Water Source Heat Pump Project

[nowty]

A few years ago I fixed a major problem under my house which involved building a reinforced concrete box culvert to slightly divert an underground river.
When fixing the problem I purposely created a catchpit to provide a pool of subterranean water,

(a)   So I could easily pump out and harvest the water.
(b)   To incorporate a heat exchanger for a heatpump as water temp ranges from a high of 14 degrees in summer to a min of 7 degrees in winter.

(a) was done some years ago with over 190,000 litres extracted do far, (averaged 500 litres per day this month for the garden)
 but for (b), I could not find much technical data for a water sourced heat pump so its been on the back burner.

Now I have found some good technical info on the Kensa heatpump website. It seems that for a water sourced heat pump in moving water, there is a 3 litres per minute per kW rule. Although it has not rained here for over two months I have measured a flow rate of around 15 litres per min which suggests I could theoretically extract 5 kW or so. Given that in winter the flow is generally better, maybe 7 or 8 kW is possible. The Kensa website also says one submersible mat with 250m of 32mm HDPE pipe can provide 6kW.

Hmmmmmm, I don’t have the space for 250m of 32mm HDPE and the manhole access is only 600mm wide but what if I used copper ?
Its thermal conductivity is 800 times more than HPDE but then 250 / 800 = 0.3m of HPDE obviously would not work as its insanely short. Therefore I think the much lower thermal conductivity of water needs to be taken into account. Then I found on the EngineeringToolbox website that the overall heat transfer for copper as a heat exchanger in still water is circa 400w / k m2.

Therefore assume a 5 degree temp difference and an 8kW extraction I would need a copper surface area of 8000 / (400 x 5) = 4 m2.

I can buy coils of 25m of 10mm copper microbore pipe, each of which are easy to lower down the manhole. Each one has a surface area of 0.79 m2. So I bought one and I found it fitted easily with enough room for 4 or 5 of them. So that’s a total copper surface area of 3 or 4 m2. Mmmmmmm, its now getting interesting.

Ground source heatpumps seem to be pricey bits if kit, but I found an 8kW discontinued and ex display one currently in a showroom on fleebay being sold by a heatpump company for around £1k. So took a punt and it arrives next week.

Heatpump Spec


Bought the rest of the copper and I have been prototyping the heat exchanger on my lounge floor, much to the dismay of SWMBO.





Obviously more to follow .....................................................

[offthegridandy]

This looks really good Nowty.  I remember following your subterranean adventures.  The copper coils look good.  Is the water clean(ish) or is the copper likely to be attacked by the water?

Keep us informed as you progress looks like a really interesting project.

Andy

[nowty]

The copper coils look good.  Is the water clean(ish) or is the copper likely to be attacked by the water?

The water is clean but it gets very cloudy when stirred up because of a layer of very fine silt. This silt builds slowly builds up in the catchpit, but gets cleared out when we have a heavy rain shower.

I have had a copper pipe connected to a small submersible pump down their for 5 years now for the water harvesting, and apart from a layer of crud on it, its fine.

I have two concerns,

1) The extra friction of the pipes causing more turbulence which may result in a slower max flow rate if we have a deluge.

2) I found a few stones in the catchpit the size of a fist so they must come through with some force, so there is potential for damage to the soft copper.

[offthegridandy]

Can you get some mesh in front of the incoming flow to catch any large debris?

[nowty]

Can you get some mesh in front of the incoming flow to catch any large debris?

Maybe, I will think about getting some stainless steel mesh if I can find some suitable for the job.

[donegal]

Looking good, i cant wait to see how this turns out. You've a lot to live up to after the last project .

Good luck

[titan]

I would think with the copper coils that tight the total surface area would be reduced considerably. Had you considered pumping the water out and through an external heat exchanger, it would mean you would not be size constrained and could have optimum flow rates.

