Just to clarify, when it goes back to zero, its just a matter of taking the magnet away and replacing it back on the iron nails and the effect reappears and gradually diminishes over time again, its not a one off effect.

With three different people each doing this magnet-against-iron (or steel) experiment, getting an initial good result then seeing it disappear over time, I do wonder whether if there is some weird unexplained effect going on within the iron.  If a repeat experiment using fresh nails or iron wire gives the same results then it could be worthwhile taking ones that have been through this above their Curie point to see whether the initial good result comes back.  That would tell us something about the weird effect.

Itsu has sent me the DC amplifier (not received yet) so I will do my own experiments using strips of high purity iron sheet I have available.  I will also have a number of ring-core flux concentrators made, connect them in series and see what I get using both DC and AC input to their primaries.  And of course I will look at putting magnets against those ring-cores.

Smudge

 

CTs don't output voltage proportional to instantaneous current. Depending on the ballast resistor (which better be non-inductive), they output the time derivative of current and have serious phase and frequency response deficiencies (not only near DC). This depends on the CT's core and inter-turn capacitances of its windings.

For this reason I have not recommended using a CT for wideband phase-coherent current measurements.  CSR is so much more accurate...

Quote from: Peterae on 2026.09.23, 10:46:58i wonder if the residual AC pickup is stopping the effect.?

How about the temperature gradients equilibrating over time and stopping any thermoelectric effects.

These are junctions of dissimilar metals after all...

Unfortunately i can confirm Itsu's finding that after 30 minuets it was back to zero.

Edit
i wonder if the residual AC pickup is stopping the effect.?

I can confirm panyumings findings.

5 iron nails soldered together using silver plated copper wire.

Setup is


and the meter reads without magnet


Testing for AC without magnet


Testing for AC with magnet


Testing with Magnet


Picture with Magnets


Testing again Moved magnets slightly


Rechecking with magnets removed


I will do a long duration test next 

Hello, esteemed senior Smudge! 
I really did forget your previous post. 
I went back to take another look. I saw Mr. Itsu said that the balancing time should be more than 20 minutes. 
Because my multimeter automatically powers off after 7 minutes, obviously when I tested before, the balancing time never exceeded 7 minutes. 
I tested again. This time I often pressed the multimeter's hold button twice to prevent it from shutting down. 
The longer the balancing time, the smaller the voltage reading; after 20 minutes, the reading went from 5 µV to 0 µV. 
Actually, this multimeter only shows 0000 when the voltage is below 5 µV. 
I wore two winter gloves to handle the action of placing the wire on the magnet. 
Even the original initial 25 µV was gone.  

@ panyuming

If your experiment is repeatable here is something you can do to finesse the significance of the result.  Try adding resistors across the output gradually reducing in value until the output voltage halves.  You then have the situation where both the resistor and the device are consumimg DC power which you can calculate knowing the voltage and the resistance value, total power = 2*(V²/R).  A demonstration of a passive device delivering power is really significant.

I did this with a device that was not passive and the problem was getting the low value (milliohms) resistor.  I had available various reels of copper wire so I cut lengths that gave me different values.  The value is easily calculated or deduced from published wire data.

Smudge

there are a few tokens and hashes that I assume are the issue,, probably some kind of verification maybe??