Author Topic: How can you "drive" a wind clock?  (Read 13853 times)

Offline BillTodd

  • Hero Member
  • *****
  • Posts: 782
  • Country: 00
  • Colchester Essex (where the lathes were made)
Re: How can you "drive" a wind clock?
« Reply #25 on: August 23, 2014, 05:30:31 AM »
One important thing to remember is that there is next to no force from the vane when pointing  into the wind. I used to have to replace the tiny ball races in the mariner wind vane regularly (the pot was magnetically coupled to the vane).

If you want and kind of accuracy you'll have to keep the friction down to a minimum.

If you go fully mechanical, think servo mechanisms driven by wind power.

bill

edit. Think how this fan tail is designed to keep the post mill facing the wind.
Bill

Offline John Hill

  • The Artful Bodger
  • Madmodder Committee
  • Hero Member
  • *****
  • Posts: 1977
  • Country: nz
Re: How can you "drive" a wind clock?
« Reply #26 on: August 23, 2014, 05:48:06 AM »
Bill, I am sure that is especially true of units on boats where the axis is rarely vertical but surely a house mounted unit could allow a sizeable vane and get a bit more force from it.


Now for an all electronic solution and that is ultrasonics.  The general idea is that an ultrasonic beeper is mounted several inches from a matching ultrasonic microphone.  Wind blowing past will change the frequency of the received ultrasonic sound waves due to Doppler effect.  Two such units at right angles give wind speed and direction with the appropriate computer processing.  I think Vaisala were the first to bring one of these to market but there are several makers now.
From the den of The Artful Bodger

Offline awemawson

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6183
  • Country: gb
  • East Sussex, UK
Re: How can you "drive" a wind clock?
« Reply #27 on: August 23, 2014, 05:54:35 AM »
Much simpler would be a 'shaft encoder' of the style that has two phases in quadrature. Many decoders available for quadrature encoders 'off the shelf'

If the shaft encoder were 360 pulses per rev then you don't even have to scale it. Here you go :

http://www.ebay.co.uk/itm/360P-R-Incremental-Rotary-Encoder-6mm-Shaft-for-Measurement-Intelligent-Control-/271492847192?pt=UK_BOI_Industrial_Automation_Control_ET&hash=item3f363c1658

You could decode and display the result with an Arduino like this chap:

Andrew Mawson
East Sussex

Offline John Hill

  • The Artful Bodger
  • Madmodder Committee
  • Hero Member
  • *****
  • Posts: 1977
  • Country: nz
Re: How can you "drive" a wind clock?
« Reply #28 on: August 23, 2014, 06:01:59 AM »
Trying to get back on track here to produce a practical solution for Eugene...

A magslip/selsyn/synchro solution is the orthodox way to go but I fear the majority of commonly available devices will have insufficient torque to move anything heavier than the lightest of indicator needles.


.....I think Eugene wants a nice analog clock face to fix to the wall?
From the den of The Artful Bodger

Offline awemawson

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6183
  • Country: gb
  • East Sussex, UK
Andrew Mawson
East Sussex

Offline John Hill

  • The Artful Bodger
  • Madmodder Committee
  • Hero Member
  • *****
  • Posts: 1977
  • Country: nz
Re: How can you "drive" a wind clock?
« Reply #30 on: August 23, 2014, 06:54:10 AM »
I think a rotary encoder could be used but it would need another bit to drive the wall 'clock'.  Something like a rotary encoder at the wind vane and another on the wall clock and some electronics to compare the two and a stepper motor to drive the wall clock.  That would at least be a fairly easy electronic and software project?

The encoders used would have to be absolute as zeroing the direction indication would be a chore and likely somewhat impractical.
From the den of The Artful Bodger

Offline awemawson

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6183
  • Country: gb
  • East Sussex, UK
Re: How can you "drive" a wind clock?
« Reply #31 on: August 23, 2014, 07:03:11 AM »
Rotary encoder up aloft suitably weather proofed with a wind vane:
Four wire cable down to ground level:
Arduino taking input from encoder:
Small Stepper Motor with pointer directly on its shaft driven by the Arduino


There is library code for the Arduino to accept quadrature encoded rotary data to my certain knowledge, and I'd be very surprised if there wasn't also library code for driving stepper motors as it's only a question of outputting a sequence on four outputs.
Andrew Mawson
East Sussex

Offline John Hill

  • The Artful Bodger
  • Madmodder Committee
  • Hero Member
  • *****
  • Posts: 1977
  • Country: nz
Re: How can you "drive" a wind clock?
« Reply #32 on: August 23, 2014, 07:06:47 AM »
I am sure you are right but what about zeroing the direction?  Are those encoder absolute?  You also need to 'know' the position of the clock needle at turn on.
From the den of The Artful Bodger

