Ralphs Water Wheel 2011

[mitcamp]

The Water Wheel is 10 feet in diameter, 6 feet wide and has 16 paddles.  I the spring of 2008 it was a 12 volt system with 16 mags  2 x 1 x .05 on each disc , 12 coils, 27 turns 2 in hand 14 ga.  In 2009 we changed the stator to 24 volt system and  in 2010 we changed it to a 48 volt system. All of the above was used with a 50-1 transmission.  We used 3 new gear boxes in 2008  and we used a different gear box supplier in 2009  Same problem, when the high water came, gear box broke up. The water wheel turned 4 R.P.M in low water and 8 R.P.M in fast water

In 2010 a local machine shop made a transmission with large sprockets and chain, same 50-1 gearbox ratio.  When the high water came, it twisted the 2 5/16 main shaft and ruined the large pillow block bearings.

As of this date the water wheel is in the shop and it will come out with 12 paddles, instead of the 16 paddles. The machine shop want to change the ratio to 25-1 and make it more effecient. They will instal a brake and will use gears only ,no chains.

I have the generator in my shop to be painted and build a new stator for the 25-1 gear change.  I used hvirtane/Hannas coil calulations and wound a coil with 150 turns of 1 in hand16 ga. Using the old magnet  hubs with the same rectangle magnets 1 x 2 x 0.5  I got 5 volts @ 108 R.P.M     Problem is the coil is now 3/4 inch in thickness in order to get all the wire in.

2 questions    1  Shoud I go with the larger 1 x 2 x 3/4  rectangles magnets to accomadate the 3/4 inch thick coil.
                    2   Would the larger magnets allow me to wind with less turns.

Comments are welcome  Mitcamp

[mitcamp]

This  is the  50-1 transmission constructed in 2010 with gears and chain.  It also had a brake to slow it down. This transmission was on the wheel when the 2 5/16 shaft twisted.  The damage was done in the middle of the night.

[electrondady1]

i hope you can sort it all out soon.
it sounds like there is alot of power to be harvested  with all that breakage
i think your stator support brackets are very long.
good luck

[taylorp035]

Wow - twisting 2.3125" shaft is impressive and scary.  Did you do any calculations to fix this problem?  Else maybe a large diameter + multi piece sprocket for a chain should be built, kinda the 84 tooth sprocket on my old supermileage car.  This way you can use a bigger leaver arm.  Maybe a 4ft diameter sprocket to a 2" sprocket using heavy chain (#50 or 60???).  You could build the sprocket in 4 or even 8 pieces.  Or a carbon fiber timing belt(just as strong a a chain for their size) --> Gates Poly Chain GT2 and the small sprocket to go along with it.  They don't stretch, so tensioning shouldn't be an issue.  The big pulley would need to be made from nylon (my car team has done this before).

Option 2 says you buy a high strength steel axle, or at least the piece between the paddle wheel connection and the transmission.  If you just broke it with a low quality steel, then you could find something with 2x it's strength.

Option 3 says to make the axle larger.

I hope you find a solution.  All of this breaking stuff sounds expensive.

[ghurd]

Nice to get an update!
This is one of the projects I think about regularly.
G-

[Madscientist267]

This is one of the projects I think about regularly.

First I've seen it (somehow)... but I have this feeling it's not going to leave me any time soon...

That's a pretty impressive piece of hardware there, dude. What kind of power output are we talking about here?

Steve

[ghurd]

[First I've seen it (somehow)... but I have this feeling it's not going to leave me any time soon...

That's a pretty impressive piece of hardware there, dude. What kind of power output are we talking about here?

The original post is here,
http://fieldlines.com/board/index.php/topic,130210.0.html


Mitcamp does not post often, but his posts are certainly worth looking at!
Might want look in his profile for 'latest posts'. Some impressive stuff in there.
Mitcamp should post more regularly!
G-

[mitcamp]

electrondady 1
Some one previously, maybe it was you, a year or so mentioned that the stator bolts look to be too long. So I will add some bracing. Thanks for reminding me.

Taylorp035
I can suggest improvements to the engineer, but he has the last say. Thanks for your suggestion.

Madscientist
300 - 400 watts,  we  all want just a little more power.

Ghurd
Thanks for the kind word's.  I,m not to sharp on computer stuff.

Update,  I wound another coil, this time I was careful with the wraps and found I can easily get 160 turns of 1 in hand 16 ga. with a thickness of 5/8 inch.  With the new transmission ratio of 25 - 1  We want it to charge the batterys at 100 R.P.M

Mitcamp

[12AX7]

Hello!   

I'm happy to see an update on this project!

