What does your JT do when the lights go out?

[TinselKoala]

We've all been amazed at how our various JTs run lights at low voltage. But the lights always eventually go out as the battery voltage drops below some minimum. So we put in a fresh "old" battery and get more light.

But what happens when the lights go out? Does your JT keep on oscillating.... and for how long?

I have here a basic JT that is Dark Alchemist's (Legalizeshemp420) mod of Lidmotor's Penny circuit. I call it the DALM thing. It is running on a single tiny LR41/AG3 button cell and it has 4 blue LEDs in series as the load. The battery started out at about 1.6V with brilliant lights and the lights went out at around 610 mV. But the circuit still keeps on oscillating.

I have a MPSA18 transistor in there right now. It seems to do the same thing with the BC337 but I don't yet know for sure about that transistor.

I warmed the battery with the soldering iron several times. This raises the voltage back up over the LED threshold and gives brilliant LEDs again, until the battery voltage drops below 610 mV or so. But the circuit keeps oscillating, with collector voltage peaks about 4 or 5 times the battery voltage, and a "mean" oscillation value returned by the scope as about 10 mV higher than the battery voltage.

So I just let it run, hooked up to the Simpson DMM and an HP 5361A frequency counter. The thing runs and runs, and as the battery voltage drops the frequency rises. It took two or three days for the battery voltage to drop down to its present level (under the load) of 320 mV or so. At this level something very interesting starts happening.

The DALM thing will stop oscillating when the battery gets low. When it does this, it removes the load from the battery, I think... and the battery voltage begins to climb, at about 1 mV/second. When it gets up over 400 mV or so it slows a bit but at 480 mV it is still rising at about 1 mV per 2 seconds. But... at almost exactly 481 mV it spontaneously starts oscillating again, which loads the battery, which drops immediately to about 350 mV and then slowly continues down from there. The oscs start at about 220 kHz and rapidly increase, and the thing cuts out and stops oscillating at something close to 800 kHz. And this cycle repeats, over and over. It takes a few minutes, and I think that the time interval is getting shorter.
It has been doing this for at least a full day.

So.... I'd like to know, has anyone monitored their JTs "after dark"? What does your JT do when the lights go out?



--TK

[Pirate88179]

TK:

Sorry, I have no info on this as I subscribe to the method you mentioned in your post...lights go out, insert new "dead" battery.  Now, I will have to check this and see.  I have many different JT circuits here and it would be cool to see what they do in this mode.

It appears that your blue leds are soldered in to you circuit board so maybe you can't do this easily with that circuit but, it might be cool to have the ability to remove one of the 4 leds and see what happens...then, when that stops, remove another one, etc. in order to see how low your input voltage can be with those transistors and still fire up.  Then, with only 1 led, when that goes out, see if it continues to oscilate below that level?

This is a very interesting discovery you have made here as I do not know of anyone else that would continue any type of tests once the lights go out.  Great work.

Bill

[IotaYodi]

That is interesting. You should isolate it from any external electromagnetic pulses from house wiring or other sources to see if that makes a difference.

[TinselKoala]

Well, I don't know how to isolate it completely and still monitor it with a scope, DMM or frequency counter.   

But it's still doing the same thing now. I've just left it alone. The voltage rise when it's not oscillating is getting slower, and the frequency rise when it is oscillating is rising faster, so it must be running out of energy. It takes about 2 or 3 seconds per mV to rise now but when it finally gets to 480 mV or so, it starts oscillating again. 

I haven't heated the battery lately, I think I'll try doing that again.

[Lakes]

TK:

Sorry, I have no info on this as I subscribe to the method you mentioned in your post...lights go out, insert new "dead" battery.  Now, I will have to check this and see.  I have many different JT circuits here and it would be cool to see what they do in this mode.

It appears that your blue leds are soldered in to you circuit board so maybe you can't do this easily with that circuit but, it might be cool to have the ability to remove one of the 4 leds and see what happens...then, when that stops, remove another one, etc. in order to see how low your input voltage can be with those transistors and still fire up.  Then, with only 1 led, when that goes out, see if it continues to oscilate below that level?

This is a very interesting discovery you have made here as I do not know of anyone else that would continue any type of tests once the lights go out.  Great work.

Bill

Tk could have his arduino monitor the led brightness and/or battery voltage and automatically switch out the leds.