pipe that has a 'high' volume to surface area or a 15mm pipe that
has lower volume, relatively higher surface area BUT has a faster transit
I would be most grateful for any suggestions.
Posted by Christian Kaiser on July 27, 2005, 1:24 pm
The smaller the better (less surface, less time in the pipe), BUT there's a
limit to it when it comes to resistance. I have a table at home about these
limits, and I can look it up in the evening. The minimum diameter depends on
the flow and on the length of the pipe, but with 50 l/h and 5 m I think 15
mm is enough.
I have 18mm pipes in my system, which can pump more than 300 l/h over about
25 m of pipe - although I know this is on the limit and I should have used
22 mm to reduce the strain for the pump, but the whole system works as
Posted by iain-3 on July 27, 2005, 11:17 pm
Christian> The smaller the better (less surface, less time in the
Christian> BUT there's a limit to it when it comes to resistance.
I'd think you would want to model your system as having some sort of
SEER ratio: how much heating do you get for the amount of electricity
expended running all the bits (pumps and controllers).
Pipe diameter comes down to trading heat loss against pump power.
The most important variable is probably the length of that return
pipe (5m), and the combined flow rate through all the panels (50 l/hr).
It probably also matters how much insulation you have on the pipe,
and how well installed it is. (Did you tape over the insulation, or
otherwise seal gaps? Is it protected from the weather?)
Finally, systems at different SEER ratios might use different pipe
urethane foam pipe insulation, 40mm outside diameter, 15mm inside, R-6
R-6/inch is .042 W/m^2-C
Rcyl = ln(outer radius/inner radius)/(2*pi*length*conductivity)
Assume a fluid-to-air temperature delta of 40 C, you get
P = 40/0.743 = 53.8 watts lost
urethane foam pipe insulation, 46mm outside diameter, 22mm inside, R-6
P = 71.5 watts lost
So, can you save 17.7 watts from the pump by going to a larger pipe?
If so, use the bigger pipe.
You can also get 22mm thick insulation, which I would seriously
consider using, and I'd use on the outbound pipe as well (if there
is any recirculation going on). It's not clear to me if the O.P.'s
flow rates will save much power going through 22mm pipe versus 15mm.
15mm: 10.4 cm/s <- this is already somewhat slow
22mm: 4.4 cm/s
Posted by Christian Kaiser on July 28, 2005, 12:44 pm
that link is nice - I will calculate my system (although I will for sure NOT
change all the pipes... :)))
Posted by daestrom on July 27, 2005, 10:45 pm
Larger pipe means less pumping work to get the flow. And higher flows
through collecters and heat exchangers means better heat transfer. As far
as heat losses go, put some insulation on the pipe. Putting 5mm of foam
sleeving around it raises the diameters to 25 and 32. That means the
difference in surface area becomes a factor of 1.28
The pressure drop through a pipe varies inversely with the diameter raised
to the *fifth* power. So the difference in pressure drop for (22mm/15mm)^5
= 6.7 times more for the smaller pipe. Mind you, the pressure drop in 5m of
pipe is probably only a tiny fraction of the pressure drop in the whole
But insulation costs are a one time thing (here in the US, foam sleeves for
common pipe sizes are very cheap), whereas pumping power and low flow are
issues that you will have forever.