[ecogeorge]

Excuse me if i am missing something here but why are you building a collector (be it mdpe pipe or copper) ? why not just pump the water through your heatpump and return to waste ??
George.

[nowty]

Excuse me if i am missing something here but why are you building a collector (be it mdpe pipe or copper) ? why not just pump the water through your heatpump and return to waste ??
George.

I did consider this as this as an option. i.e. do an an open loop design instead of a traditional closed loop collector. But everything I have read recommends against it unless the water is very clean and deep, even then there are issues.
I would need a large settling tank and more pumps and filtration, even then you can feck your heat pump heat exchanger very quickly with corrosion or blockages. Lots of maintenance.
An open loop design requires considerably more energy to pump out the water.
The water could freeze in the heat exchanger as there is no antifreeze.
If I returned the waste water to the river then licencing with the Environment Agency may be required.

I currently pump small quantities out for harvesting purposes and its not without its problems so attempting to do it on a larger scale and filter it etc, I just don't want to go there.

It would be best to have a direct heat exchange design where the refrigerant gas is circulated in a small pipe in the the water but thats unlikely to be a DIY project.

[nowty]

I would think with the copper coils that tight the total surface area would be reduced considerably. Had you considered pumping the water out and through an external heat exchanger, it would mean you would not be size constrained and could have optimum flow rates.

You could be right about the surface area being reduced, I did originally consider coiling it like a slinky and fixing it to a couple of strips of aluminium, but its just the practicality of doing it, and getting it all down there down a narrow manhole. I am hoping it won't make a huge difference as the thermal conductivity of copper is so huge compared with the water.

If you think about it, imagine it was a solid block of copper with hollow water channels within the block for the water to circulate. I think the most relevant factor would be the surface area of the water channels rather than just the surface area of the copper block exterior.

With regards to simply pumping the water out, I think I have answered it to my reply to ecogeorge.

[titan]

With regards to simply pumping the water out, I think I have answered it to my reply to ecogeorge.

Not quite, I was suggesting through a heat exchanger, closed loop for your HP, with anti freeze, open with no contamination for the cooling water. Also the effective cooling area for the copper coils is that which is in contact with the coolant, the coils will be much less effective in tight coils.

[knighty]

I think they're coiled a bit tight too... not enough room to let water flow between the coils?

could you coul the copper in slinky fashion - but keep the coils tight torture

run a few of lengths of copper/aluminium/stainless/whatever bar the full length and cable tie the coils to it

coils sized to fit down the length of the drain... then post them down? - guess you might have to cable tie the coils down in the hole to let you slide theme

you could rap all 4 coils at the same time, or have one after another... I guess if you're running them in parallel coil them all at the same time, if you're running them in series run them one after another?  (and flow water through them in opposite direction to the water flow through the drain)

put them in the drain pipe flowing away from your box area, then you can add some kind of rock catcher to the box area?

[nowty]

I think they're coiled a bit tight too... not enough room to let water flow between the coils?

could you coul the copper in slinky fashion - but keep the coils tight torture

run a few of lengths of copper/aluminium/stainless/whatever bar the full length and cable tie the coils to it

Thats exactly what I had intended to do but when I went down the hole with a single 10m coil to see how it would fit as a slinky I realised its not really possible. Its difficult to explain, but the access, the shape, the depth, the claustrophobic space, working in cold water, all conspire against you.

The compromise was to use 25m coils all individually dropped in and tied together, at different depths and offset against each other so the water flow runs in and around each coil. After 4 of them, I found I had enough space for a 5th so I added another. The water is now packed with copper and I could not add any more. They are all paralleled up to give a similar cross sectional area to the main 22mm flow and return pipes.

If the ground loop temp keeps dropping and evidently its not working then I will have to re-visit the design and try to loosen out the coils, but I am hoping its just over engineered enough not to have to.

[Nickel2]

Gravy-mine springs to mind... 

[offthegridandy]

I've sussed it.
It's a ruddy great still he's making.

Send us a jar nowty!!