Magnetic Anemometer?

Discussion in 'General Electronics Chat' started by pntrbl, Jul 20, 2008.

  1. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Years ago in another life I worked for a company that made disc drives. The heads operated on a spindle and for servo feedback there was a round magnet mounted on the spindle too. The magnet rotated back and forth inside of a fixed coil and created electricity in direct proportion to the speed and direction of the spindle.

    I'd like to make an anemometer using the same principle. I won't need rotational direction but the faster the wind spins the magnet the more juice I oughta see .....

    Seems reasonable. :p

    I've googled toroidal coils so far but all I've found are transformer types. I can't see where I'll need a secondary, but I'm always open for opinions!

    Wind my own? I've got a lathe so I can turn up a precise core to wind around. I'm thinking neodymium's for magnets and hopefully I can generate enough voltage to not have to climb the pole.

    Any and all thoughts are welcome.

    SP
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    If you don't want to have to wind your own stuff, you could use a small DC motor as a generator. It already has the windings and commutator set up for you. All you'd need to do is build the propeller or whatever you're going to use for making the thing turn.

    Just for the heck of it, I connected up a small 24v DC motor I've had sitting around to a variable speed electric drill. It's about 1 1/2" in diameter and 1 1/4" long, not including the shaft. Voltage output seemed to vary pretty consistently with speed, went up to 8v output and decided that was enough.

    Stepper motors will generate some voltage too, and that way you could count the pulses. They'll have more drag than a regular DC motor, though.

    You'd probably do reasonably well with a motor rated for >12v.

    Do you have a spare auto heater blower motor around? Give that a whirl. You'll need a good sized fan to turn that beastie, though.
     
  3. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    I'm looking to measure wind speed and I need to be reasonably close so some calibration will be in order. I can just see me hanging out the window now!!! LOL!

    Never thought of a DC motor and why must I always over complicate everything?:rolleyes:

    Fresh out of blower motors tho and man is that a sore spot. I've had about 5 in my current van over a period of years. All new manuf Siemens but "Hecho in Mexico" and I wish there something better out there.

    Wiper motors? I got some of those. For blades I'm gonna dig up some soup ladles and bend the handles. Right now I'm gonna dig up a flashlight to see how hard it'll be to get me a wiper motor! :D

    SP
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Hmm - soup ladles is an interesting idea! Twist the bowls 90° from the flat handle? Or do you have the kind with the round handles?

    Wiper motor? That's a maybe. I was thinking like windmill-sized fan for the blower motor. Soup ladles won't give much torque; you will need a pretty small motor for those to be able to turn it at a decent speed.

    I have some of these motors:
    http://www.goldmine-elec-products.com/prodinfo.asp?number=G16810
    Didn't get mine there, and they cost more than that sad to say. But that might be a decent choice for what you want to do. Since you'll be using it as a generator instead of a motor, you could probably put the thing inside of a small can to keep the weather out.

    Electronic Goldmine has a $10 minimum +shipping order policy, but they have lots of neat stuff.
     
  5. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    Google "homemade anemometer dc motor".
     
  6. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Like most things in my life Sarge, I can't take credit for the soup ladles. Saw it on the internet, :D, but yeah, get 4 matching ones at the dollar store or wherever and bend the handles at 90 degrees.

    Surprised myself when I found a blower motor on a non-AC van that I never robbed but alas, it seems to have too much stiction. 1350 rpm with a drill motor did 6.5v and it seemed very sensitive to any rotation, but it takes a fair amount to get it moving. I'd probably need to find 4 foot long soup ladles.:eek:

    I beamed one of those Johnson motors up from Goldmine. That's cheap enough for a test.

    SP
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Oops, they have a $10 minimum order.

    Have a google at Ron H's suggestion. I landed on this page:
    http://www.otherpower.com/anemometer.html
    They tried the small DC motor route, and found that the bushings wore out fairly quickly, and the response was non-linear; when it got up to a certain speed, the voltage generated tapered off rapidly. There's also their point about the length and size of the wire; the output will change with wire size/length.

    Actually, bushing wear was one of the reasons I pointed that particular motor out. It's fairly long, so the loading on the bushings will be less than that of a short motor. But naturally, precision sealed ball bearings will last a heck of a lot longer than bushings will.

