Hydro Power

I am giving some serious thought to hydro power.

Preliminary investigations suggest there must be a catch, 'cos the return can't be that good.

As a start point: 3kW turbine £3.5k A wild guess at £10k to install ~ £13.5k.

3kW, 24 hours/day, 365 days/year = 26280kWh, at 20p/kWh =£5256 per annum... paid for in less than three years.


What have I missed? Apart from the water abstraction license, and the cost of MCS certification.

Dunno - but I'll watch replies with interest as Hydro interests me greatly..

Where would you be instaling your turbine? A lot depends on water flow and height of the drop, would also be interested in the replies.

The power available in a stream of water is
P = H x Q x n x g

P is the electrical power output in watts (W)
H is the head in metres (m)
Q is the flow rate in litres per second (l/s)
n is the hydraulic efficiency of the turbine system (%)
g is the gravitational constant which is 9.8m/s/s

A low head is considered less than 10m. A micro-hydro system ranges from 60-85% effeciency.

A 1kW micro system with effeciency of 50% needs a head of 23m at a flow rate of 10 l/s.

The seasonal variation in flow also needs to be considered.

The Crabble Corn mill has two micro systems each of 1.2kW. It only has a head of 2.63m (replacing the original water wheel) but it has a flow rate of 0.43 m3/s...a lot of water. This system is proven to have 65% effeciency.

Cost effective systems are where the water is actually next to the building to which it is providing the electricity. The other big cost can be the civil enginnering works of weirs and pipe system (penstock).

Problems needed to be considered is storing water to the smooth seasonal flow. Good housing to ensure no water ingress or leaks into the electrical system. Very good design of the penstock to ensure no loses under pressure and no loss of head due to too small a pipe.

There is a british-hydro.org web site that may help make you mind up!

We would have installed a micro system if the house was below the lake but its 400m away! That why I've looked at the info...
I have thought of putting the freezer down there etc but it doesn't really make any sense at the moment.

Zoe wrote:We would have installed a micro system if the house was below the lake but its 400m away!

You just need a very long cable.

Zoe wrote:A 1kW micro system with effeciency of 50% needs a head of 23m at a flow rate of 10 l/s.

Or, presumably a 2.3m head and a 100 l/s flow rate or any other appropriate combination of numbers.
Which is all very well, but what does 10 l/s look like?
OK it is not hard to estimate the speed of a stream, but my I've not seen many regularly shaped streams, so the cross sectional area is very hard to measure.

Moving on to the next question. I have a river, and I have a generator.
The generator needs to spin at a given speed to produce mains electricity.
So how do you design your turbine to deliver that speed?

Very good design of the penstock to ensure no loses under pressure and no loss of head due to too small a pipe.

The energy is free, does it matter so much if you waste a bit of it?

Hairyloon wrote: What have I missed? Apart from the water abstraction license, and the cost of MCS certification.

You've missed lots. I'll be able to help with a lot of your questions at least to get you pointed in the right direction. But let's start with what puzzles you most, cost.

If that you have i an old mill site, the dam, the pond, the channel or pipe, etc. are all still in place and all you are talking about is the turbine and generator you are in the right ballpark as to cost. But if not, if you just have the stream and are otherwise starting from scratch, the cost of the turbine/generator will be only a rather small fraction of the total cost.

OK, from the beginning, you need flow and you need drop (head). We'll get to how to measure these to see if you have a potentially useful site but first some terminology. I will be using the term "turbine" for all rotating devices, not distinguishing "wheels". Thus an ordinary "overshot wheel" is "an impulse turbine where its size is not negligible compared to the total head" (an ordinary impusle turbine is tiny compared to the total drop so we ignore the change in gravitational potential while the water passes through the device -- with the "water wheel" we ignore the impulse part except for getting the proper speed of the wheel).

