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

[Omnibus]

@LarryC,

To accept that current measured by dividing the resistance of a shunt resistor into the voltage measured across that resistor is the actual value of the current means that we are talking here about a major finding. I don't see anything else that can come close as an objection to the reality of the effect compared to what a critic might say about the current measurement. I discussed that part, the current measurement, with several colleagues and their opinion summarized coincides with yours. Would be interesting to see their reaction when they hear what this really is all about. Especially in these new times when current measurements are almost completely forgotten -- because of the microelectronics is in the focus most measurements in these electrical engineering departments are voltage measurements. Now I have to figure out how to present this new finding. I'll continue the contacts with my academic colleagues and will probably ask some of them to assign the measurement of the power balance of a transformer to a grad student as a small educational project or, even better, have them do the balance. Too bad no one here in this forum found this simple project to be worthy of experimental attention. Of course, lack of adequate equipment may be the main reason. Anyway, next step is assesment of the propagation of errors and summarizing the work done so far. Any ideas about the error analysis?

[Omnibus]

Here's the schematic of a setup which appears to solve the inductance problems. Now the OU effect is easily reproducible at very low frequencies such as 1Hz and is much more pronounced (many times) than previously. I haven't tried it with other transformers but I guess OU will show trivially in all of them. It seems the incomplete energy balance is the reason OU has not been discovered so far in the various energy transforming devices, including transformers. Would be nice to hear comments about the new setup and maybe even have someone try it.

[gyulasun]

@Omnibus

I wonder the physical distance between the connecton point of CH4 and that of CH3, if it is 2 -3 centimeter, then there can be no extra info in the voltage shape between the two points?  Especially in the low frequency range the transformer is able to cover, such a small distance does not introduce transmission line effects.
I can ask the same for CH2 too: why is it justified to measure the same voltage again on a separate channel?  Maybe you wish to mount the BNC sockets further from each other? (Even so it may warrant this at some hundred MHz.)

One more question: I tried to figure out from its user manual whether your DSO scope 4 input channels have got independent grounds each or just common ground as most of the multi-input scopes have but I have not found it. (I think they do not have independent grounds from each other, once you created the measurement setup with all inputs having the common ground.)

Otherwise the setup seems ok for me.

rgds,  Gyula

[Omnibus]

@gyulasun,

You're quite right about the redundant measurement of voltage at CH1 and CH2 as well as CH3 and CH4 (the distance between CH1 and CH2 is about 5cm while between CH3 and CH4 it is 8cm). I did that for consistency and to use it for some other configurations, if needed. But, yes, you can get away with only 2 voltage measurements rather than double them. Just to mention -- one has to be very careful when using the 6252point (REDUCED) mode vs. 125,000point (FULL mode) when dumping the data into the flash drive. For some reason Tektronix have decided that the 6252point dump has to be into a 9 column spreadsheet adding four additional columns for the peak voltage values at each channel. The 125,000point dump is supposed to be only 5 column dump but at some x-axis resolutions it becomes 9-column without a warning. This may result, as obvious, in strangest results if you apply a 5-column macro. 1,000,000point measurement always yields 5 columns but it is impractical because, the results being essentially the same, it reqiures much longer time to crunch the data let alone plot it. As for the ground, my understanding is that it is a common ground for all four channels. What is interesting in the last setup I showed is that because of the large area of the ground plate the phase difference between the current and voltage under no load is completely gone -- practically no inductance. That's exactly what we need. That improved the results immensely, like night and day. As I said before, now you can observe OU reproducibly at frequencies as low as 1Hz and maybe even lower. 

[Omnibus]

@gyulasun,

Your remark is also important in a sense that people who do not have four-channel oscilloscopes and probably do not have any standalone oscilloscopes but only softscopes downloaded free from the internet may try to replicate these results.