Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Governing of torque or rpm

  1. Mar 3, 2014 #1
    Hi all, new to the forum and hoping I might get some ideas for a project I am working on, not an Engineer or very well technically educated, so please go easy on me :smile:.

    What I wish to do is to build a waterwheel which will supply electrical power to my home. I have been doing some research in some old books on waterwheels, since it is an old technology not used much any more, being replaced by water turbines for the most part. Obviously, the waterwheel is going to turn at different speeds, depending on the flow of the river which varies by season and such. But electric generator heads don't like variable RPM's to them as far as I know, although someone who builds waterwheels answered a few questions and one of them was that once connected to load, the generator head would 'govern' the wheel to a certain degree.

    I was thinking of using an undershot waterwheel directly in the stream, which is why you would get variation in the turn speed of the waterwheel. In the old books, they often talked about using a 'sluice' I would supposed you call it, and they had ways to raise and lower the sluice gate to admit more or less water into the sluice - thereby regulating the amount of water getting to the wheel and adjusting it's RPM's. But often these were adjusted by hand, you had to go to the waterwheel and adjust the gate manually, although I think there were some 'automated' attachments you could use for raising and lowering the sluice gate 'automatically' to adjust flow to the wheel.

    I was just thinking that with all our new technology there should be some way to automate this process, and I would like to do so, so that I don't have to go to the wheel all the time and make adjustments.

    I don't know the wisest course of action to follow when it comes to governing the RPM's either. Build something onto the wheel that would act like a centrifugal clutch or something of that nature, and regulate it's turning there? Or use something at the other end to govern the turning of the pulleys?

    I have kind of been wracking my brain here, and I have thought of a few possible solutions, but I don't know how well each one would work. Also keep in mind, I don't know the river or the site on the river I will be using, so I don't know if the governor will have to work in ranges of 20 to 50 HP or 150 to 300 HP, and I don't know what the torque levels would be either.

    I have thought about getting a transmission of some sort from a vehicle, say an automatic or CVT transmission of some type - a manual transmission wouldn't work because I am wanting this to be 'self adjusting' and unattended in nature, or with the minimal attention required. But I don't know if a vehicle transmission would govern the RPM's or not. And if so, how would it do so, or how could it be arranged to do so?

    As I said earlier, I also thought about a centrifugal clutch arrangement of some sort. But what would be the best way to do so? And if it were engaged for a period of days I can see the possibility of it burning out, say when the river or creek was in flood stage.

    And, with my minimal engineering and mechanical background, I am convinced the possibility I am not finding or thinking of a good solution of some other type might be quite high.

    Any help you could give on this problem would be very much appreciated. Thanks, Steve:smile:
  2. jcsd
  3. Mar 3, 2014 #2


    User Avatar
    Gold Member

    You have a classic situation where engineering is essential, because there is no cookie cutter solution available.
    That said, you can narrow down your specs a little.
    Supplying a home modestly takes about 5000 watts, which translates to about 7 horsepower. That will comfortably run your lights, entertainment and small fridge, no big A/C or well pumps. Please do note that a horsepower is a pretty big unit, 33,000 ft pounds/minute. Water weights about 64 pounds/ cubic foot, so equal to lifting 500 cubic feet of water up one foot in one minute. Your wheel will extract only a fraction of the energy of the falling water, so you need to think in terms of several thousand cubic feet of water going through your sluice every minute in your setup. Can your stream supply that reliably and are you able to make the installation required to use it? Also, a 5-10hp generator is not cheap or easily transported. The normal designs use a small internal combustion engine as a driver, so they are driven at around 2000rpm, way faster that your water wheel. Adapting that to your setup will require gears, not all that cheap either as you are looking for about a 100:1 speed reduction, even assuming your wheel whizzes at 20 rpm.
    So there are some design issues to be ironed out before you can worry about variable speeds or stream flows.
    Good luck and please keep us posted.
  4. Mar 3, 2014 #3
    additional information

    The 'gearing' and other things can be taken care of by using a technique I came across on another website, it was cleverly done by the gentleman who built it. He basically used chain and sprockets on the side of it to turn the waterwheel itself into one big sprocket. So, the gearing and such isn't the problem I am facing. I will of course be building my own model to test out his technique, but he showed picture and video of his working models running so it should work.

    So, for the sake of the problem, let us assume that his technique does work for me, then my hurdle is as I stated in my original post, governing the wheel or the pulley at the other end. Some way to maintain the generator head at it's designed RPM target, usually 1,800 to 2,000 RPM with most generator heads, with a waterwheel that will turn at variable speeds depending on current flow conditions. It must be governed. If there is no mechanical means of governing it, then it must be governed by regulation of the flow of water to it.

