Amplifier frequency response

[Old cart]

Calculates 2.25 Hz(.1 and .033 in parallel). The thought has been to use the capacitor coupled output to find the target and look at the last direct coupled stage with a voltmeter to pin point. Offset shouldn't matter non motion, just looking for peak change.

That should work too since you are in non-motion.

[green]

The target is swinging across one of the coils in the scope pictures I have been posting. Including a picture with the US nickel swinging across both coils with the swing center where the two coils touch.

[Old cart]

The target is swinging across one of the coils in the scope pictures I have been posting. Including a picture with the US nickel swinging across both coils with the swing center where the two coils touch.

Those are very clean traces. The low pass might be a little to low though. You might try a higher cutoff frequency like about 8 Hz. Actually doesn't the MAX make it easy to adjust cutoff frequency? If so you could make it adjustable from say 2- 10 Hz so it could be optimized for the situation.

[green]

Those are very clean traces. The low pass might be a little to low though. You might try a higher cutoff frequency like about 8 Hz. Actually doesn't the MAX make it easy to adjust cutoff frequency? If so you could make it adjustable from say 2- 10 Hz so it could be optimized for the situation.

I played with the frequency some. The absolute valued trace looks like two half sine waves, if I increase the filter cut off to much I start to see the opening especially at lower sweep speeds. Not sure the absolute value filtered is the way to go yet. The absolute value goes - instead of + because I used a LM358 for the circuit and the LM358 input includes the - rail. You can use an external oscillator to adjust cutoff with the MAX7410, or vary the external capacitance.

[Old cart]

I played with the frequency some. The absolute valued trace looks like two half sine waves, if I increase the filter cut off to much I start to see the opening especially at lower sweep speeds. Not sure the absolute value filtered is the way to go yet. The absolute value goes - instead of + because I used a LM358 for the circuit and the LM358 input includes the - rail. You can use an external oscillator to adjust cutoff with the MAX7410, or vary the external capacitance.

I can see that you would see the gap in the absolute value trace because lowering the bandwidth increases the pulse width. The trade off here is low bandwidths will make the signal very quiet but also very small if you are sweeping too fast for the filter to respond. I once had a White's VLF that would go very deep but required slow speed to get get that depth. It used a 4th order filter.

That is largely why I suggested a variable frequency low pass filter. It would be very cool to automatically change the cutoff frequency with changes in sweep speed. Faster sweep speed = higher cutoff frequency. This would require either a micro and code or some kind of elaborate analog circuit and an accelerometer.

[green]

I can see that you would see the gap in the absolute value trace because lowering the bandwidth increases the pulse width. The trade off here is low bandwidths will make the signal very quiet but also very small if you are sweeping too fast for the filter to respond. I once had a White's VLF that would go very deep but required slow speed to get get that depth. It used a 4th order filter.

That is largely why I suggested a variable frequency low pass filter. It would be very cool to automatically change the cutoff frequency with changes in sweep speed. Faster sweep speed = higher cutoff frequency. This would require either a micro and code or some kind of elaborate analog circuit and an accelerometer.

A variable frequency filter might be a good idea. I know zero about micro's so I did some scope traces at different sweep speeds to see how bad it is. Probably don't have the best high and low pass filter cutoff frequency's for the integrator and post amplifier. The high frequency glitches are there even if I ground the scope probe. The disk is glued to the swinging bottle not the coil, target glued to disk.

[Old cart]

A variable frequency filter might be a good idea. I know zero about micro's so I did some scope traces at different sweep speeds to see how bad it is. Probably don't have the best high and low pass filter cutoff frequency's for the integrator and post amplifier. The high frequency glitches are there even if I ground the scope probe. The disk is glued to the swinging bottle not the coil, target glued to disk.

good work. Looks like I would expect. Do the HF spikes go away when you turn off the detector? When you say ground the scope probe how are you doing that? If you want to see where they are coming from you can make a tiny , say 1" coil of wire and hook it btwn Gnd and signal at the end of the scope probe. Then wave the coil over your circuit. This makes a pretty sensitive magnetic field probe.

if you use VCO audio with your circuit then you may gain some info about the target if adjust the frequency of oscillator so that it can go both up and down in frequency as the target passes. This when combine with a variable frequency filter and changes in sweep speed may help with rough target ID.

[green]

good work. Looks like I would expect. Do the HF spikes go away when you turn off the detector? When you say ground the scope probe how are you doing that? If you want to see where they are coming from you can make a tiny , say 1" coil of wire and hook it btwn Gnd and signal at the end of the scope probe. Then wave the coil over your circuit. This makes a pretty sensitive magnetic field probe.

if you use VCO audio with your circuit then you may gain some info about the target if adjust the frequency of oscillator so that it can go both up and down in frequency as the target passes. This when combine with a variable frequency filter and changes in sweep speed may help with rough target ID.

The spikes happened when the control one shots switched. The control and analog circuits are on separate boards. Each has it's own 5 volt regulator for the power. The analog supply has a TLE2426 for a virtual ground. The integrator switch is mounted on the analog board and used the analog supply for power. After rewiring the integrator switch supply to the control supply the glitches seem to have disappeared except for when the coil turns off. Maybe due to avalanche on the coil driver. The scope probe ground lead is always connected to the virtual ground, when I say ground the scope probe I connect the probe to the virtual ground.