Circulation pump

[2807]

Hello list

I have 96 x 58mm evacuated tubes together with 160 (soon to be 180) 47mm tubes heating the radiators in my home either directly (when there isn't much energy available) or via a thermal store (when the sun is shining).

At present I have a grundfos 15/50 circulation pump which is not really up to the job I am asking it to do.

Today for instance, there have been sunny intervals & thunderstorms, I have just been out to my shed (where the controls for the system are) & the temperature sensor in the manifold was reading 47 degrees (because the sun was shining) whilst the radiators were still cold to the touch (about 40 meters away from the manifold).

I have looked on the interweb, but am confused.

Can anyone tell me which is the next most powerful pump and the next and the next etc, so I can decide what to replace the existing pump with.

Thanks in advance

2807

[desperate]

15/60, it,s the same size but a bit higher pressure, then an 18/50, then 18/60, can't remember after that. What pipe size are you using? are you sure there is no air in the system?

Desp

[Justme]

A, was the pump running?
B, are the pipes well insulated?
C, what bore size?
D, whilst the array header might have got to 47c it still does not mean that there was any real energy to collect. A few second pumping cold water into the header will soon cool it. The header heat & cool cycle could go on for hours & still no heat makes it to the rads. What is the heat load / loss of all the rads & pipe work? 

[2807]

Hello desperate

Many thanks for the info - exactly what I was after.

I am using 10mm copper for the 1st 10 meters, either side of each set of panels (for ease of fitting), 15mm copper where it is underground for 12 meters in each direction to & from the pool shed roof to the house where 60X58mm together with 100X47mm tubes are situated.  Then 15mm plastic for the plumbing inside the house (again for ease of fitting).  I have also used 15mm plastic pipework (for ecconomic reasons) above ground between another roof where I have the remaining 60X47mm and 36X58mm tubes installed.

I am fairly sure that there is no air in the system, I have (well insulated) bleed valves at every high point in the solar loop & it is plumbed directly to the booster pump which supplies the garden irrigation system from my well, so it is a fairly simple matter to flush the system all the way through, just by opening the pressurising valve then walking around opening each of the bleed valves in turn.

Hello Justme

A. Yes, the circulation pump is currently on override
B. I think so, I have 13mm wall 10mm bore HT armaflex surrounded by 35mm bore 19mm wall O class armaflex round the 10mm copper pipe (I had to split the 35mm bore insulation to get it in place, but the 10mm bore is in one piece) & 13mm wall 15mm bore HT armaflex surrounded by 42mm bore 19mm wall O class armaflex around the 15mm pipe.  Last winter, the radiators in the house were at 60 degrees plus, whilst the external pipework still had frost on the outside of the insulation.
C. See above.
D. I have 2X30 58mm tube manifolds, 2X18 58mm tube manifolds together with 8X20 47mm tube manifolds.  Maybe 17 or 18 liters of water being continuously heated all the time the sun is on the tubes.  Obviously, there is a lot of water in the radiators and other pipework but surely I can extract the energy from those 17 liters which are being heated.  I understand about the losses from warm water standing in dead legs, hence me leaving the circulation pump on override on days like today, when we have thunderstorms intersperced with periods of bright sunshine.

Many thanks for your input.

2807

[Antman]

Are the manifolds in series or parallel?
You state "the temperature sensor in the manifold was reading 47 degrees ". Which manifold is it in and are you sure that the other maifolds were also at 47 degrees?

If in parallel connection, I suspect an airlock in the panel with the sensor fitted.

If series, then you really want all panels in 15mm at least - the 10mm will probably throttle the flow.

The viability for space heating this time of year is for separate discussion...

Antman

[2807]

Hello Antman

The manifolds are plumbed in 4 lots of parallel, 60x58mm, 100x47mm (soon to be 120x47mm), 36x58mm & 60x47mm.  Mainly because of the order in which they were bought & added to the system.

I only have one manifold temperature sensor, it is fitted at the exit end of the 100x47mm tube run.  Unfortunately, at this time of year, that array of tubes is partially shaded until 2.30pm, hence why the circulation pump is left on override (I have suffered from burst plastic pipework due to stagnation in other panels at this time of year.

60 of the 47mm tubes and 36 of the 58mm tubes are mounted vertically & shaded in summer so as not to "over power" the system in the summer.  The 36x58mm tubes being the offenders in the stagnation events of last year & 3 weeks ago.