    I read an older book on waterwheels and I don't remember exactly how it was done, but somehow they used to use the flow rate of the stream to adjust the sluice gate of the sluice up and down, 'automating' the amount of flow allowed to reach the wheel and thusly regulating the RPM of the waterwheel. I don't remember what the mechanism was, I suppose I will have to try to find it again if no other means can be thought of to govern the system automatically.

    But I am betting on there being a suitable mechanical method for governing the system, keeping it running at a constant speed. Modern wind turbines are governed, due to variation in wind speed. Centrifugal clutches are used to run go carts and were used in my youth to run mini-bikes at variable speeds with direct drive from motor to wheel sprocket.

    I saw in a book where a waterwheel company even used a device working on centrifugal force to govern the speed of a Pelton Turbine. The device had 3 disks, a central disk which had the governor system built on it, and the 2 outer disks had the 'cups' mounted on them. When the wheel turned too fast, the centrifugal force caused the central disk to 'push' the outer disks away from it, causing the cups to separate, allowing less water to hit the cups and slowing down turbine wheel rotation. This wouldn't work in my case, as undershot waterwheels and Pelton Turbines work differently, but it did show that they had some type of centrifugal governing system in place way back then - at least for this type of device.

    Is there some type of centrifugal force mechanism that could be used with a waterwheel to speed up and slow down the turning of the wheel to keep it within a certain range?

    A CVT transmission keeps the motor of a car working at the maximum efficient torque the engine was designed for. Could a CVT transmission work in this case? If not, could the concept be modified in some way so that it would work to regulate the turning speed of the waterwheel?

    Assume that all other needs have been met in the design. What I am stuck on is how to govern the wheel or the pulleys or something in the system such that the final pulley hooked to the electric generator head is always turning at the target RPM range - or as close to optimal as possible. And it would be nice if I didn't lose too much of my torque from the stream between the waterwheel and the generator head :smile:. Thanks
  5. Mar 3, 2014 #4
    Have you considered using a DC generator, batteries, and an inverter? No worries of matching rpm. Storage for those times you need more power or need to shut down the wheel for maintenance. And much cleaner power from low to high load. Just a thought. :smile:
  6. Mar 3, 2014 #5
    You should look for proven practical ideas...search online. You seem to have three choices...water, wind, or solar or a combination. Why water??

    I have no idea of relative costs among the three, but home sized solar systems are roughly $4 or $5 per watt.....maybe a bit less these days.....an example of 'green energy' being at a severe cost disadvantage. Here in NJ, USA, a kilowatt [1,000 watts] from the utility is about 18 cents. So without government subsidies it takes about a thousand years for payback of a home colar system....just a joke, but no need to do the calculations, it simply does not pay.

    That is a huge obstacle!! HUGE!!!

    As an example, my home washing machine will not cycle properly when on my emergency gas generator. The electronic controls are apparently VERY sensitive to either waveform shape or frequency or both. When aboard a friend's boat recently, his square wave inverter blew up his KEURIG coffee pot....apparently the digital controls did not like the electric source. It would not work at all. A simple on off no timer electric coffee maker worked fine as will all resistive loads.
    So an electric stove or oven will work fine unless it has digital controls or digital displays.
    50 cycle European appliances do not like US style 60 cycle source power, and vice versa...You can burn up some type electric motors by feeding them low or high voltages or power.

    I skimmed here

    and found some interesting water wheel insights.

    One simplifying idea might be belts instead of gears.....gears have much longer lives but and suited to heavy loads but are probably more complicated to install....

    As noted, a battery ['sump'] can serve to regulate available power somewhat, storing some at low power requirements and providing some at high power times. That's a basis of solar and wind power sourced electric.
  7. Mar 3, 2014 #6


    User Avatar
    Science Advisor
    Gold Member

    I would go with DC and an inverter. It may well be advantageous to regulate the amount of water to a certain point. I would use a PIC processor to sense speed and linear motor(s) to change the flow of water.
  8. Mar 3, 2014 #7


    User Avatar

    Careful - you're mixing your units here. The utility sells you electricity by the kilowatt hour (which is an amount of energy, not an amount of power), the solar cells are usually sold by the watt (which is an amount of power). If your figures are accurate, the solar cells would pay themselves off in about 25,000 hours of full power output (between 13 and 14 years if you get ~5 hours of full power per day).
  9. Mar 4, 2014 #8


    User Avatar
    Science Advisor

    Your engineering design problem needs to be cut down to size.

    How much drop over what distance do you have in your stream?
    How much flow do you have in your stream?

    Once those parameters are known, the solution will be more obvious.

    Solar panels now cost about 1$ per watt.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook