improve the hysteresis loop

[siti]

Hi Henrik,

I use 50Hz directly from powerline. From your explaination, do you mean FFT is the option to solve the coarse loop? I'm sorry I didn't get your point on this -'Clean' higher order frequencys are due to your hysteresis (n* 50Hz), all other will make your trace move around-

What is "n" actually?

Thank you Henrik for your attention.

[Henrik_Volkers]

---

siti wrote:

Hi Henrik,

I use 50Hz directly from powerline. From your explaination, do you mean FFT is the option to solve the coarse loop?

No, it's just to gain more Information about your signal.

I'm sorry I didn't get your point on this -'Clean' higher order frequencys are due to your hysteresis (n* 50Hz), all other will make your trace move around-

What is "n" actually? n is the order of the harmonic distortion of your (pure?) 50Hz sinus stimulation due to the nonlinearity (hysteresis) of your system.Your response will have 50Hz AND n*50Hz (100Hz, 150Hz, ....)
 All other signals (noise, other catched signals) will make your lines thicker.

---

[siti]

Thank you very much Henrik for your explaination. I thought the problem is due to the unstable voltages measured at the primary and secondary side of the ring core. Do you have any idea how to make the hysteresis loop more smoother and static?

[Henrik_Volkers]

Have you tried to sample with a higher samplerate? Your pics show 1k S/s so the hysteresis loop is build from 20 points. Try 20k S/s or higher to smooth.

Your are interested in the relation of x to y, so if you cant provide a stable sin voltage, one thing you can try is to cut out single periodes and normalize them to the input. 

Another hint for calibration: We used a pure Nickel probe 😉