Joule Thief

[xee2]

@  hazens1

I noticed something when measuring using capacitors. Appears that my DMM does not measure peak voltage since when I remove the cap the voltage measures 70v and the cap measures 100v. I'm guessing the cap charge maxes out at the peak pulse voltage and the DMM is measuring an average voltage rather then a peak voltage when measuring pulsed DC across an unknown time period    I checked my DMM manual and could not find any documentation that concerned pulsed DC measurements.

Bridge circuit (full wave rectifier) charges capacitor across DC output terminals to half of the peak to peak input voltage. When DMM is set to AC it measures the full wave rectifier voltage (half of peak to peak) and then multiplies by 0.707. This would be the RMS voltage of the AC input if the input AC was a perfect sine wave.

When the positive and negative peak voltages have equal magnitudes, then a full wave rectifier will output half of the peak to peak voltage.  When the positive and negative peak voltages are not equal, then the full wave rectifier will have an output equal to the largest peak voltage.

Thus you could have anything between 100 volts peak to peak up to 200 volts peak to peak.

Assuming you are using the capacitor at 100 volts, then each of the 25 LEDs in series is getting 4.0 volts. But the LEDs will probably limit the voltage to 25 X 3.6 volts.

[hazens1]

@Hazens,
good job. 25 in series.
Did you stop when the lights dimmed?
Soon we will be able to light those christmas strings from our own wound toroids!

Nope, just what would fit easily on the breadboard. I need to get some bigger boards. I'm sure that circuit could handle more, same as the 43 in Parallel setup could handle more.


@xee2
Thanks for the info. Makes sense..  Also, I removed the cap after the rectifier, only used it for measuring.


@All
Based on previous test circuits, I've got an idea that should push the 1 inch Goldmine toroid to about 1k volts. Just need to find the time to wire up the config that is in my head and test it. Also going to need to have a plan to measure it since my DMM only goes to 1k. Goldmine has a 35k volt cap and maybe a voltage divider     http://www.goldmine-elec-products.com/prodinfo.asp?number=G16864

[altrez]

@jeanna

Thanks! Oh and BTW i ordered a scope just like yours it will be here Monday! I figured it would be a good one to learn on and should last awhile.

Take Care,

-Altrez

[xee2]

@ hazens1

Also going to need to have a plan to measure it since my DMM only goes to 1k. Goldmine has a 35k volt cap and maybe a voltage divider     http://www.goldmine-elec-products.com/prodinfo.asp?number=G16864

Make you own voltage divider with two resistors. They should be over 100K each. For example, if R1=R2 then the voltmeter will read 1/2 of Vin. If R1 = 9 x R2 then voltmeter reads 1/10 of Vin.

[TheNOP]

@theNOP
I don't know what the 1.04kHz was you were thinking of.
I think it was the MK .8
But that was not 1.04 kHz.
Anyway The MK .8 with 20 up and 20 down across 20 up 20 down in 2 examples of base resistance shows:

we can't compaire this toroids values with the other one you are testing.
too many things are different.

So, I can get higher frequency or higher voltage.
One goes up when the other goes down.

my point is that this quote will not be true anymore when the frequency will begin to be too low for your toroid.


btw
to be able to compaire the effect of removing or adding turns on the jt coils, you must remove/add to the pickup coil in order to keep the same turns ratio.
if you don't, you will only be measuring the ratio difference, without any clues on the real gain or loss.



there is also something else to take in consideration when trying to achive very high frequency.
since we are not dealling with signal amplification,
there is a point where the frequency will limit the currents flow.
this is because of the circuit's resistance, mostly the coil impedance, but also the transistor's internal resistance.

since the currents rise with time and it rise more slowly with greater resistance,
if you don't give it time to reach it maximum, then you are limiting it.
this is where duty cycle can help us to some extent.