University Waterwheel Help

[Dannbodge]

Hi Dan, just wanted you to know that the Dannbodge comment was meant as a compliment as you clearly do not bodge things.

Out of interest what are you going to use as a load when you are testing, you mention that you think you will be able to get 24 volts at 2000 rpm from memory but the load is the critical thing. Can I suggest that you use something resistive such as a small light or a group of bulbs on switches (wired in parralel ) as you will find that it will give you an easy way to measure your power at a certain voltage ( Ohms law etc ) and as the load will increase with revs the wheel should start easily as the load is very low untill some revs are produced. I made up a board with 20 switched light fittings with 100 watt bulbs , had a voltmeter hooked up and found out what she could do, I think the most we got was about 1.8 kw so thats how we gauged the size of the heater elements at 2kw. Before we had a PMG we tried another generator that was high speed ( bad ) and also voltage regulated, this meant that the wheel and the load had to be perfectly balanced or else the wheel would speed up and up or grind to a halt ( also very bad and scary) Another thing that is helpfull is that with a resistive load ( as long as you don't have a voltage regulator on the pmg) is that the wheel will balance itself against the load in relation to how much water or flow you have , take a volt reading and your load resistance and you can work from there to see if you can make it more effecient. Here is a handy calculator.

http://masterslic.tripod.com/olcalculator.html

I am sorry if I am teaching you to suck eggs, I am no uni student and had to learn all this stuff by having a go so bear with me if its already obvious to you.

Regards

Rob

I wasn't taking badly, just thought I would explain where it came from. I am grateful for the compliments.

I have a variety of bulbs ready to use, ranging from small E10 bulbs to 100w 24v ones. Of which we will be trying to load the wheel.
The most I think this will produce is around 2-300W if lucky, I have a voltmeter that will be used at the same time.
I have no worries about people saying things I might already know, after all you guys have more experience with this than I do and everything you say is read and taken into account.

[guydewdney]

when measuring power, you will find an analogue meter will help a lot, as it probably will cycle a bit - or fluctuate. digital meters are much more accurate on steady state measuring, but wobbly leccy needs a wobbly meter Both ammeter and volt meter if I were you, and make a 'best guess' from the apparent mean of the needle position.

[Dannbodge]

So I am back after my weekend of testing in Okehampton and.......

It didn't work
I am gutted to say the least.

The torque needed to turn the wheel when connected to the pmg was too much, and the river didn't have the power.
However with the pmg disconnected and just using the wheel on its own, it worked beautifully. Giving a speed of 60rpm.

The issue we have found was the wheel was just too small for the gearing and pmg I put onto it.

However, Project wise it has been a brilliant success and my tutor was extremely helpful and happy with what I have achieved.

[Lurk]

You got a result, perhaps not what you had hoped for but a result so the tests were a success. Will you change the gearing and re-test ?
It has been great to follow the thread - thanks for sharing - best of luck with your studies.
Lurk

[guydewdney]

what did you connect to the pmg? Lighten the load a bit maybe?

[Dannbodge]

You got a result, perhaps not what you had hoped for but a result so the tests were a success. Will you change the gearing and re-test ?
It has been great to follow the thread - thanks for sharing - best of luck with your studies.
Lurk

I did get a result, and can continue on my project with it. I feel I will do better with a failed result as I have lots of things to explore and show what I have learnt which will get me into the higher marks.
Going forward, I will be changing the gearing, trying a different pmg and potentially building a larger wheel. As we found that is the route of all the problems.

what did you connect to the pmg? Lighten the load a bit maybe?

The pmg had nothing connected.
With the pmg disconnected and the counter shaft connected the wheel still failed to turn. Once the counter shaft was disconnected, the wheel spun brilliantly.
The bearings I have used are fairly tight still and the shafts are probably over engineered for what it needed too.

I collected some more data from the river to allow me to calculate the rough amount of torque it gave to the wheel, so using my figures I can measure the torque needed to spin the wheel with pmg connected and keep increasing wheel size until I get one that will spin it. (The project markers love a bit of Maths and theory work)

[guydewdney]

somethings badly wrong if the bearings in those pillow blocks are 'tight' - they are self aligning, so there shouldnt be any tightness at all. Try loosening the chains off - they will withstand being very loose indeed for a while. check alignment of the two shafts - check parallel-ness?

[Dannbodge]

Quick question for anyone.

I have a measurement of for the force that the river exerts on the wheel.
How do I work out how much torque the wheel will produce?

[guydewdney]

torque is measureed in Newton . Metres

newtons are a force.

Metres are a distance.

You have the force which acts on the ?centre of area? of the paddle. You have the distance of that from the axle.

example
force 20N
Centre from axle 0.3M
torque = 20 x 0.3 = 6Nm

Then for the gearing, its a simple ratio - so you have a 10:1 speed up, you have a 1:10 torque down ratio - so in this example you would have 0.6Nm


Power of course is torque x speed, but you know that, don't you, as you are doing a degree in engineering....

