Curious phenomenon / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Curious phenomenon

1 2 next

Move Thread

LAN_403

TheMerovingian

Thu Nov 12 2009, 02:34PM

Registered Member #14 Joined: Thu Feb 02 2006, 01:04PM
Location: Prato/italy
Posts: 383

In order to prevent people stealing my oranges growing on a tree in my property (with the branches exiting partially my property) i put on a plastic mesh to protect the fruits, but i'm aware it is not enough. Carefully avoiding potentially hazardous and probably illegal or boring solutions (like electrocity or audio/video surveillance) I switched on the noble art of bluffing and decided to build 4 two transistor low power led flashers powered by 9V battery shared by all circuits to simulate some movement sensor security system (people know me, and know what i have build in past). The circuits are very simple consisting in the classical two transistor fashion with a cpacitor and resistors like timing components flashing at a speed about 1 Hz for 20ms. The tolerances of resistors are 5% so i predicted that the phase of the flashes will drift and they will be totally uncorrelated even if starting at the same time.

this prediction was wrong........

All the circuits sharing the same power supply (decoupling 1000uF on battery and 22uF on each circuit) flash almost at the same time, drifting a bit and then resinchronizing after a while without any adjustments!! Even if i intentionally slow one of the circuits to change the phase, after a while they are in phase again. Cannot explain this effect, the circuits have 1meter of wire(s) between them and the circuit proximity doesn't affect this effect. Maybe the power fluctuation induced by led flashing may have a role in this( but the circuits are decoupled....) . They are sinchronized every time even after a night that they are flashing... it is a total nonsense... i will try to power them with individual batteries to understand this phenomenon.

Do you know any possible explaination?

Steve Conner

Thu Nov 12 2009, 02:53PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

It's probably entrainment

This was what people used to sync oscillators before the PLL was invented.

WaveRider

Thu Nov 12 2009, 03:44PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

yes, this is a classic phase-locking effect. Christiaan Huygens noticed this effect in pendulum clocks .. Only a very slight coupling is needed to observe this effect. It can be used in injection-locking of electronic oscillators (as Steve noted)...

In order for this to work, some form of nonlinearity must be present (saturation nonlinearity in transistor amplifier is the classic case).

Many oscillators exhibit this behaviour...even biological ones. I have read that the menstrual cycles of women in office situations will synchronise as a result of working in close proximity over the course of a few months.

Mates

Thu Nov 12 2009, 03:57PM

Registered Member #1025 Joined: Sun Sept 23 2007, 07:53PM
Location: Czech Rep.
Posts: 566

That is quite interesting, because as I understand the entrainment principle (like the classical example with clocks) there must be some kind of connection (communication) between the two oscillators. In case of pendulum clocks the phenomenon was apparently only on those which were all inside the same cupboard. The cupboard was the mechanical link between the oscillators.
But how is in the case you are describing? It would be interesting to try if the same effect will be present in case the two flashing oscillators would be powered from two separated power sources (ideally batteries). That could tell us whether the connection between your flashing oscillators is via cables or via some other more mysterious link...

Regarding this issue, once I saw on TV a document about modern military technologies. They presented two huge metal balls (at least 500Kg maybe more), rotating with high speed next to each other. These balls after some time started to rotate synchronically via the entrainment effect â€“ not a surprise, but what they showed next was somehow too much. They isolated one of the balls spatially (moved it away â€“ donâ€™t remember the distance exactly) and then they hit the one of the rotating ball with a hammer. The second rotating ball was scanned by a laser beam and the hit by the hammer was detectable on the second ball as small change in the rotation speed/direction. In the document claimed that the distance for the two balls "linkage" is unlimited and this principle works on modern submarines as kind of super secret communicator. I know it sounds crazy â€“ but it was on TV â€“ so it must be true

No seriously, did somebody of you heard about anything like that?

MinorityCarrier

Thu Nov 12 2009, 04:54PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

Run them on their own individual batteries and see if this sympathetic effect goes away. Do it for science's sake.

TheMerovingian

Thu Nov 12 2009, 07:20PM

Registered Member #14 Joined: Thu Feb 02 2006, 01:04PM
Location: Prato/italy
Posts: 383

i will try to power it with different batteries. Maybe the perturbation the led load introduces (voltage fall due to increase current consumption) is the "signal" of the sincrhronization. Even slight differences in frequency (5% tolerance resistors) are compensated by this effect.

UPDATE: separating the supplies inhibits the fenomenon, at least on a 2 hour scale, so the perturbation must the the voltage drop induced by led lighting

MinorityCarrier

Thu Nov 12 2009, 09:44PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

Put a decoupling capacitor at each blinker circuit, say 1uF to 10uF, then run them all with a common power source and see if they go sympathetic again.

GeordieBoy

Thu Nov 12 2009, 10:29PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

As Steve C said it is a case of entrainment (or injection locking.)

All oscillators are technically unstable systems, and the only thing that stops their amplitude increasing to infinity is the "something non-linear" that saturates. The same non-linear element that defines the amplitude of the oscillation also makes it susceptible to injection locking.

If you think how most relaxation oscillators work, they charge and discharge a capacitor between some pre-defined thresholds. These two thresholds are usually defined to be some ratio of the power supply rail. As the capacitor voltage gets close to one of these thresholds and the comparator is just about to trip, you can see that it wouldn't take much of a momentary glitch on the threshold to make the comparator trip a little bit early. That is one example of how injection locking can occur.

The way to cure it is to run them off separate supplies, or at least locally decoupled supplies. I'd suggest putting a 100uF decoupling capacitor across each of the timer IC power rails, and then feeding each of these locally decoupled rails from the main common supply rail via a 10 ohm resistor. Then each timer can trash it's own local power supply with glitches but they won't get through the resistors and appear at the other oscillator.

Another option to prevent entrainment is do implement the oscillators digitally. i.e. Program a microcontroller to generate the handfull of frequencies that you want to be close but not the same. Unless you program it into the source code, then digital oscillators won't entrain!

-Richie,

IntraWinding

Fri Nov 13 2009, 04:30AM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

There's an "electronic firefly" circuit out there where each firefly is an led driven by a slow oscillator, just like yours, but with each circuit electrically independent. To promote phase locking each includes light sensors aimed at nearby identical circuits. I daydreamed designing ones with the light sensor being the LED itself - LED's do generate a photoelectric current I found.

Taking it a step further, perhaps you could also use the LED to charge the circuits battery during the day to get a few hours display each night! Make a few hundred built for long life and sealed in resin buttons and distribute them over trees and bushes for years of dynamic entertainment!

Alternatively, move to Malaysia for the real thing

Wolfram

Fri Nov 13 2009, 07:09AM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Check out this

video for a very striking example of entrainment.

Anders M.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob