Jules

I have a homemade desktop swing, with a small permanent magnet on the swing. I want to keep it swinging when pulsed by a hidden electromagnet circuit underneath, so it keeps moving with no apparent help. The swing pulse needed is about 1Hz, which I need to adjust slightly to fit with the natural resonance. I have seen some ideas about an electromagnetic clock pendulum, but don't need this sort of accuracy. The power supply would be an AA cell (or two). Has anyone made something similar or know of a simple circuit?

Diver300

The toys that work like that use the voltage generated by the magnet and the coil to trigger the power pulse to the coil.

You could try that, or you can use and RC timer. I prefer to use a 74HC132 as it only needs a resistor and a capacitor to oscillate. You will probably want a short pulse and you can put a diode in parallel with the resistor to get a very small duty cycle.

Jules

Have you gt a simple schematic of this? Seems very economical. I had thought of a 555 as a monostable giving a short pulse, but needs more components than your fast CMOS idea.

kchriste

Here's a site with a PIC powered pendulum:
http://home.comcast.net/~botronics/index.html#pend
You can also make something very simple with just a single transistor, resistor and center tapped coil. I made a very simple brushless motor this way years ago. I got the idea out of a magazine, which is long gone, but I've managed to dig up the old project and it still works after 25 years or more! I can post a schematic if you want, but I'll have to reverse engineer it first.

__________________
--- The days of the digital watch are numbered. ---

Diver300

Here is the oscillator circuit that will work with a schmitt nand such as the 74HC132. The mark-space ratio is about 50% and you can swing that by adding a diode with a resistor in series, in parallel with the resistor.

Attached Images

osc.gif (2.6 KB, 3 views)

audioguru

A 555 needs a minimum supply voltage of 4.5V (a run down 6V battery) and has a typical idle current of 3mA. AA alkaline cells would be dead in 1 month if the circuit didn't do anything.

A 74HC132 has a minimum supply voltage of 2V (two run down alkaline cells), has a typical idle current of zero and an operating current of 100uA.
AA alkaline cells will last 2.5 years if the circuit did nothing except power the Cmos oscillator.

__________________
Uncle $crooge

Diver300

HCMOS gates tend to use a lot of power when switching, and that includes when the input voltages are near the middle. It is not as bad for schmitt gates, but the current could be quite a lot in the oscillator circuit that I suggested. I have never used them in battery powered circuits where long life was needed.

audioguru

I have 2 chaser projects and a soil moisture indicator project that use 74HCxx Schmitt trigger oscillators and very bright blinking LEDs.
Ones with red or yellow LEDs use two AAA or AA alkaline cells and ones with bright green or blue LEDs use four AA alkaline cells.
The oscillators run all the time. One is at about 3kHz and the others are very slow.

If the brightness of the LEDs is max (extremely bright, at night it lights up the houses across the steet) then the battery lasts for months. If the brightness is turned down every night or is medium then the battery lasts for about 1 year. If the brightness is turned down to zero then the battery lasts for years.

__________________
Uncle $crooge

saturn1bguy

But using a non-CMOS 555 would be silly. An LMC555 will operate down to 1.5V at 50uA supply current typical, both less than the aforementioned 74HC132.

FYI,
Corey

audioguru

Yes, but with a 1.5V supply its typical output high current is a whopping 0.25mA. Even with a supply voltage of 5V its output high current is only 2mA.

I don't use both inputs of a two-inputs gate to make an oscillator because the 2nd input is just adding to the supply current.

__________________
Uncle $crooge