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

[teslaalset]

@teslaalset,

Now, if we suppose that the data in posting #3123 were correct then it would definitely be an even more interesting case of OU than the reansformer case. Indeed, in #3123 we don't even need to know what's going on in the secondary coil -- the negative value of the power is enough to conclude that we have an OU device based solely on the data for the primary coil. The negative sign of the power values indicates that energy has been returned to the device, in addition to the energy that has been dissipated through Ohmic losses and losses in the core (hysteresis, eddy currents, magnetostriction, etc.) The novelty here is that during the functioning of the device one not only gets energy produced in the form of these losses but gets back electrical energy. This is a self-sustaining device par excellence. Therefore, it is crucial to ascertain that we're not dealing here with simple measurement errors.

@omnibus, when do you spend time to sleep? I did'nt expect such fast response  .

To further align our views, I took the diagram of posting #3117 and the excel data of #3123.

In column D, I don't see the absolute values, which should be used according to my indications.
In my view you should use ABS(((B(x)/1.0762)*C(x)) to calculate D(x) of column D.
Reason: it doesn't matter whether the current through the input impedance is positive or negative, it consumes power in both cases.

[Omnibus]

@teslaalset,

Reason: it doesn't matter whether the current through the input impedance is positive or negative, it consumes power in both cases.

I agree with that but only when it concerns dissipated power. Note, however, that current is squared when Ohmic losses are calculated so we don't need to take its absolute value.  When, however, electrical energy applied is concerned we have to take the signs of I and V as what they actually are. In the usual cases the signs are such that the energy during a full cycle is positive, that is, energy has been spent. Recall, however, the case when we have induction and we are to assess what energy has been spent during a full cycle to overcome it. As is well known, the net energy spent during a full cycle to overcome induction is zero -- whatever energy has been spent in the first part of the cycle is returned to the source when the cycle is completed. Thus, the net energy of induction is zero if we look at the entire cycle. Notice, we will not reach this (correct) result if we had used the formula ABS(((B(x)/1.0762)*C(x)) you proposed. Hope the above illustrates why one has to deal with the real data registered by the scope and not with their absolute values.

[LarryC]

@Omnibus,

Interesting statement from wikipedia: 'Current Shunts are considered more accurate and cheaper than Hall effect devices.' at
http://en.wikipedia.org/wiki/Shunt_(electrical)

Regards, Larry

[Omnibus]

@Omnibus,

Interesting statement from wikipedia: 'Current Shunts are considered more accurate and cheaper than Hall effect devices.' at
http://en.wikipedia.org/wiki/Shunt_(electrical)

Regards, Larry

@LarryC, thanks for the quote. Yesterday's experiments with the P6021 probe reassured me further that the method I'm applying for measuring current is correct. Unfortunately, as I said before, because the P6021 probe is a transformer type probe it is inferior to the method I'm using -- it cannot measure the DC component of the current (the offset) and burdens the circuit with additional inductance which is exactly what we want to avoid. Nevertheless, I saw that the form of the secondary coil current trace taken with that probe is the same as the form obtained by the method I'm applying. The primary coil current form appeared slightly different (without offset, of course) with the current probe. Unfortunately, now looking more thoroughly into the Agilent probes I've ordered I'm finding out they won't do the job in terms of resolution -- 100mA minimum isn't enough. What I really need are two TCP0030 probes (Hall effect based), over $3,000 each. That may be an overkill.

[LarryC]

@LarryC, thanks for the quote. Yesterday's experiments with the P6021 probe reassured me further that the method I'm applying for measuring current is correct. Unfortunately, as I said before, because the P6021 probe is a transformer type probe it is inferior to the method I'm using -- it cannot measure the DC component of the current (the offset) and burdens the circuit with additional inductance which is exactly what we want to avoid. Nevertheless, I saw that the form of the secondary coil current trace taken with that probe is the same as the form obtained by the method I'm applying. The primary coil current form appeared slightly different (without offset, of course) with the current probe. Unfortunately, now looking more thoroughly into the Agilent probes I've ordered I'm finding out they won't do the job in terms of resolution -- 100mA minimum isn't enough. What I really need are two TCP0030 probes (Hall effect based), over $3,000 each. That may be an overkill.

Just my opinion, but I'm not sure why you wish to continue with the pursuit of a hall effect current probe. It is clear to me from the available information, that a shunt resistor would be more accurate, less costly and it will measure the DC and AC component with your voltage probes.

With my scope a .1 Ohm resistor is the lowest that I can use for a small amperage circuit, any lower and the gain needed would cause too much static. With your high end scope a .001 Ohm could be used. This is important as it lowers the power wasted by the resistor.

Regards, Larry