[Dannbodge]

torque is measureed in Newton . Metres

newtons are a force.

Metres are a distance.

You have the force which acts on the ?centre of area? of the paddle. You have the distance of that from the axle.

example
force 20N
Centre from axle 0.3M
torque = 20 x 0.3 = 6Nm

Then for the gearing, its a simple ratio - so you have a 10:1 speed up, you have a 1:10 torque down ratio - so in this example you would have 0.6Nm


Power of course is torque x speed, but you know that, don't you, as you are doing a degree in engineering....

Nope Degree in Design. Touched a bit of physics in my 1st year and at A level but forgotten 90% of that now.
I measured the river's force at 49N.
Now If the wheel was 100% efficient I should get 14.7NM of torque.
I know that I had no where near that as It couldn't spin the counter shaft which requires ~ 1NM.

Does the curved blades affect the amount of force converted into torque?

I was wondering if there was a way of actually calculating the torque the wheel produced.

[guydewdney]

ah - a "do you want fries with that" degreee   

A crude method would be to attach a spring balance to the wheel, and plonk it in, but thats not strictly right, as its a dynamic thing, not a static thing. Ideally, you want a dynometer - which is essentially what you have built - replace the pmg with a simple DC electric motor, brushed. Buy a  cheap rev counter (optical) on fleabay (unless you can blag one off the uni). Spin up the wheel, as the dc motor shouldnt have any drag worth talking about with no voltage on it. Then apply a variable voltage  and known current to slow the wheel, and measure Volts, amps and rpm. Volts x amps = power. Plot graph of power against rpm. note best speed/power point - use that SPEED OF TIP OF WHEEL NOT RPM to base calcs for larger version.

[Dannbodge]

ah - a "do you want fries with that" degreee   

A crude method would be to attach a spring balance to the wheel, and plonk it in, but thats not strictly right, as its a dynamic thing, not a static thing. Ideally, you want a dynometer - which is essentially what you have built - replace the pmg with a simple DC electric motor, brushed. Buy a  cheap rev counter (optical) on fleabay (unless you can blag one off the uni). Spin up the wheel, as the dc motor shouldnt have any drag worth talking about with no voltage on it. Then apply a variable voltage  and known current to slow the wheel, and measure Volts, amps and rpm. Volts x amps = power. Plot graph of power against rpm. note best speed/power point - use that SPEED OF TIP OF WHEEL NOT RPM to base calcs for larger version.

Not at all, it's just not as scientific as an engineering one. I "learn" more about the actual product side rather then how it works. So through the conception of a product through the manufacture and testing etc.

I might pop into the physics department and see if they have any ideas about it/ get some physics books about it too.

[petertc]

Dan,

Just dug out an old book called energy primer. isbn 0-914774-00-x printed 1974

got a few figures in it.

NHP Net horse power.
Q cubic feet per second
H actual head in Ft
E efficiency of the device
correction factor 8.8

NHP = Q x H x E
                 8.8


Poncelet wheel ( which i think yours is)
Low speed High torque (not suitable for power generation so the book says)
Minimum dia 14 ft (up to a maximum of 4 x head ).
head range 3 to 10 ft
optimum rpm = 42.1 x  square root of H ( head in ft) / Diameter if wheel in FT
60 to 80% efficient
 Not sure if this helps  or not.


You should be able to work out the NHP on the flow that is flowing into the wheel. if you know your speed or maybe use the optimum rpm figure from that you should be able to work out the torque from the wheel. there are lots of online calculators for giving you torque figures from a speed and power.

[RobNute]

Don't give up mate, as I said in my first post ,

1. They are tempermental
2. They are torquey
3. The gearing is the key to success, keep it simple and as slow as possible ( the final generator speed, that is).

and also addictive.

Here is something that I came across that may help you, its all a mystery to me but you may be able to make sense of the maths.

http://sleekfreak.ath.cx:81/3wdev/VITAHTML/SUBLEV/EN1/WTRWHEEL.HTM

One thing that I think we all perhaps have been guilty of is expecting more power than is actually achievable, you are now in the club, welcome!

You may not need to replace the pmg, it should work just as well slower but with less volts, as you are not running it long term it should not be a problem , its not as if you are going to kill it with that amount of current. Gear it down, loosen the chains as Guy said, check the bearings ( run it for an hour or two with an electric drill to loosen them up if neccesary) and have another go. It would be great after all that effort to be able to measure some actual production however small, could you change the bulb for a resistor and measure the current that way ?  Well good luck anyway, I guess the first goal is to get the pmg spinning with no load and work from there.