Second, we will be doing "micro-hydro" so some of the things in the books describing the usual megahydro aren't appropriate. Thus it isn't really right to say things like "above 10 m is high head and below low head" because that is relative to the flow. If that's 10 m head and a flow of 20 l/sce we will want a high head sort of device but if that 10 m head had a flow of 20 c/sec they'd be using a low head device. NOTE: we will not be discussing mini-hydro like that 20 cubec example I just gave). Given the relationship head X flow = power and that we will be talking about micro-hydro (at most a few kw) low head will be less than a couple meters, high will be anything over 5, and in between medium.

OK, head and flow. The potentially available drop as the stream flows across your land is an ordinary surveying problem. Measuring the flow is more likely to be the problem. Cross section and current is for when you are doing sizable rivers. For micro projects you will build a weir with a sharp edged* notch from which formulas based on measurments will tell you the flow (what are the dimensions of the notch and the depth of flow over it measured well upstream - before the surface of the water curves downward as it speeds up to pass through the notch). In other words, you will need to build a temporary dam forcing the water to pass through a rectangular (or triangular) notch and then you will know the flow.

If you have a high head application (lower flow) instead of a notch may well use a (sharp edged) circular hole instead of a notch**. I can give you s simple experiment so you can see the "contraction" of which I speak. Take a large tin can or a bucket that is cracking and so you can spare. Drill a moderate size hole through it (say 1 cm) and sharpen the edge of that hole (from the outside use a flared reamer so the hole is larger on the inside of the bucket -- then the water coming out won't touch the surface of the hole. Now fill the bucket and look closely at the jet of water coming out. See it curve in so the jet diameter is smaller than the diameter of the hole. That's the "contraction of flow".

* The edges should be sharp as this gives a known "contraction of flow"

** We'd have to look up notch tables but could directly calculate flow though a sharp edged circular hole in a wall.

Mike wrote:If that you have i an old mill site, the dam, the pond, the channel or pipe, etc. are all still in place...

Firstly, there are quite a lot of such places, and secondly, why can you not simply get a big roll of fat PVC pipe; stick one end in the river upstream and run the other end down to your turbine?
Obviously not quite that simple, but that is the basic deal. If we plan to only use a fraction of the river, then it shouldn't need much of a dam/pond.

Hairyloon wrote:

Mike wrote:If that you have i an old mill site, the dam, the pond, the channel or pipe, etc. are all still in place...

Firstly, there are quite a lot of such places, and secondly, why can you not simply get a big roll of fat PVC pipe; stick one end in the river upstream and run the other end down to your turbine?
Obviously not quite that simple, but that is the basic deal. If we plan to only use a fraction of the river, then it shouldn't need much of a dam/pond.

I'll separate these two.

The pipe: Let's assume that the stream gradient is 3% (that's steep) and you have available a drop of about 10 meters. That's a pipe about 300 meters long. What do you think it would cost for say a 1/4 to 1/2 meter tight pipe that length? Include the grading, possible support structures over dips, etc. The sort of diameter piping we are familiar with in househld plumbing are delivering amounts measurable in a few liters per minute at signifcant pressure drop. Takes a much larger diameter pipe to deliver tens of liters per second at less loss of head.

BTW -- now if the available pipe size is a given, it sticks in my head that the optimum power at the turbine is when 1/3 of the head is used to drive the water through the pipe. I lost the book where this was demonstrated and can't remember the proof.

The dam: No, this depends mainly on the size of the river and not the amount of water to be extracted. You need to be able to raise the water level enough for the intake structure. In a very steeply dropping stream with enough depth and an excavatable bottom you might be able to use a buried (rock and gravel covered) intake and no dam but then the pipe would be in the stream channel for some distance and that's likely practical only is the stream isn't flashy*. Perhaps you might be able to excavate leading the pipe off at an angle to the bank (getting it out of the stream channel fast) but a major excavaton project like that won't be cheap, especially when you consider that the inital sections below water level.

PS -- on the topic of pond storage. This is likely to be cost effective only if already in place. For example, assuming a 5m drop an acre.foot of water isn't much over 10 kwh storage. Compare the cost of a dam, the land taken up by the pond, etc. with the cost of a batteries capable of providing that much storage and the cost implications for the turbine which now must be able to handle highly varying flows at good efficiency**.