    Balancing your soup ladles is going to be tricky. That'll have a major effect on the life of the bearings - not to mention that if you're off by more than a smidgen, the whole thing could vibrate itself to death.
     
  8. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    I had googled DIY anemometer before. Including DC motor brought up a whole raft of different results!

    We just discovered a dead CD-Rom. Must be a motor in there somewhere .....

    SP
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, if you really wanted to make your own...

    I think I'd make one out of a couple of precision SS sealed ball bearings with 1/4" center holes. Use some 1/4x28" threaded brass or SS rod. Make up a bracket using some 1 1/2" aluminum angle. Use locknuts on the threaded rod to make sure that the threaded rod was locked to the bearings' inner races. Use a galvanized steel nut on the bottom. Put a Hall-effect sensor with a magnet on the other side of it near the steel nut. For every rotation of the threaded rod, the Hall-effect sensor would output 6 pulses.

    Rather than using the soup ladles, I think I'd try for something lighter in weight. Sure would make it easier to balance. Using the round half of plastic Easter eggs was a novel idea, but not very durable. Trimmed-down rounded end caps for PVC pipe seems like a possible idea; sure would be more durable than the plastic eggs. Easy to work with, too. They do have cheap black plastic soup ladles around...

    It wouldn't have to be a complete half-hemisphere; basically just convex on one side, concave on the other.
     
  10. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    For reduced weight, what about plastic soup ladles instead of metal ones? A heat gun or big hair dryer should facilitate bending.
     
  11. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    This idea's looking better all the time Sarge. Per my original idea I got some powdered iron toroids and found out how difficult winding a toroidal coil really is! LOL! I haven't completely given up yet, but right now I'm stuck on how many turns of what size wire to get a suitable output. Basically shooting in the dark at a difficult target .....

    Here's the setup;

    1"x1" neo N50 cylinder magnetized axially. Got a 3/16 hole in the center.

    The toroid is also 1" tall and has 1.250" on the ID, 2" on the OD. There will be a .125" air gap to put wire in.

    Probably looking at 30 rpm in a 1mph breeze.

    Output wise I'd like to feed a ucontroller analog input that will be maybe a 100 ft away .....

    Am I trying to defy the laws of physics once again?:rolleyes:

    SP

    p.s. Walmart calls these coffee scoopers but they sure look like anemometer blades to me.

    [​IMG]
     
    Last edited: Jul 25, 2008
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Holy cow, those toroids are HUGE! :eek: You're not building an anemometer, you're building a magneto. ;)

    The trouble with a magneto is that at very low RPM's, there may not be enough voltage generated to trigger your circuit. Conversely at high RPMs, you'll have to limit the voltage using a clamp arrangement (think Zener diode). The trouble with clamping the voltage is that it will put a load on the shaft (think brakes) that will cause the output to be non-linear, similar to the difficulties that they were talking about with using motors as a generator on the web page I posted earlier.

    Hall-effect sensors are great. Three connections; Vcc, ground, and signal out. They're pretty cheap too, around a buck each. Really reliable. They're widely used in automotive systems nowadays, particularly in ABS (automatic anti-skid braking) to check the relative speeds of the four wheels. If the wheel speeds are more than about 10% different, the ABS automatically decreases the pressure in the braking system to stop the wheel from skidding.

    Some Hall-effect sensors report the relative strength of the field, and some are strictly proximity (a yes/no kind of thing) - you want the on/off type.

    After giving the "detecting points on a nut" idea more thought, it would be just as easy to use a chunk of say, 1/16" or 1/8" flat steel stock with a hole in one end, and on the end towards the hall sensor just grind to something more or less like a "V" point. That would give one pulse per revolution, and have practically no rotational resistance.
     
  13. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Well if you're gonna be a bear ya might as well be a grizzly! :D Liable to burn down the house tho huh? I have this tendency to try and re-invent the wheel .......

    So hall-effects it is. I had the impression they switched from a passing magnetic field, but after spending some time at Wikipedia I think I've learned they only have to be in the presence of a magnetic field. Then any passing piece of ferrous material makes them switch. If I'm wrong in that assessment would someone please advise ....

    Throwing a dart at the online Mouser catalog I'm currently looking at an AH175. In the general description of the datasheet I see language that leads me to believe it might create it's own magnetic field. And I quote;

    "The device includes an on-chip Hall voltage generator for magnetic sensing, an amplifier to amplify Hall voltage," and etc. If a Hall voltage is part of a magnetic field I'd think it would be making it's own flux lines, but again, if I'm reading more into that than is there I'd sure appreciate a heads up.