One thing that I think I would try would be to put a lining around the inside of the wheel that would change the blades into buckets like an overshot wheel, I know that is a different design but I don't think that yours works quite like a poncelot is intended , here is a quote from this page which is a very good explanation of the poncelot design.

http://www.waterwheelfactory.com/poncelet.htm

"If one has a hydro site with as little as 2 feet of head and with a sufficient flow of water, a Poncelet designed for this combination may be the most practical wheel available."

It would seem that at least some head is still required, and that the buckets should be able to dump the spent water into the tailrace free of turbulance. Could you make up a short floating leat positioned where you have the most drop and get yourself enough head, you would only need 100mm or so to get you clear. Time for some bodging I feel!

This also worries me

"The idea of my design is that the front "scoop" lifts the water slightly so that the exit is level with the water."

Surely that will negate the effect of the poncelot design as you are reducing the velocity of the water before it gets to the blades which is supposed to their job? I like the idea but it seems that maybe the two concepts oppose each other.

Anyway, thats enough thinking about waterwheels, I will get into trouble if I don't do something usefull now!

Best wishes

Rob
 

[Dannbodge]

Dan,

Just dug out an old book called energy primer. isbn 0-914774-00-x printed 1974

got a few figures in it.

NHP Net horse power.
Q cubic feet per second
H actual head in Ft
E efficiency of the device
correction factor 8.8

NHP = Q x H x E
                 8.8


Poncelet wheel ( which i think yours is)
Low speed High torque (not suitable for power generation so the book says)
Minimum dia 14 ft (up to a maximum of 4 x head ).
head range 3 to 10 ft
optimum rpm = 42.1 x  square root of H ( head in ft) / Diameter if wheel in FT
60 to 80% efficient
 Not sure if this helps  or not.


You should be able to work out the NHP on the flow that is flowing into the wheel. if you know your speed or maybe use the optimum rpm figure from that you should be able to work out the torque from the wheel. there are lots of online calculators for giving you torque figures from a speed and power.

Potentially that is lots of help. Thanks

Don't give up mate, as I said in my first post ,

1. They are tempermental
2. They are torquey
3. The gearing is the key to success, keep it simple and as slow as possible ( the final generator speed, that is).

and also addictive.

Here is something that I came across that may help you, its all a mystery to me but you may be able to make sense of the maths.

http://sleekfreak.ath.cx:81/3wdev/VITAHTML/SUBLEV/EN1/WTRWHEEL.HTM

One thing that I think we all perhaps have been guilty of is expecting more power than is actually achievable, you are now in the club, welcome!

You may not need to replace the pmg, it should work just as well slower but with less volts, as you are not running it long term it should not be a problem , its not as if you are going to kill it with that amount of current. Gear it down, loosen the chains as Guy said, check the bearings ( run it for an hour or two with an electric drill to loosen them up if neccesary) and have another go. It would be great after all that effort to be able to measure some actual production however small, could you change the bulb for a resistor and measure the current that way ?  Well good luck anyway, I guess the first goal is to get the pmg spinning with no load and work from there.

One thing that I think I would try would be to put a lining around the inside of the wheel that would change the blades into buckets like an overshot wheel, I know that is a different design but I don't think that yours works quite like a poncelot is intended , here is a quote from this page which is a very good explanation of the poncelot design.

http://www.waterwheelfactory.com/poncelet.htm

"If one has a hydro site with as little as 2 feet of head and with a sufficient flow of water, a Poncelet designed for this combination may be the most practical wheel available."

It would seem that at least some head is still required, and that the buckets should be able to dump the spent water into the tailrace free of turbulance. Could you make up a short floating leat positioned where you have the most drop and get yourself enough head, you would only need 100mm or so to get you clear. Time for some bodging I feel!

This also worries me

"The idea of my design is that the front "scoop" lifts the water slightly so that the exit is level with the water."

Surely that will negate the effect of the poncelot design as you are reducing the velocity of the water before it gets to the blades which is supposed to their job? I like the idea but it seems that maybe the two concepts oppose each other.

Anyway, thats enough thinking about waterwheels, I will get into trouble if I don't do something usefull now!

Best wishes

Rob
 

After reading that link, I think there is some re-designing to do on the actual wheel.
I could move the wheel onto our neighbours stretch of river and use just before some rapids as a leat.

The idea of the scoop was to gain some head on the wheel, so that the back wouldn't get choked by the tail water. But after having tested it, the river is moving fast enough that the water is clear of the back anyway, another reason was so that a bit more of the flow is directed into the wheel.

For my uni project, I am all but done. Apart from changing some sizes and sprockets to theoretically make it work.

As a family project, it is going to continue. Firstly by changing gearing, the blades then making a bigger wheel.


By the way.
A massive thanks to everyone that comments on this thread. I take on board and read every comment that is posted.
I plan to mention you in my acknowledgements and thanks page for my project.