I am using a discarded hard drive as the base of for an anemometer. I have it running and can generate up to 200 mV a.c. that are ( I assume) proportional to the wind speed. The motor has 4 connections to the outside and I can get approximately the same a.c. voltage from any combination of pairs. My question is: Can I use some type of combination ( delta, star etc.) that would increase the generated voltage? This connects to a high impedance circuit that converts the a.c. or d.c. to RS232. A computer plots the result.
Sounds like a dose of FM radio is needed: Amplify the piss out of it and let it clip and clamp so you get a variable frequency square wave.
Of you can use an AGC circuit ...
A PIC, though, will do you no good.
I take it you meant "a signal [whose amplitude] is constant with speed."
Depending on the age of the disk drive there may be a hall-effect tachometer pick-up built into the motor. If it is really old there will be an aluminum ring with slots in it ... with a phototransisor and an incandescent lamp ...
Second this suggestion. With luck you will only have two wires coming from the anemometer with no (almost no) active circuitry up there on the mast.
--
Nicholas O. Lindan, Cleveland, Ohio
Consulting Engineer: Electronics; Informatics; Photonics.
Remove spaces etc. to reply: n o lindan at net com dot com
psst.. want to buy an f-stop timer? nolindan.com/da/fstop/
The trouble with your tachometer is that output voltage is proportional to speed. I have played with such tachometers and found it hard to come up with a reliable pulse detection over the large speed range of > 100 : 1 you need for wind speed.
The opto and hall effect types are easier to use, because they have a signal which is constant with speed. Personally, I would drill a circle of holes in the disc and put a LED and phototransitor on either side of the disc - now you have digital level transitions ready to go into a schmitt trigger (40106 or 74HC14 etc) and then into your microprocesor.
Some thoughts on using the motor windings :
Because the signal is proportional to speed, I would feed it into a filter with diminishing gain as frequency increases - ie an integrator response. That will give you a flat overall response. That means an awful lot of gain at low frequencies. You will have to stop increasing gain below some low frequency or will just have rubbish noise as output.
A schmitt trigger after the filter will give you pulses ready to feed to a microprocessor. Some microprocessor smarts may be needed to reject random transitions at very low speed.
Have you thought of sending the pulses straight down the wire with no RS232 needed ?
The trouble is at low frequencies. Imagine a very low amplitude 1 Hz waveform. You amplify it and clip it. Thus at the waveform zero crossover points, you have a high gain amplifier with input moving slowly through the active range. The output will slam up and down according to bearing wobble, noise, you name it - just like an FM limiter does with no signal. Means that simple edge counting won't give you speed. A micro could work out the difference between long cycles with bouncing edges vs short cycles.
The integrator slows down the low frequencies and does help. I have played with this stuff and though it can be tamed, I reckon its too much trouble for a one-off.
The beauty of the Hall / Opto approach is that the analog signal increases smoothly as the magnetic field / light builds up and diminishes, so a simple schmitt cleanup does the job.
The reed switch suggested in another reply has been used a lot. A small reed switch will do > 1 KHz !! which for 10cm cup spokes is about 628 metres per second or 2261 km / hr !! The reed switch does bounce, but the bounce is fairly consistent across the speed range and you get a nice hysteresis built in. Only 2 wires needed.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.