STEORN DEMO LIVE & STREAM in Dublin, December 15th, 10 AM

[Omnibus]

@Pirate8879,

Thank you for the kind words. I went to two universities today and called a third one in an effort to borrow a current probe. Unfortunately, it isn't as common as one might think. It appears  some colleagues at the physics departments don't even know such critters exist at all (not that it was a household item for me either when I started this research). Now, while figuring out what to do to verify the reality of the current I'm observing (as I said, a current probe will give a direct answer to that) I used the following schematic to study what might be the threshold frequency where the purported parasitic inductance might kick in. You can see two screenshots taken at 50Hz, 1kHz (current and voltage traces superimposed) and 50kHz (note, I've changed the scale of the one at 50kHz to show you in detail what's going on with the voltage and its current response). As you can see, there is no frequency dependence of the current. The change in shape of the current trace at 50kHz is due to the change of voltage trace shape which is due to the peculiarities of the pulse generator -- current only follows that. As was mentioned before, it is the scope readings that are to be trusted regarding the values of the supplied voltage, no matter how weird they might appear -- the function generator may exhibit peculiar behavior at higher frequencies, as is the case here, apparently. Our main concern are the values of the current which are derived values. Some may say that it may include parasitic inductance voltage values, as discussed, and that would distort the derived current value which is only supposed to be the quotient between the Ohmic voltage drop and the Ohmic resistance across which it is measured. As seen, however, from the data presented, it appears not to be the case and the obtained current values may be trusted at all studied frequencies. That, of course has to be confirmed independently by measurements with a current probe.

[Omega_0]

Joining back after a break due to travel etc.
At this time I've no explanation of this behavior. I guess this has more to do with some weird unknown measurement errors than with true OU.

I've seen resistance changes of +/- 1 ohm only due to loose connections alone. When the resistance is so low, everything becomes unreliable. So for example when you have a low resistance measured precisely with high end meters upto 6 decimal places, it may not remain of that value in the circuit, due to loose connections, heating up during run or moisture etc etc.

So there is a need to use either high values of resistances and voltages or place the system in tightly controlled atmosphere in a lab with gold connections etc. Best would be a calorimetry test.

[Omnibus]

@Omega_0,

As I'm repeatedly stating here all the questions as to whether or not this is an OU effect will go away when current probes are used instead of passive ones. The culprit is the current. It should not be overestimated which may happen if possible induction isn't accounted for. The problem is that current is in fact cubed at the output while at the input it is in its first power. Even at this stage, however, it appears that the eventual inductive effects can be neglected. My last experiment was designed to explore how accurate the current measurement is when done indirectly (by dividing voltage across an active resistor by the resistance of this resistor). In other words, whatever voltage change might occur when adding such resistor is unimportant. That ultimately is taken care of when measuring the overall voltage in the actual experiment. We're not doubting the voltage measurements. What is being doubted are the values of the current which are derived quantities. When looking at the voltages from which we derive the current in my last experiment, it is seen that these voltage undergo a very slight glitch at the very beginning even when the spice of the applied voltage (not the voltage used to assess the current) is quite big -- the response of the current to a high voltage spike is negligible. Thus, I think that even prior to measuring it with a current probe (which bring an end to this major doubt we're having as to the reality of the OU effect) it appears that current data for frequencies as high as 100kHz and even higher are trustworthy. The contribution of the glitch in the current you see in the 50kHz pic I posted is negligible which can be seen when you integrate the current over time for a, say 100Hz frequency and compare its slope with the integrated current at 50kHz frequency. What I haven't done is to see what the square and the third power of this frequency will give, that is, is the glitch in question negligible at these conditions. Whill have to see that an will post results.

As for calorimetry, I don't think it's necessary once you do the electric measurements right. Even the slight variations of the resistance values don't exert so great an effect and I suspect the error due to the glitch in the current at high frequencies brings about an error of that magnitude which is to be neglected. Also, realize, with current probes the resistors we're using now to measure the current will be gone and we will be carrying out a power balance on a bare naked transformer whereby we'll only have to worry about the accuracy of the transformer input and output resistance. That's why current probes are so much needed for reaching a definitive answer.

[Omnibus]

Also, realize, conversely, once it's confirmed that active resistors can be used to measure the currents then it would be possible for more folks to replicate the experiment without the need to get into the expense of buying current probes. It would be nice if four channel software oscilloscopes can be downloaded free from the internet (I've only seen 2 channel software scopes so far, using the sound card and wonder if 4 channel softscope exists at all). If we can achieve reproduction of the OU effect discussed here (it appears the definitive answer is almost around the corner) by many participants here and in other forums that would bring the OU effort to a new level. Of course, the transformer I'm studying can be found probably only in the US but I have several of these and I can send it to anyone who might express interest and has the equipment to study it as an OU machine. Don't know if customs in Holland would charge a fee otherwise I can send one to @teslaalset as long as I know the address to send it to. If you're not in the US I can send it to you too. Just let me know.

[LarryC]

@Omnibus,

This open loop current transducer from digikey is $18.00 US. http://search.digikey.com/scripts/DkSearch/dksus.dll?lang=en&site=US&WT.z_homepage_link=hp_go_button&KeyWords=csla2cd&x=24&y=20

I have this setup. This item with a 9 volt battery in the circuit and connected to the voltage probe is all you need for accurate current testing.

Plus, I don't understand why a .1 ohm carbon resistor has an inductance problem. It is good enough for JLN and in most of my testing, but not for you?

Just curious, how do you have an expensive high end 4 probe tek scope (~15,000) and can't afford a much cheaper current probe (~500)?

Regards, Larry