I'm not sure how to phrase this, so here goes anyways..     

You mentioned that the main shaft 2 5/16 was "twisted" and also mentioned 300-400 watts.

This reminds me of a chat that I had with Zubbly a number of years ago, subject being,  power/watts/horsepower.
If the waterwheel/stream has enough torque/power to twist a shaft of that size, wouldn't there be much more than 300/400 watts worth of energy there to be captured?

ax7
Mark

[Madscientist267]

That's what I'm sayin!

400W is what, a little over 1/2HP? Yet tons of torque; shaft twisting of that nature doesn't just happen "on its own" hahaha

I'm no physics major either, but surely there's more than 400 watts to be had at that shaft.

Also mentioned (in the original link) is the snapping of a tree branch of the 3" persuasion. What are we missing?

Steve

[mitcamp]

12AX7

The machine shop believes by using a lower ratio gear box it will be more efficient.  The wheel in it,s former state had tremendous power. I was on site when the boys had to stop the wheel. They cut a 3 inch green maple tree, 12 feet long and rammed it between the water wheel spokes and it sheared it off clean.   I stayed on the bank on this operation.   good idea, less swearing, less muck in your shorts. Last year they added a brake to stop the wheel.   I,ll keep you posted

Madscientist267

When the waterwheel had 16 paddles there was a lot of splashing, sounding like it was fighting itself, clunk clunk clunk.  When the first paddle was filling, the second paddle was full (straight down),  at the same time the third paddle was pushing water, (trying to get out of the water)  This is causing the wheel to slow.  I believe with 12 paddles, it is going to be much smoother.  We will see when we put it back to work in 2 to 3 weeks.               

[taylorp035]

Lets do the math here:

4 - 8 rpm  --> lets say 6 rpm
6 rpm = 0.6283 rad / s
output power is 400 watts.
if we are at 50% efficiency, that make 800w at the shaft

Power = Torque * rad/s = 800 = T * 0.6283
T = 1273 N *m 

Ok, from my experience, 1273 Nm is a lot of torque, some where around school bus territory or turbo diesel pickup trucks.  But I don't know what their diameters are.

At 2.3125" diameter, we get 0.0294 m radius --->  Shear Stress = T * radius / J,  where J = (Pi /4) * radius ^4 =

Shear Stress = 1273 * .0294 / ((3.14/4)*.0294^4) = 63781754 Pa = 63.7 MPa

hmm... according to my sheet of metal properties, the yield strength in shear for A-36 steel is 100 MPa.  If we knew what type of steel it was, then maybe we could get a better idea of what is happening.

Of course, the numbers could be higher due to more torque or lower rpm or different efficiency (I think my 50% maybe of been too high considering a 2 or 3 stage chain gear box?).



I hope this helps.

[REdiculous]

The machine shop believes by using a lower ratio gear box it will be more efficient.

It may be slightly more efficient mechanically but 100-200rpm is a very narrow range and is very slow for an alt. I get the feeling that it'll be less efficient overall.

Even the original 50:1 is pretty narrow and slow. I think 75:1 (300-600rpm) would be better. Cool project anyway.

[DanG]

Twisting the shaft - I can see that, gross flood river flow hydraulic turbulence-eddy trapping the wheel in pulses, snatching the leverage arm from low/no or reversing trend input to... whatever the iron needed to relax trying to spool up the gears and alternator.  Maybe a judicious shear pin would be easier replacing than reworking gears?

[Madscientist267]

Power = Torque * rad/s = 800 = T * 0.6283
T = 1273 N *m

Ok, from my experience, 1273 Nm is a lot of torque, some where around school bus territory or turbo diesel pickup trucks.  But I don't know what their diameters are.

At 2.3125" diameter, we get 0.0294 m radius --->  Shear Stress = T * radius / J,  where J = (Pi /4) * radius ^4 =

Shear Stress = 1273 * .0294 / ((3.14/4)*.0294^4) = 63781754 Pa = 63.7 MPa

Oh SUUUURE, bring out the math! 

Got to thinking about this, and thought about it in terms of shearing the maple log.

If you think about what a log splitter does, then it almost makes sense.

That ram moves at a very slow rate, powered by a (typically?) 5HP engine. The engine's shaft is spinning MUCH faster than this water wheel, yet has much less torque. All the while, the same work gets done.

Guess what I'm trying to say is if you only had a 1/2 horse engine, and were willing to wait long enough, that engine could shear the 3" maple log as well, no problem.

Thinking about it in terms of say a buck converter (analogies in parenthesis) ; Higher voltage (speed) at lower current (torque) = Lower voltage (speed) at higher current (torque).

