Let's Make Robots!

Help! Recalcitrant Jiggle Bot

companion bot...jiggles once in awhile to keep you company


Help! What am I doing wrong? I have checked the connections for bogus soldering, checked the circuit plan a zillion times, checked the transistors to be sure they were in the right place, checked with a battery that the motor works and that the wiring worked in the circuit, checked to see if the recycled solar cell was giving juice.


BUT what I did not follow in the diagram is the use of just one capacitor. Are there hooked up screwy? I thought they were in parallel...and I thought that was OK. I just wanted a good strong motor response to cause the jiggling.

The negative cap leads are on a shared wire that leads to what it should.  Ditto the pos cap leads.


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

(Like JAX says, the caps will accumulate the 0.6v trickle all day long into the bucket, but at the end of that day, it will still be a bucket full of 0.6v. )

How do solar swingers manage to work even firing every now and than in a shady position using 6 1/8" square solar cells generating about 1/2 volt @ 1.9 mA in sunlight?

this message is written out of ignorance so please don't get offended

They are usually more productive than that.    I suspect there is something wrong with the panel if it's only generating 0.6v.  that size - 30mA @ 3v would be more like it.      

Regarding the 6" square panel, I have several of the epoxy coated panels.   some of them have erroded around the edge and dont perform.  some of them generate 100+mA @ 10v or 1 watt.    you can do a lot with that.

looks very cool! your robots totally looks like an avant garde art! cheers!

School vacation is coming up next week so I can fix my non-jiggling bot one way or another. 

Thought some of you would be interested in the art that inspires my work...these two always make me smile.

Paul Klee's " Twittering Machine"


While germanium may be hard to obtain, Schottky diodes are in alle the catalogues. They also have very low forward voltage drops.

"in the range 0.15 V to 0.46 V"

What sort of circuit change did you have in mind with the schottky diodes? --just the motor and diode in series across the capacitor? That might do it.

I am looking at a pdf for some schottky diodes (http://www.ixyspower.com/images/technical_support/Application%20Notes%20By%20Topic/FREDs,%20Schottky%20and%20GaAS%20Diodes/IXAN0042.pdf) and on the last page (pp.4) the junction voltages listed range from 0.32 to 0.8 volts.  This is only one manufacturer, though, so maybe another mfg. will have ones with even lower junction voltages.  You mentioned some in the 0.15 v. range.

I had no additional suggestions beyond replacing germanium with schottky. The seem to have the same advantage (low V drop) but are easier obtained. I hope. NB: my wisdom is none. I was merely quoting wikipedia. The 0.15V version may very well be a rare exception. Nullifying the availablity argument.

It's just another (new?) word to help Emma google stuff up. Or anyone else.

Ok, that's cool.  I also thought of Shockley diodes, but I don't believe they work at that low of a voltage, so I had not brought them up. Basically they can cascade on at one voltage and then stay on until the voltage drops and they switch off.  The cap would then begin to charge again, until it reached the turn-on voltage.

The equivalent circuit is almost identical to the circuit that was already being used.

If there were any that had a Vs 'on' voltage of about 0.6 v. they would work perfectly.

While the easiest and best method is to add more solar cell(s), I was just thinking about this project and thought of a different way to make it, which should work at the prescribed 0.6 volts by using germanium transistors for switching. I looked around the internet and find they are still available.

What you could use is a very tiny coil, (probably a coil you wind yourself on a short piece of iron nail). That will go where the motor sits in the circuit now.

Basically, wind a tiny coil out of the finest (enamel insulated) wire you can find.  The finer it is, the more turns of wire you can make in the same space, increasing the magnetism*, so it will pull in even with only a couple tenths of a volt. The piece that pulls in will need to be a small piece of iron (the head of a nail, perhaps?) that is tied (super-glued) to the antennae by a "wiggle bar".

The iron piece will not move far (maybe a 1/16 of an inch at most) because of the very small magnetic field, but if done carefully, the movement it does have will wiggle the antennae.

However, another thing you will need to do is change the silicon transistors (2N3904 & 2N3906) to germanium transistors (one NPN -arrow point out and one PNP -arrow points in) [or replace them with a germanium unijunction transistor].

As I recall the magnetic flux is equal to (the permiability of the core it is wound on) times (the current) times (the number of turns of wire) divided by the length of the coil (shorter and fatter is better). Voltage does not enter the equation other than indirectly due to its limiting the current through a given resistance coil.


I hope that helps some.

Dan M

I agree. If the solar panel is only generating 0.6v in sunlight, that's the problem. Like JAX says, the caps will accumulate the 0.6v trickle all day long into the bucket, but at the end of that day, it will still be a bucket full of 0.6v. Extra food for thought: a solar panel in the shade or darkness will drain your caps or a battery since it will read 0v and volts will flow in reverse from battery or caps to solar panel. So depending on the the project, solar panel circuits should have a diode to prevent reverse flow. But, like an LED, diodes will cost your another 1.6v volt to operate. for example, if your solar panel generates 6v and you put a diode in series, you will yield 4.4v downstream.