* Or you lose that section of pipe every severe spate. Thus on Clesson Brook I would have that situation where the stream now flowing say 20-30 l/sec might go to several cubecs when in spate. A very significant project to anchor the initial sections of pipe until above flood flow. Could be done but talking about perhaps 50m protected/anchored by large (several ton) rocks -- in spate the stream often moves rocks under a ton.

** For the impulse turbines this is mainly just a matter of size, so while more costly, only somewhat so. They can have a reasonably flat efficiency curve over variance of flow 10:1. But if a reaction turbine is needed:
a) Fixed blade axial flow types have very severe drop in efficiency even slightly off the design point.
b) Fixed blade radial flow types can provide a range of perhaps 2:1 at good efficiency but range is best for the hig head sort (a low specific speed type Francis with a high ration of in diameter to out diameter).
c) The reaction types capable of efficient operation over a wide range of flow need baldes that adjust in addition to the guide vanes adjusting. The Kaplan and its kin. These are expensive!

Something like just the larger diamter PVC pipe might be usable if what you had was a comparative trickle (a few liters per second) but flowing down a steep mountainside so you had a drop of 50 meters. If even more head than that the lower sections would have to be stronger pipe (well the black plastic PVC pipe sold here is rated* for 90 psi). But if you had that 50 meter drop, using 10-15 meters of it might drive enough water through 2" PVC leaving 35-40 meters of head for the jet (at this head/flow ratio you would definitely be using an impulse turbine, either a Pelton or a Turgo)

* I don't know what the "safety factor" of that rating is or how much more a buried pipe might be good for. If you had a shut off near the top probably could take some chances as the flow isn't so great that a pipe burst would be a dangerous disaster. My ground here is very rocky, would not erode quickly from a burst pipe, especially if flow limited to some max at the intake.

Mike wrote:The pipe: Let's assume that the stream gradient is 3% (that's steep) and you have available a drop of about 10 meters. That's a pipe about 300 meters long. What do you think it would cost for say a 1/4 to 1/2 meter tight pipe that length?

I'm still looking for a supplier, but I frequently see great rolls of the stuff at the side of roadworks... I'll stop and ask next time I see one convenient.
The rig I was looking at asks for delivery through 150mm diameter pipe; I don't know if you want to feed down through something bigger?

Look --- this is likely to be more useful if you start by describing your potential site.

Estimate the flow. Are you talking about a few liters per second, a few tens of liters per second, or a few hundred liters per sec? (any more and even at low head you are "mini", not "micro", at least in the size of turbine you would need if not in power produced).

Estimate the head. And over what horizontal distance is this head obtained.

Describe any pre-existing hydro structures such as dams, ponds, leats, etc. (if any).

Describe how "flashy" is the stream, how high does the water rise when it's in spate. Describe your climate if it freezes hard in winter (do you need to worry about ice).

There are a lot of other site details but those can wait as some of these relevant of not depending on what sort of turbine.

Mike wrote:Look --- this is likely to be more useful if you start by describing your potential site.

I have at least three sites in mind. Probably more if I can get the hang of it.

Estimate the flow. Are you talking about a few liters per second, a few tens of liters per second, or a few hundred liters per sec?

Yes, yes, and yes. :?

Estimate the head. And over what horizontal distance is this head obtained...

I'll have to get back to you on that.
If it helps. as a starting point, I was looking at this turbine, or one of its family.

Hairyloon wrote:

Mike wrote:Look --- this is likely to be more useful if you start by describing your potential site.

I have at least three sites in mind. Probably more if I can get the hang of it.

Estimate the flow. Are you talking about a few liters per second, a few tens of liters per second, or a few hundred liters per sec?

Yes, yes, and yes. :?

Estimate the head. And over what horizontal distance is this head obtained...

I'll have to get back to you on that.
If it helps. as a starting point, I was looking at this turbine, or one of its family.

Look -- I was working on a long answer and lost the connection and I'm not going to start over.

Describe these three sites. I could then tell you for which of these this turbine would be suited and for which it would not. There is some flexibility but that might mean changing the generator and the nozzles.