And you had to use math... Silly uber-efficient car maker... 

J/K - LMAO

Steve

[taylorp035]

Well, I was doing my Strength of Material homework anyways, so I might as pull out the equation.  If it were up to me, everything would be made of carbon fiber, including the water wheel and the shaft.

[RP]

On bending the shaft:  Consider the momentum of the wheel at speed if the thing was stopped abruptly.  That'd be considerable more torque than a few hundred watts would provide.

[hydrosun]

I don't know if it is too late but if you are reducing the number of paddles it seems you could also reduce the diameter to increase the rpm? Or are you reducing the number of paddles to increase the distance between them to reduce the interference.   With an overshot wheel the diameter is determined by the height of the water drop. With an undershot it would be the size where one  paddle doesn't interfere with the next. How did you determine the size the wheel needed to be?  The critical dimensions seem to be the width and depth of the paddles to intercept the most water. And close enough together to avoid a gap in capturing the energy of the stream. The diagrams of poincet undershot wheels I've seen had the paddles close together.  On pelton wheels there are three buckets always in contact with the water jet. One starting, one full on and one leaving the flow.   My area of experience is with higher head hydro so I'm just asking silly questions for these types of wheels. I looked at the original posting without seeing how the wheel was designed.
Chris

[taylorp035]

On bending the shaft:  Consider the momentum of the wheel at speed if the thing was stopped abruptly.  That'd be considerable more torque than a few hundred watts would provide.

I agree.  Obviously we need a decent safety factor, but that would be difficult to calculate without an radial deceleration rate.

T = I*alpha.  If T ~ 1300 N*m,  I need an approximation to the moment of inertia or roughly the mass of the wheel.
If we go with 1/2 m R^2 = I (roughly), then m ~ 500kg? and R = 1.524m -->  I = 580.644 m^4
Which means alpha could equal 1300 / 580.664 = 2.24 rad / s^2 for the same amount of torque as the wheel in operation.

Of course, this extra moment adds to the normal torque, so for a total of 2600 n*m or about 127 MPa.


In conclusion, there are definitely some serious consideration to be taken for the acceleration of the wheel.  2 rad/s^2 would mean stopping the wheel from 6 rpm to 0 rpm in 0.314 seconds....

[ghurd]

Lets do the math here:....

I hope this helps.

It helps aspirin company stock values, and LR44 calculator battery sales.
All it did for me was cause a killer headache and a trip to RatShak for 3 sets of LR44s!   

Pleases me to see some of the youth do not spend all their time exercising their Atari thumb.   
Atari?  Google it. 
G=

[taylorp035]

Yeah, I'm one of the few.  I spend 12 hours a day punching numbers into a calculator (too much hw), so it is a good thing that I like it.   Edit: Today I was designing a zip line that pumps water, so lots of number crunching.  I am by no means a math major, just an ME student who loves to be creative and solve problems with power (windmills, this lovely waterwheel, and my supermileage car).

If we up the ante, you could get the steel that they make socket heat cap screws out of, which is good for 1240 MPa.  That should solve your problem.  Or some titanium like I used in my helicopter combat robot, which was much stronger than an equal sized steel shaft.  But ~3" diameter titanium is about $1000 a foot...  I bet a well wrapped carbon fiber shaft would be better.

To solve the shearing issue with normal material, you could use my earlier idea of a larger diameter timing belt or chain using a 1 stage gear system for your ~50:1 ratio.  Obviously this would be a custom job for a cnc machine, but it would be a sure fix.  It can be done (and I have seen it done).  Definitely let the engineer redo his calculations on the design before anything happens.


Does anyone think that making an alternator good for 48v @ 4 rpm and 400 watts be possible?  It would need a lot of poles and many turns of wire.  I would almost put one on each side of the paddle wheel so you only have to hit 24v.  Either that, or put magnets all the way around the 10 ft diameter

[12AX7]

Here's a low speed alternator.

http://www.youtube.com/watch?v=InZ9XtiKpXo&feature=related

ax7
Mark

[taylorp035]

^ If you could get the wheel up to 8 - 12 rpm, then that motor would be perfect.  Plus, there would be no losses in the gearbox.  And at 13kw, you would never burn it up.

1.81 rpm / v  * 8 rpm = 14.54 volts ~ cut in on a 12v.  Unless you could wire it for an even higher voltage....  Or a 5:1 metal single stage gear box would be ideal.

