Joule Thief

Pirate88179 · December 20, 2009, 01:43:18 PM

Xee2:

I am probably showing my ignorance here but, what the heck. My question is are you not creating losses in the circuit by adding the diode, cap and resistor? Would this not alter the true output of the circuit as it would have been before these components were added? All of those components have losses that the original circuit did not have and it just seems to me that this will alter the true efficiency of the circuit that does not normally use them.

My analogy would be if you had a horse that you wanted to see how fast he was compared to other horses and you strapped telemetry and measuring equipment to him weighing 100 pounds, and then tested his speed, one could conclude that that horse was not faster than the other horses except the other horses were tested by distance and time rather than with the heavy telemetry equipment which probably slowed him down.

Thanks,

Bill

PaulLowrance · December 20, 2009, 02:13:52 PM

Quote from: xee2 on December 20, 2009, 12:23:41 PM
@ PaulLowrance

Can you post details of how you measured the efficiency? Schematic, readings, calculations, etc.

I have posted efficiency test before and only got about 25%. Attached is the technique I use for measuring output power. Input power is just amps out of battery times battery voltage. This particulate drawing was meant to show how to determine best number of turns for the coils, which is the combination that produces the highest output power rather the the highest output voltage. But, the same technique works for any output power measurement.

Hi, I've used various methods. This was outlined in the ultracap thread.

Method A:
A Keithley meter was used to measure the DC voltage across a 0.10 ohm resistor to get the DC current. A AM-240 meter to measure the DC voltage across the ultracap, which was at ~ 1.3 volts. See the ultracap thread for the exact numbers. So I started the experiment with the ultracap at ~ 1.3 volts. As you know, the power is I*V. Also the voltage & calculated current was measured from the battery or power supply. I tried numerous experiments using different types of batteries, and also a power supply. The initial efficiency was close to only 5%, but I made adjustments such as changing the cap & adjusting the resistor pot, which got it up to 55% efficiency. IMO it would not take much to get it closer to 90%. Perhaps more windings. A better core. I don't know since IMO the JT is not worth much further investigation.

Method B:
This was also detailed in the ultracap thread. For this method I used a fix resistor as a load instead of the ultracapacitor. This would allow for an effective higher resistance load than the ultracapacitor, thus increasing the voltage coming from the magnetic core while decreasing the current, which means less losses in the transistor, etc. Using the ultracap as an effective load has the advantage of removing the voltage spikes, but a fixed resistor alone will have noticeable spikes from the JT circuit. So for this method I placed a small 402 SMD thermistor on the load resistor to measure the rise in temperature. This experiment was followed by a control experiment where the load resistor was removed from the JT circuit and placed across a power supply where I adjust the power supply till the load resistor temperature increased to the same temperature as in the JT experiment. This provided the amount of power that the load resistor was dissipating. The efficiency was ~ 80%.

Thanks for posting your efficiency measurements so others can see that I am not lying about my measurements of the JT not being cop>1. Your efficiency of 25% sounds very believable.

Regards,
Paul

Pirate88179 · December 20, 2009, 02:31:56 PM

I do not believe that any one here ever claimed that the JT has a COP>1 and I have read the entire topic. So, if you are setting out to prove this I have no idea why.

It would be nice to have COP values to be able to compare one circuit with another, and that would be helpful. I just suspect that the testing methods are altering the real efficiency of the circuit although, by how much I don't know. Certainly not enough to keep it under COP 1 as I don't think a JT is over 1, or even close to 1....yet. (I am an optimist)

Bill

PaulLowrance · December 20, 2009, 02:35:12 PM

Quote from: Pirate88179 on December 20, 2009, 02:31:56 PM
I do not believe that any one here ever claimed that the JT has a COP>1 and I have read the entire topic. So, if you are setting out to prove this I have no idea why.

It would be nice to have COP values to be able to compare one circuit with another, and that would be helpful. I just suspect that the testing methods are altering the real efficiency of the circuit although, by how much I don't know. Certainly not enough to keep it under COP 1 as I don't think a JT is over 1, or even close to 1....yet. (I am an optimist)

Bill

Bill, you know what cop>1 is. People don't have to use that exact lingo to say the same thing. Gadgetmall makes that claim almost on a daily bases. See his last video where he says the battery voltage is increasing in both source & charging batteries. See gadgetmalls claim in the overunity $15,000+ prize money. The requirements for the prize money is very clear.

Paul

Pirate88179 · December 20, 2009, 02:39:29 PM

Paul:

But that is a totally different circuit he is talking about. You just tested a sort of replication of his JT circuit alone. Again, I know of no one that claims COP>1 for just a JT circuit by itself. I just want to keep the facts clear here for others coming into our topic. Gadget's circuit for the OU prize is much more than just a stand alone simple JT so, in my opinion, you are comparing apples and oranges here.

As I have said, I do appreciate the testing you have done over in the b-cap topic. Testing the proper way is good for all of us and will help guide us in the direction we want to go.

Bill