BTW -- for "flow" I meant "at low water", what you could count on almost always having available, not the maximum flow when the stream was in spate. Seriously, I cannot imagine you could have available (that you would be allowed to use) a site flowing at hundreds of liters per second when at "low water" that dropped 20 meters in a relatively short distance. Would be a very scenic waterfall or a rapids that at higher water the "white water" folks would be using!

Mike wrote:Look -- I was working on a long answer and lost the connection and I'm not going to start over.

It is a right PITA when that happens.

Describe these three sites. I could then tell you for which of these this turbine would be suited and for which it would not.

As I said, I'll have to get back to you, I'm in the wrong area. But they do a whole range of turbines, so I'm sure there is one to suit each place.

Coming back to pipes, would these be suitable?
Roughly £5/m for 150mm diameter, so only £1500 for the 300m we talked about earlier.

Seriously, I cannot imagine you could have available (that you would be allowed to use) a site flowing at hundreds of liters per second when at "low water" that dropped 20 meters in a relatively short distance.

I didn't say it had many hundreds of litres, just a few.

I'll retry the longer comment

Lets start with this turbine. It's a Turgo (axial flow impulse turbine) and the type can be used with up to 8 (easier a max of 6) nozzles each in size up to 1/8 the pitch diameter. In other words, as impulse turbines go, potentially fairly high specific speed*. A very flexible type as the flow can be anything less than the maximum just by suing fewer and/or smaller diametrer nozzles.

Except when buying a particular model as is shouldn't think of it as "fixed" in capacity. It's turbine and generator together and with a Turgo, how many and what size nozzles. If piped for the maximum of 6-8 and the nozzle tips can be screwed off and replaced with ones of different size a different flow/head capacity and the power produced is flow*head so might need a different generator attached to it.

In other words this same turbine (with the number/size of nozzles it has now) isn't limited to 20 m head and a flow of 20 liters/sec. If the head were 80m the flow would be 40l/sec and the power ~12 kw (the 3 kw generator couldn't handle that, would need a bigger one). Or at 10 m head the flow would be around 14l/sec but you'd need to replace the 3kw geenrator with a 1 kw.

OK, now the pipe. I will repeat, just because at the inlet it's 150mm does not mean that a unit like this should be used with a pipe that small its entire length. In long pipes the frictional losses beciome excessive and in long pipes the "water hammer" severe when the flow inot the turbine is altered (with a multi nozzle Turgo that's usually by shutting off nozzle by nozzle --- nozzles with an adjustable "needle valev" are expensive).

Consider if the pipe length were 3000m instead of 300 (and no, that's not an unrealistic length for a 20 m drop if you are talking about a river). What is common in a long pipe installation is that you have most of the horizontal distance graded to just below the hydraulic gradient. To this point the pipe can be low pressure stuff and if "vented" at the lower end (vertical branch up just over the inlet height) and of large enough diameter losses in this section will be lower and water hammer not severe (if suddenly shut off to below there isn't a large mass of water in the vertical vent pipe to be accelerated so the moving water can escape). This upper vent is placed where the distance down to the trubine is relatively short and steep. That part of it must be very strong pipe and probably you have a second pipe leading back up to a vent so the force needed to accelerate the water in that pipe is the limit of the overpressure when flow to the turbine is cut.

The "big boys" don't shut down flow to impulse turbines quickly! Theirs are usually equiped with adjustable deflectors that can move quickly so the water jet(s) can be made to miss the turbine as quickly as needed and then the valve(s) can be closed slowly enough that the "water hammer" presure won't burst the pipe. Remember, the rapid shutting of a valve in a long horizontal flow is how the "hyrdaulic ram" can force water up 10-20 times the initial drop.



* The cross flow turbine is capable of even higher specific speed but that's because it has no defined axial length. If the axial length of a cross flow turbine isn't much longer then its diameter then it's specific speed won't be higher than a full flow (8 nozzle) Turgo. Again, specific speed is best thought of as a measure of how large a flow a given diameter trubine can handle. So yes, the cross flow is described as very high specific speed (for an impulse turbine) but that's with axial lengths 4-8 times the diameter.