[hydrosun]

What if instead of making the shaft stronger you make the wheel diameter smaller. That would decrease the torque on the shaft but have the same horsepower with a higher rpm.  The extended arms holding the buckets don't extract any power from the water. With curved buckets that start from a smaller radius and radiate out the wheel would have the same contact surface with the water. I went back and looked at a similar idea on this site in 2006. There is a link to a czech site that gives design numbers for ratio between different size wheels. mve.energetika.cz/vodnikolaDD/poncelet.htm
Chris

[REdiculous]

hydrosun, my dad was thinking along the same lines but I'm not sure it'd be worth it since he'd basically need to build a whole new machine. 

I'd be interested in knowing more about how the original gearboxes fell apart and how they were constructed. Stripped the teeth off the sprockets, broken chains or what? Got any pics? tia!

[mitcamp]

hydrosun

Yes we are reducing the number of paddles to increase the distance between them to reduce the interference.  The 3 rd paddle was making the noise, you could hear it clunking 50 feet away.  The paddles are not C shaped,  they are shaped like a 6 sided nut (for a bolt) cut exactly in half.

Taylorp035

I got the same killer headache that ghurd got.  Thats over my head.  You have a gift, use it.

REdiculous

The first 4 gearboxes were similar (see first picture)  They stripped the small pinion gear.  In the second picture is the gearbox that was fabricated in the shop with gears and heavy chain.


Mitcamp

[REdiculous]

I was hoping you had pics of the insides so we could see how they were built and how they failed.

[Madscientist267]

Either that, or put magnets all the way around the 10 ft diameter

I wondered about that too... you could place the 'alternator' at the top of the wheel to minimize contact with the water.

Thinking of it in terms of something that resembles a brake calliper. I can't see any benefit to having coils going around the entire circumference, and might even have detrimental effects.

You'd certainly get past the losses, shaft twisting, and moving part count by going this direction.

25:1 at quick glance looks very plausible (ratio of circumference of the magnet centers on the wheel to circumference of the existing alternator centers)... ?

Anybody tried it?

Steve

[REdiculous]

But if there's any ferrous bits in the water your mags could get fouled which could end up destroying the stator.

[Madscientist267]

Hmm, true.

Maybe a somewhat smaller diameter ring bolted to the wheel so that the mags are not in the water?

Say, 8ft? Still could probably hit the ratio needed...

And possibly a magnetic 'guard' at the bottom, down just below the rotor ring, that is simply a few magnets between two steel plates to catch anything that gets splashed up?

Food for thought.

Steve

[12AX7]

Not sure this will work, an old up load from long ago.

Just a indicator that real power can be harvested at low speeds.

http://www.otherpower.com/images/scimages/4822/Direct_drive_gener.pdf

Whats wrong with a few MEGA watts @ 20 RPM?

[Madscientist267]

I don't know about megawatts, but there's bound to be more than 400W available from a 10' undershot.

I'd be inclined to believe that, yes, some of the problem is that third paddle in the water, but the losses in the transmission can't be trivial either.

Speed isn't the only thing that increases in a 25 or 50:1 transmission. Bearing losses, gear friction, parasitic drag (eddy effects, etc) at the output shaft all get multiplied by that amount as the input shaft sees it.

It's the whole point behind doing direct drive for windmills... why would water be different? Granted the rotor turns slower, but this can be compensated for by making the rotor bigger, something that isn't (generally) practical on a windmill.

Sure, there are optimizations that could be made all over the place, but this one is a little closer to the 'more obvious' department. No?

Steve

[hydrosun]

My comments may not apply to this project because the turbine is already done.
 When I looked at the historical record for undershot wheels, it was a huge improvement when going to a curved blade instead of the flat paddles.The efficiency went from 30 to 70% efficiency. The overall design was still of a large diameter wheel like those used for overshot sites. So people had adapted a known design to utilize no head sites.  Since the wheels were used for grinding grain and other high torque slow speed uses there wasn't much need for smaller wheels. Then the author wondered why the use of these poncelet wheels went out of fashion. Thinking about it, I came to the conclusion that it was the switch to the smaller diameter cross flow wheels to create higher rpm to generate electricity that made the large wheel obsolete.  Even without the cross flow it seems that the smallest diameter wheel that can accomodate the volume of water would work the best for electricity generation.  In pelton turbines the diameter is kept the smallest possible that can handle the round jet of water. There are even designs using multiple peltons on one shaft to keep the diameter smaller with more water. So using a smaller diameter but wider undershot would do the same thing.

Is the 3rd paddle clunking because of the flat bottom of the bucket hitting the water thrown up by the bucket before it instead of cutting through. It may be why other designs had many close gently curving paddles instead of buckets. Smacking the water in the wrong direction would just slow the wheel down. Maybe just cut off the lowest part of the bucket and lower the entire shaft a few inches instead of taking out some of the buckets.
Chris
Chris