Offline awemawson

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6183
  • Country: gb
  • East Sussex, UK
Re: How can you "drive" a wind clock?
« Reply #33 on: August 23, 2014, 08:23:23 AM »
That one I pointed to is incremental, but it doesn't take much to see which way the wind vane points and turn the pointer to suit  :scratch:
Andrew Mawson
East Sussex

Offline BillTodd

  • Hero Member
  • *****
  • Posts: 782
  • Country: 00
  • Colchester Essex (where the lathes were made)
Re: How can you "drive" a wind clock?
« Reply #34 on: August 23, 2014, 12:37:20 PM »
Guys, I think the design brief puts us into the relm of mechanical or at best electro-mechanical. CPUs aren't an option here ;-)

Bill

Offline awemawson

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6183
  • Country: gb
  • East Sussex, UK
Re: How can you "drive" a wind clock?
« Reply #35 on: August 23, 2014, 12:51:11 PM »
Guys, I think the design brief puts us into the relm of mechanical or at best electro-mechanical. CPUs aren't an option here ;-)

OK then Bill, back to the Pitot tubes - mechanically very simple
Andrew Mawson
East Sussex

Offline Eugene

  • Full Member
  • ***
  • Posts: 151
  • Country: gb
Re: How can you "drive" a wind clock?
« Reply #36 on: August 23, 2014, 03:41:30 PM »
Quote
.....I think Eugene wants a nice analog clock face to fix to the wall?
Yiss, Sorr, dats roight so t'is.

Im guessing at an 8 - 10' dia clock face. We have a cheapo wireless weather station that has an LCD display; no wind direction function (they are available) but it's surely possible to render the input in analogue form. If you are cleverer than me that is!  Andrews rotary encoder / Arduino /stepper sounds great but nailing it together would be beyond me without a very close to hand mentor. However I'd be quite happy for someone else to put the circuit together and then fit it myself.

Eug

Offline BillTodd

  • Hero Member
  • *****
  • Posts: 782
  • Country: 00
  • Colchester Essex (where the lathes were made)
Re: How can you "drive" a wind clock?
« Reply #37 on: August 23, 2014, 04:48:53 PM »
do you want the pot and motor shown in my video?

you may be able to fix / clean the pot (i have not even tried to open it)  I think the motor is ok but may need a front bearing.

PM me your address and I'll post it to you.


Bill
Bill

Offline Eugene

  • Full Member
  • ***
  • Posts: 151
  • Country: gb
Re: How can you "drive" a wind clock?
« Reply #38 on: August 26, 2014, 04:44:59 AM »
Bill,

That's extraordinarily kind of you. See PM.

Eug

Offline BillTodd

  • Hero Member
  • *****
  • Posts: 782
  • Country: 00
  • Colchester Essex (where the lathes were made)
Re: How can you "drive" a wind clock?
« Reply #39 on: August 26, 2014, 02:49:24 PM »
ok Eugene it's on its way.

My guess is that the fault with the pot is mechanical. Probably, one of the wipers has come loose - P&G usualy use a sort of wire brush that wipes both the main track and the input track.

bill
Bill

Offline Noitoen

  • Full Member
  • ***
  • Posts: 135
  • Country: pt
Re: How can you "drive" a wind clock?
« Reply #40 on: September 02, 2014, 03:44:06 AM »
It's very easy to make a contactless potentiometer that will last a very long time. Use a LED to shine on a LDR through a plastic disk with a "spiral" pattern painted on it. If you use 3 LDR's in a 120 configuration you can even get that three phase pot mentioned earlier.

Offline Eugene

  • Full Member
  • ***
  • Posts: 151
  • Country: gb
Re: How can you "drive" a wind clock?
« Reply #41 on: September 02, 2014, 04:15:49 AM »
Bill,

The parts are now to hand; thanks a lot. I'll play around and see what I can make of them.

Eug

Offline BillTodd

  • Hero Member
  • *****
  • Posts: 782
  • Country: 00
  • Colchester Essex (where the lathes were made)
Re: How can you "drive" a wind clock?
« Reply #42 on: September 08, 2014, 10:47:59 AM »
I found a wind  vane and arm, it'll be on its way to you, so you'll have no excuse when it arrives.

bill
Bill

Offline Eugene

  • Full Member
  • ***
  • Posts: 151
  • Country: gb
Re: How can you "drive" a wind clock?
« Reply #43 on: September 08, 2014, 05:27:02 PM »
Bill, You are not only very kind, but nuts too!

What voltage / current do I apply to the pot to make things circumulate?

Eug

Offline BillTodd

  • Hero Member
  • *****
  • Posts: 782
  • Country: 00
  • Colchester Essex (where the lathes were made)
Re: How can you "drive" a wind clock?
« Reply #44 on: September 08, 2014, 06:25:45 PM »
From memory, mariner used 10v but it will probably work with 5v. it's a1k pot.

bill
Bill