With all the bleed valves I have in place, plus the ease of bleeding the system, I am fairly confident that an airlock is not the problem.

I am beginning to come round to the 15mm pipework idea, but will try a larger pump as a first (more ecconomical?) measure.

As for the viability for space heating.  I live in SW France, with no mains gas & the only really viable alternative being to light the woodburners (which I will need to supplement the solar in December & January anyway).  I moved here in 2005, since then, my electricity consumption has reduced from 3,000€ to 700€ p/a and my firewood requirements from 12M3 to 3M3.  The horrendous first year lecky bill being caused by using the electric underfloor heating. Like most renewable systems, mine is never going to completely wean me off other energy sources, but for the same cost of an oil fired boiler, I have a system which heats my hot water for 9 months of the year, Hot Tub for 7 months of the year & swimming pool for 5 months of the year whilst costing me very little on an ongoing basis.

2807

[pj]

Hi 2807,

You mention 15mm plastic in your system. If this is of the Speedfit variety, and you are using the SuperSeals, these seals are about 9mm bore. They can represent a significant pinch point to flow calcluations around the whole system, limiting the max flow to quite a low figure. I beleive the Polyplumb variety uses steel inserts - the bore is over 11mm.

[Justme]

I think a good start would be to go round testing the temp of all the headers & the pipe work.

Such a large system with one temp probe is bound to cause issues.

I doubt that low flow is the issue.

On that point have you actually checked that the pump is pumping & not just making a noise?

[2807]

Hello pj

I haven't done any flow calculations, my system has become too big & complicated for that.

I would have thought that 40M of 10mm copper would have already caused enough of a "pinch point".....

Hello again Justme

I have been round with an infa red thermometer, it is difficult without ruining the insulation, but, at the moment all the manifolds seem to be between 35 at the coldest and 76 at the hotest, this tallies with the temperature of the hottest radiator at 72 and the coldest at 36.

From this, I would again conclude that the pump is indeed pumping, but the flow rate is not high enough.

Grundfos have come back to me & suggested a UPS 25/55 pump would be more effective, I have loked on Fleabay & they seem v-expensive - anybody know whether a 25/60 might be OK?

Thanks all for your input, will try a different pump & see what, if any difference it makes.

2807

[Justme]

Still dont think its the pump.

You have hot water flowing out to the rads & cold (ish) water coming back then getting re heated to go round again. The heat being extracted via the rads is exceeding the input heat. Faster flow will just give you colder water going round the system as it wont be in the header long enough to be reheated.

Still sounds to me like insufficient sun or a problem within a header or two. You need more heat input not faster water speed. Faster water will reduce the peak temperature (and raise the lower return temp) but you will still have the same total energy available to heat the rads.

[pj]

I would have thought that 40M of 10mm copper would have already caused enough of a "pinch point".....

Good point, well made...

[Justme]

Just a thought, wont a higher pressure & flow pump just cause more turbulence?

[desperate]

Hi 2807,

I second Justmes post no 9 up up there, but also the difference in manifold temps suggest that there is a different flow rate through your paralleled groups. Running groups of panels in parallel is not as easy as it appears, one needs to consider the pipework resistance and length and tee off layouts in each circuit, just putting a bigger pump in the system may well improve the flow to an acceptable rate in the manifold that reads high, but would also increase the flow to the other manifolds too much. Could you try to control the flow rate in each group with a gate valve, increase the resistance in the colder circuits in relation to the ones that run hot.

My gut feeling is that the pipework may be a bit on the small side but a standard 15/60 pump should do the job, but that is a bit of a guess.

Desperate

[Philip R]

2807,

Get rid of the 10mm microbore. Remember the pump is really a low head circulator and you are running it in a stalled condition ( Not the impellor) but referring to the flow condition where the impellor is inducing excessive turbulence in the water, as it nigh on impossible to shift round the consrticted system.

You do not need a bigger pump, just less pipework restriction. Re-plumb the 10mmm in 15mm, remember the comments on the inserts from PJ, these restict also.

PhilipR

[SteveH]

 I made this mistake too... As Philip R says it's a circulator & not a pump... It produces no significant pressure differential across its input/output. All your problems are down to the restriction in flow over the extended pipe runs you have.

 Just my thoughts.

 Steve.