That's why I asked as to what was in the middle of the coil that I saw. The sketch I found just shows a magnet being moved over the coil. The magnet on the end of the flame blade isn't going to be doing a great deal of moving relatively compared to a pendulum, very short radius of swing from its axle. Is there an iron element in the center of the coil that might be moved by the magnet? Also, I keep seeing these kinds of oscillators with a capacitor (which I expected but is missing here). Balancing all of this out to make it work is quite a little feat, even as simple as it looks. I am impressed.

I don't understand the reference of your comment. Which circuit are you talking about?

 

The TS coil. The blade of the "Flame" is not very large and the distance from the pivot axle to the magnet on the end of the blade is very short. Therefore there is not a great deal of movement across the coil as there would be from a long radius pendulum (in comparison). So... Could the magnet also be moving an iron mass in and out of the coil? Can't tell from the photos so far.

 

The TS coil. The blade of the "Flame" is not very large and the distance from the pivot axle to the magnet on the end of the blade is very short. Therefore there is not a great deal of movement across the coil as there would be from a long radius pendulum (in comparison). So... Could the magnet also be moving an iron mass in and out of the coil? Can't tell from the photos so far.

So, let me revise what I originally said. I didn't spend quite enough time analyzing. While it looks very much like a version of the pendulum circuit I showed above, there are some problems with that idea.

1. There doesn't seem to be connections for a sense coil, which would be necessary for the scheme I had in mind.
2. The period of the small pendulum would be quite short, and it would move very quickly (maybe it does).
3. It might or might not be self starting and I expect the candle would be.

I was a little too quick in my conclusion, though it still might be the case.

 

It pretty much agrees with the circuit I posted, except it is missing a capacitor (which I expected to see in this type of circuit). The circuit I posted is missing the LED but it is easy enough to add one as the load. IF the TS device is actually using a magnet to move an iron mass in and out of the coil.... Not sure exactly how but would think it would be of some aide. Curious device!

 

If the coil had a core then the dangling magnet would be liable to be permanently attracted to it and fail to swing.
I built a clock from balsa wood with a magnet in the bottom of the pendulum and a small circuit driving a coil underneath. It was a single coil but used two transistors. I don't know how it is done with just one transistor. Could the 'transistor' actually be an IC?
There are several designs on WWW.

 

Indeed, and they all use a capacitor which this circuit is missing. Or did I miss it?

 

hi Sam,
I am thinking this LED;s internals look a little unusual, I suspect the 1Hz timer is in the LED.
E
Clear view of the PCB track

 

Sry E, I can't see it in that blurry pic.

I suspect the 1Hz timer is in the LED.
E

In the LED itself or in the circuit? Some clearer pics to try and follow the tracks and scribble a schematic would help. I think we have pretty much figured out what and how, just not the details. LOL Caught my interest as a new trick for an old dog!

 

Hi all,
Posting the high-res close up images of circuit. Based on the discussions going here, I guess this is one strange circuit design...

On top of the circular core, it mentions "LT"


Another view :
In the middle, it says "L1" (inverted in the pic).
And in the sides, "089" and " 151"

 

hi Sam,
I am thinking this LED;s internals look a little unusual, I suspect the 1Hz timer is in the LED.
E
Clear view of the PCB track

The LED has nothing inside other than basic components.

 

Could you take a sharp photo of the LED that shows the internals from the top and side(s)?

 

OK, the L1 marking is clear but the results for it very confusing. Could be voltage detector, PNP, NPN, etc...

 

OK, the L1 marking is clear but the results for it very confusing. Could be voltage detector, PNP, NPN, etc...

I think it might be a MOSFET. Could the gate capacitance be used to oscillate?

 

OK, this "Voltage Detector" IC is a new one for me. If it is in fact what it is...

 

Based on the L1 and the SOT-23 shape the 3 main components are the V Detct, NPN, and PNP. Not coming up with an exact match so grabbed a typical.

The SMD components marking codes database - Starting with 'L1' (yooneed.one)

 

Could you take a sharp photo of the LED that shows the internals from the top and side(s)?

Sure. Here they are...

 

Thanks. It does appear that only the LED die is in the cup.

 

Based on the L1 and the SOT-23 shape the 3 main components are the V Detct, NPN, and PNP. Not coming up with an exact match so grabbed a typical.

The SMD components marking codes database - Starting with 'L1' (yooneed.one)

This is also marked L1 as well as a few other MOSFETs. http://www.hotenda.com/datasheet-html/1542489/1/BSS670S2L+H6327.html

 

This is what I get attempting to trace the circuit:


If this is correct, the SOT23 is an IC that pulses the coil every 1 second.

Bob