500 nano second sampling

[green]

1 meter coil is for deep targets, like 1 meter deep for small (relative) targets and 2 meter deep for large (relative) targets. The targets could be voids in the ground or different conductivity of the ground/soil itself.
This machine is to measure the characteristics of the soil, not metallic targets. However, in the lab I often use metallic targets of different TC's to test the system.

Thanks

[green]

These measurements were made with the probe 10x1

The scope shows red trace 1 volt/div. Is the gate changing 3.2 volts or 32 volts? Does the - supply have the same negative over shoot as the gate signal?

[Tinkerer]

The scope shows red trace 1 volt/div. Is the gate changing 3.2 volts or 32 volts? Does the - supply have the same negative over shoot as the gate signal?

I don't remember just what I did at the time of the picture, so I have to reconstruct what I see on the picture. I know that I always use the probe on the 1:10 setting. I also know that the pulse of 3.2V must be from the MCU, that runs at 3.3V. This pulse is fed into a mosfet driver running at 12V.
I guess I wanted to show how the oscillation migrated to the MCU PSU battery ground.
Since you are so kind to offer your help in solving the problem, I suggest we start over with a proper documented experiment, so that we have reliable data to base our work on. I have a circuit and coil on the bench, not exactly the same, but close enough to show the battery ground noise problem. We can use that one to try to remedy the problem.

[green]

I still fighting oscillation problems. Some caused by the power supplies, coil shielding, etc. I'm hoping we as a group can solve your problem and learn something while doing it.

[Tinkerer]

I still fighting oscillation problems. Some caused by the power supplies, coil shielding, etc. I'm hoping we as a group can solve your problem and learn something while doing it.

Noise is our greatest problem. We can remove noise with filtering and integrating, but the best way is to avoid the noise by design.
Attached are 2 pictures showing noise.

The "Noise" picture shows blue trace, at 20mV/div, shows the minimum noise with no power to the circuit, measured on the battery minus input.
The red trace shows the feed-through of the noise on the same battery minus input at 200mV/div, with the circuit running.
The yellow trace shows the effect of the noise on the output of the pre-amp.

The IN_OUT& noise picture
The red trace shows the OUTPUT of the pre-amp with a very short TC target after some 1MHz filtering.
The blue trace shows the OUTPUT of the pre-amp without target.
The yellow trace shows the noise at the RX INPUT.

I think the first step should be to reduce/eliminate the feed-through noise on the battery minus input.

[green]

Hi Tinkerer, Good example. I'm wondering if we can come up with a simple test circuit to drive the power supply, so we can compare results. Example circuit, 555 generating a 10khz square wave.

[sinclairuser]

Eric used to use a 2ohm resistor in the negative lead to decouple the battery from the TX

did he put it on the - coil lead or - battery lead?.
thanks

[Tinkerer]

Hi Tinkerer, Good example. I'm wondering if we can come up with a simple test circuit to drive the power supply, so we can compare results. Example circuit, 555 generating a 10khz square wave.

The way I understand it, it is a matter of the magnitude of the step applied to the PSU, as explained in the bypassing(6) pdf above.
aggravated by the use of LC decoupling networks. For example, consider the decouplingnetwork shown in Figure 7. When a 20 mA step is applied, the resulting ringing had a peak amplitude of over 500 mV and a time constant for the decay of over 40 us

The attached picture shows the TX current step, blue, then the current step on the Capacitor bank,red, then the noise on the battery input lead, yellow. This is using the battery directly, no PSU.

The current pulse is 6Amps, measured across a 70mOhm resistor.

We see a voltage step of about 90mV on the capacitor, red, then no more voltage drop on the lead, after the filter, yellow, but still the switching noise.

[green]

Hi Tinkerer, Do you think the high frequency noise is feed through from some thing resonating or the power supply resonating after a load step? Have you tried a ceramic capacitor across the supply? If it's the power supply I think the frequency should change. I'm guessing, maybe someone that knows could help.

[Tinkerer]

Hi Tinkerer, Do you think the high frequency noise is feed through from some thing resonating or the power supply resonating after a load step? Have you tried a ceramic capacitor across the supply? If it's the power supply I think the frequency should change. I'm guessing, maybe someone that knows could help.

Added some low ESR capacitors and changed the reverse voltage input protection diode to a Schottky. Remarkable change, will give more details tomorrow, getting too late here.

[Chet]

Hi Tinkerer

If the added capacitors and diode change aren't the final fix then the following may help.

From the scope shots it looks like it may be noise from the oscillator in the voltage converter (555 chip or 7660A chip). One way to identify this as the source would be to pull the 555 or 7660A and substitute a small 12v (8 AA cells) battery to directly drive the 5v regulator. The 12v battery is preferred over a bench supply as it might inject other AC noise or ground noise.


Two of the Figure 9 circuits, from the .pdf document in Post #3, would be my choice as a possible fix. One will feed the transmitter circuit and the other feeding the voltage converter circuit. Adding inductors in to power filters is very effective. It will take some experimenting with different values to get it right.

Have a good day,
Chet

[6666]

did he put it on the - coil lead or - battery lead?.
thanks

From memory the 2 ohm R was between the neg batt lead and the circuit, he made a drawing will try and find it.

[Tinkerer]

Hi Tinkerer

If the added capacitors and diode change aren't the final fix then the following may help.

From the scope shots it looks like it may be noise from the oscillator in the voltage converter (555 chip or 7660A chip). One way to identify this as the source would be to pull the 555 or 7660A and substitute a small 12v (8 AA cells) battery to directly drive the 5v regulator. The 12v battery is preferred over a bench supply as it might inject other AC noise or ground noise.


Two of the Figure 9 circuits, from the .pdf document in Post #3, would be my choice as a possible fix. One will feed the transmitter circuit and the other feeding the voltage converter circuit. Adding inductors in to power filters is very effective. It will take some experimenting with different values to get it right.

Have a good day,
Chet

Thank you for the feedback. This was just the first most obvious try to fix the first problem. The TX runs directly from a capacitor bank, fed by a 12Ah, 12V battery. As the noise feeds back to the negative battery terminal, it needs to be eliminated before anything else.

[Tinkerer]

Hi Tinkerer, Do you think the high frequency noise is feed through from some thing resonating or the power supply resonating after a load step? Have you tried a ceramic capacitor across the supply? If it's the power supply I think the frequency should change. I'm guessing, maybe someone that knows could help.

This noise shown, emanates from the TX board. It is something on the board itself that is resonating, because the board is driven directly from the battery. From the Battery negative terminal, the noise then migrates to the analog PSU and every other board. I have drawn up a TX PCB, but decided to define the specific source of the noise first, to find a remedy before etching and building the PCB board.

I had originally a 1.8 Ohm resistor on the battery minus input to the board, but removed it trying to find the source of the noise. But since it is a perforated board with a bad circuit layout, I will try to add the resistor again, re-routing the bat minus input, to see if it helps.

[Chet]

Hi Tinkerer

Could you post the power supply and transmitter circuit diagram? You may wish to protect it for commercial reasons. If so I understand.

Chet