Hysteresis curve's Y axis seems to be primarily measuring the DC dual power supply of the opamp integrator

Question

I am trying to find the hysteresis curve of a transformer's iron core.

The curve looks good, but I do have a doubt:

The LM741 op-amp integrates the output voltage signal from the transformer's secondary winding. The integrated output voltage Vout is connected with channel 2 on the oscilloscope and is reflected on the y axis. It is proportional to the magnetic field density B, and H is on the x-axis. From there I can obtain a B-H curve.

The op-amp needs a 9V dual power supply, which I provided.

What I observed is that the y-axis, which is supposed to reflect the measurement of Vout, actually also measures the DC supply voltage.

As can be seen here, the "height" of the curve increases as I turn up the DC supply. I am a bit confused and curious about this:

Isn't the y-axis only supposed to measure Vout? Then why does the value on the y-axis increase when DC supply into the opamp is increased?

I observed this when I disconnected the input voltage from channel 1, meaning there was no input signal (the signal in red is input.) I even removed the secondary winding (I am using a dissectible transformer) to check if maybe the output signal comes from the residual magnetic density within the iron core. It turns out that it is the DC supply that interferes with the signal.

As can be seen, despite there being no input signal and no secondary winding, there was still output signal from channel 2.

This signal increased as I turned the knob and increased the DC voltage. The V RMS is weaker compared to when there is a secondary winding and an input signal, but it does vary with the DC power supply. (The output signal is yellow.)

Why is this happening?

Answer 1

When you say that the 741's output amplitude increases as you increase its power supply, immediately I think that it's saturating.

In other words, the input signal is too large, and the output is getting stuck against the positive and negative supply limits.

It looks to me like you have a gain of -(1MΩ / 10kΩ) = -100. With power supplies of ±9V, you can't expect the 741 to produce an output greater than ±7V out, which corresponds to a maximum input amplitude of 7V / 100 = 70mV. If your input exceeds 70mV in either direction, the output signal will be clipped.

If this is what's happening, you have two options:

1. Reduce the gain, by reducing the 1MΩ resistor or increasing the 10kΩ resistor.

2. Reduce the input signal amplitude.

Note: a clipped hysteresis curve would look like a normal one, just shorter vertically, and I suspect this is what you are seeing here.

Note 2: the gain bandwidth product of the 741 is 1.5Mhz if you're lucky. That corresponds to 15kHz at a gain of 100. If your signals are approaching that limit, you can expect severe attenuation.