    IF so tho, the 6 points on a typical nut are looking better all the time. I assume it'd just a question of the right proximity ....

    Thanx to one and all and you in particular Sarge, for any advice.

    SP
     
  14. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Why use a 2x4 when a 4x12 will do? ;)

    Basically, they need a small magnet on the back side of them, and then they'll detect ferros metals passing near the front of them. If you pop the distributor cap off a '75 thru '83 or so GM distributor, you'll likely find a "reluctor" instead of a 6 or 8-lobed cam. Pointy things.

    Yeah, you're reading a bit more into it than is there.

    If you look at it closely, it shows a N magnetic pole on the side of the IC away from the marking. So, glue the N pole of your magnet (using original J-B Weld) to the back side of the Hall-effect sensor (the side that doesn't have any lettering.) If you don't know which side of your magnet is N, then hang it on a string. The N pole will point towards San Francisco. ;)

    Compared to the magnet, any other ferros non-magnetic material will be S.

    Correct! You'll have to "tweak" it to get reliable switching.

    That's why I suggested the single "V" shaped pointer in my prior post. It would be much easier to adjust the Hall-effect sensor.
     
  15. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Thanx for the clarification Sarge. Me and JB been friends for quite awhile and I got some of them pointy things out back.

    SP
     
  16. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The original J-B Weld is one of the very few permanent epoxies that are marketed to the general public. J-B Quick (from the same manufacturer) has less than half the strength, and isn't as good over temperature. About the only problems with the original J-B Weld is that it needs to remain undisturbed for the entire curing period (no biggie if you can just leave it the heck alone for a whole day) and won't hold up at temps less than -55°F (yes, I've tested it in an environmental chamber. It cracks.)

    If something's worth doing, it's worth doing right - preferably, the first time. ;)
     
  17. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Got some pointy looking things for hall-effect switching;

    [​IMG]

    But as usual I've got some more questions regarding the AH175 I plan on triggering. This device is gonna be on a pole up in the sky. By the time I route things around to get to a data logging setup I might need a hundred foot run. Will an open collector with 25mA of current sinking ability be able to do that?

    SP
     
  18. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Sure, why not? 25mA at very low frequency. You could even use phone wire, but I think I'd go for something more rugged, and preferably stranded wire. Stranded will hang together better if there's vibration. Run the wire down the north side of the pole; it'll spend less time in the sun that way. I'd anchor it every foot to 18", and stretch it snug between anchor points to keep it from flapping around in the breeze - either that, or put it in some electrical conduit. PVC conduit is cheap and really easy to work with.

    I was doing some more thinking about twisting the handles of those coffee scoops. The trouble with that idea is the section with the quarter twist is going to act a bit like a propeller, and it will cause the assembly to vibrate as it turns. I think you would be better off to snip the handles clean off at about 1" long, bend the remainder of the handle so it sticks straight up from the bowl and drill a hole through it, balance them individually for weight, and use something like 8" or 10" long bolts with threads just at the end. Use a nut on either side to secure it.

    I think four scoops is overkill. Three will work just fine, and it'll be quicker to respond to changes in wind speed, as there will be less inertia.
     
  19. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    25mA for 100' seems like asking a miniscule amount of energy to do a whole lot of work ... but if the Sgt. says so ... I believe it.:D

    Would a pull up back at the power supply be a good/necessary idea. I'm thinking 12v on the supply. 12/.025=480 ohms but 25Ma is the max. 510 ohms?

    We're thinking alike on those flat arms. Currents plans are for 3/8 Alu rod. I'll snip the flat arm off at an inch and slit the end of the Alu to accept the stub. Probably use 2 countersunk screws for attachment because it's all about overkill. :rolleyes:

    I've got some 1.5"sq. Alu for the hub. Piece of cake to center up and get perpendicular on that, but it does 4 mean vanes. To finish and get the hub aero ready, lol, I'll turn the corners off and probably taper the bottom extended portion down to 1.25". I'll post a pic when it's ready to fly .....

    SP
     
  20. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    100' of #24 conductor, there and back again, will have a resistance of 5.14 Ohms. Voltage drop across the line will be on the order of an eighth of a volt.
     
Loading...