SamSuka
bigclive
bigclive

patreon


Scoping the coil of a perpetual motion unit

These things were called perpetual motion ornaments in the 1970's.  They must have been one of the first electronic ornaments, with a single transistor to make them work.

They still sell them on eBay, and the circuitry is exactly the same, but with newer transistors.

https://youtu.be/VcoC9nYBVpU

I had to get the oscilloscope out for this one, and it revealed a very interesting coil waveform with a significant demonstration of why it's advised to use a back EMF diode on relays and solenoids.

Scoping the coil of a perpetual motion unit

Comments

Interestingly simple design... I'm wondering whether that negative spike at the emitter could increase power consumption by loading the battery for a split second through collector-emitter leakage of the transistor, which is exceeding its max Vce. Also, wouldn't there be an even larger negative spike on the base since it's the same field collapsing inside both windings but the base has more turns?

Gordo

The one downside to modern digital scopes is that they can do so much that it can initially be frustrating to use them. It's a matter of playing with them and getting a feel for the controls.

Big Clive

I really look forward to more scope videos from you. You make this stuff so understandable and I recently got a Rigol 1054Z (after glowing reviews of it from Dave), but I haven't really done anything with it yet.

Phillip Cosgrove

I think that the energy will take the lower impedance path, in this case the 25 ohm winding and diode.

Jeff Larson

It could, but there's very little energy available so it would take a long time to charge.

Big Clive

That actually sounds quite intriguing.

Big Clive

That could be an idea.

Big Clive

Very interesting. Good to see the 'scope being used

Why put the diode across the small winding only? Wouldn't there be a much larger spike across the entire coil, slamming into the base of that transistor?

Circuitmike

Thank you for cracking out the o-scope, very interresting.

I wouldn't mind seeing more videos of you deep diving into the world of electric pendulums!

Jeremy Impson

My grandad had a heap of these things in 1980's, and they were as you said nothing but coil and a transistor. Oddly IIRC they had just a simple coil of wire in a pancake form that was made from two differrent enabled wire (red and yellow) without any "iron" core. This was then just glued to the plastic underneath the pendulum's magnet. I assume they were glued into place then the wires were soldered in-situ when the glue dried. Thanks for hooking the oscilliscope up to it, for something so incredibly simple, it revealed some strange traits.

Anton

Interesting. Maybe put small cap on the base to get a single pulse? I'll have to get (or make) one to play with. I've seen these taken apart before but they never really caught my interest.

Bill Kerr

I guess I went a little over board with my pendulum clock @ https://zeedijker.blogspot.com/2014/04/pendulum-controlled-arduino-clock.html

Zeedijk Mike

What part charges the battery, the spike of positive voltage or the negative transistor clamp❓ Seems wasteful to need the backfeed diode and a MOV to clamp the transient voltage spike

Cerity

i wondered if there would be a video about this after the stream discussion :D

Jesse_Fish_45

I have a clock that runs that way. It has a pendulum that is kicked each crossing by the electromagnet. Here is a website with more information about them. https://sound-au.com/clocks/kundo.html

Elliott Krimchansky

I like John Bedini's SSG motor design a lot better. They cheapened out big time on the electronics. SSG works with only one pulse, so it's more efficient. I built one about 15 years ago, and played with it for weeks. It was such a fascinating device and I learned so much about pulse motor/generators. I still have it somewhere!

MarkM

When a pulse is created to have a very sharp turn on (or off), it will create ringing as you observe. Used in diagnosing underground cables. I think they do this to charge the battery back with the pulse. One test would be to let it run with, and then without the diode, and drain down 2 new batteries, and see which one last longer. Warning; this can take weeks on each.

MarkM

John Bedini's school girl motor (SG) was claimed to be a free energy machine. Then it was his pendulum motor. Of course it's all bullshit. Nothing but pulse motors set to teeter totter at resonance. Extremely low energy use, but also no torque to speak of.

MarkM

I often wonder why "top down" video producers don't work on toughened glass or polycarbonate tables, so you can put things underneath, giving more space without moving the camera.

Cool from Keysight that they're giving you that scope. And I guess a handheld Hantek DSO1060 (very affordable) would still be the more suitable option in your videos.

I was wondering when a scope would appear.. Any chance you could pull it out and revisit the Quartz clock pendulum movement. I am curious what it would look like I suspect the intent of the circuit is similar to what this one does.

Joe Watt

I loved this exploration! That waveform changed so dramatically with just the diode added. Yes I think they missed a trick or over-economized or both.

Michael Thompson

I'd wager that it does, but that being said no doubt the circuit could be further refined.

Michael Thompson

Does the addition of the diode suppress the awful RF interference that these generated?

Andrew Donaldson

How much energy is in that back emf spike? 80 volts but how many amps? Can it charge a cap through a diode? Or too little amps?

Hey Clive, I believe that 20us delay right before the large negative voltage spike may actually be due to the time it takes the signal to propagate through the 2k ohm section of coil! So, the first small dip when the transistor turns off (you suggested it was magnetic inertia) I believe is back EMF from the 25 ohm section. Then there's a 20us delay; based on principles of time-domain reflectometry, I'll take a wild guess that the signal travels through the coil quite slowly at maybe 5m/us, since it's a massively high-impedance transmission line. So a 20us delay suggests there is roughly 50 meters of wire in that coil (I'm thinking the signal might travel up the coil, then reflect back, effectively doubling the distance? Not really sure about the physics going on there). But, while probably inaccurate, that guestimation seems to be about the correct order of magnitude.

Cant find it, I'm either mistaken or it's because trying to use mobile YT. Any case, the solar flower is similar. https://youtu.be/pwR--SBEM9U

Kadah

I think ElectronUpdate decapped one of these a few years ago. His might have been a bit more complicated.

Kadah


More Creators