HV temporal sampling ?

[Riss]

https://www.youtube.com/watch?v=JlKkECM6sWE - !?? I can't believe these two top models are so noisy in real search. This woman probably doesn't know how to adjust the ground balance properly. They "hear" all the hot rocks!?? In Europe, such behavior of detectors is when they are tuned to the ground (a general background that has weak fluctuations) and hot rocks are heard against this background.

[moodz]

Hi Paul,

I've done a few experiments and there really is a distinct difference between ferro targets and conductive targets.
I determined what was the optimal ratio of the two pulses and the corresponding delays of the samples (as far as I could, of course), but...

But there is no goal, again a beam.

Samples from ferro targets are very close to the results of low-conductive targets (this could be assumed because in collapse time they are have a similar nature).
Later, maybe next month, when I really have enough free time, I will change a little of the detector code to look for a combination between these samples and samples after the long pulse (I think four additional samples will suffice).
If I manage to achieve some satisfactory results, I will post them on the forum.

Hey Paul, thank you for the help you provide and for the extremely useful things you share in the forum.
Thank you again.

Regards,
Vangel

Thanks Vangel ... also see my transmit circuit in the FPGA thread.
I use a version of that to produce the magnetic permeability ( FERRITE ) channel and the conductance channel ( R ) and there is a third channel ... the phase channel. ( LEAD/LAG ).

[boilcoil]

OK, thanks,
of course, I will read it very carefully.

[boilcoil]

Hello again,I did many experiments, but unfortunately without success.
In monocoil mode, when operating on just one channel and with 5 passive discriminating samples I was not able to achieve separation of ferrous/non-ferrous targets.The low-conductivity targets almost completely repeat all the samples of the ferro targets.
It is very easy to separate targets into shortest, short, medium, long and longest time constants, but not ferrous/non-ferrous.
The only active samples for ferrous/non-ferrous are the samples measuring change of coil's current ramp, but this is so insensitive that it only works for large pieces of iron.

Of course, in the other modes ( work with IB coil), separating the ferro targets is very easy.
In both IB modes:
- full discrimination, where the discrimination based on time constant and ground balance is carried out in the channel receiving information from the transmitting coil (as in mono-coil), and separation of ferro/non-ferro is done in the IB coil channel.
- simple discrimination, where both channels receive the signal from the IB coil and discrimination is only ferrous/non-ferrous, but the depth is greater due to the greater inductance of the receiving coil.

It is obvious that for the mono-coil mode, which is the most interesting for PI MD, a new way needs to be found.
Not a new method, but a new principle.
It should be continued with the thinking in this direction.

[moodz]

Hello again,I did many experiments, but unfortunately without success.
In monocoil mode, when operating on just one channel and with 5 passive discriminating samples I was not able to achieve separation of ferrous/non-ferrous targets.The low-conductivity targets almost completely repeat all the samples of the ferro targets.
It is very easy to separate targets into shortest, short, medium, long and longest time constants, but not ferrous/non-ferrous.
The only active samples for ferrous/non-ferrous are the samples measuring change of coil's current ramp, but this is so insensitive that it only works for large pieces of iron.

Of course, in the other modes ( work with IB coil), separating the ferro targets is very easy.
In both IB modes:
- full discrimination, where the discrimination based on time constant and ground balance is carried out in the channel receiving information from the transmitting coil (as in mono-coil), and separation of ferro/non-ferro is done in the IB coil channel.
- simple discrimination, where both channels receive the signal from the IB coil and discrimination is only ferrous/non-ferrous, but the depth is greater due to the greater inductance of the receiving coil.

It is obvious that for the mono-coil mode, which is the most interesting for PI MD, a new way needs to be found.
Not a new method, but a new principle.
It should be continued with the thinking in this direction.

Just for interest ... can you tell me why the monocoil PI is more important to you ... Is it only because the coil is easier to make ?

[boilcoil]

Frankly speaking, I don't know either, but I feel closer to it.

[moodz]

yeh ... its my favourite type of coil :-)

[boilcoil]

I just had an idea - probably a foolish one, but still I want to share it:

1. In order to have a very steep falling characteristic after stopping the impulse, we need to ensure maximum energy loss during damping. The power loss is equal to I^2*Rd, where the current drops nonlineary, but unfortunately we are limited from increasing the value of the damping resistor due to the increase in voltage above the avalanche breakdown voltage of the FET and sufficiently high Q-factor of the coil (ringing). So we can only reach the critical value of Rd. But what would happen if we added loss through an additional resistor powered by a current transformer connected to the transmission circuit? The main losses in it will occur in the first 3-4 microseconds (when di/dt is at a high value).

2. If we sample from this short period with a suitable divider on this additional resistor, we might, maybe, detect remnants of the active component of the signal, as a DC will be filtered by the CT and will not be present in the samples.. This would help solve the ferrous/non-ferrous discrimination problem in monocoil mode.

Paul, please help.


EDIT: It might be more reasonable to investigate the interval t0-t1 up to t0+t2, where t0 is the moment the pulse is turned off, and t1 and t2 are intervals on the order of short microseconds,in which interval the current from the current transformer passes through zero point .

EDIT: My mistake - the current will not pass through the zero.

LAST EDIT: Actually, it will pass if the current transformer is connected to the damping circuit, but then there is no point in investigate the signal before the stop of pulse.

REALLY LAST EDIT: Forget about point 1 - the the decay time will not be reduced by an additional resistor due to the reverse action of the CT in the primary circuit, but early samples can be taken.



NO MORE EDITS, that was really a strange post.
BTW where is Aziz?

[moodz]

I just had an idea - probably a foolish one, but still I want to share it:

1. In order to have a very steep falling characteristic after stopping the impulse, we need to ensure maximum energy loss during damping. The power loss is equal to I^2*Rd, where the current drops nonlineary, but unfortunately we are limited from increasing the value of the damping resistor due to the increase in voltage above the avalanche breakdown voltage of the FET and sufficiently high Q-factor of the coil (ringing). So we can only reach the critical value of Rd. But what would happen if we added loss through an additional resistor powered by a current transformer connected to the transmission circuit? The main losses in it will occur in the first 3-4 microseconds (when di/dt is at a high value).

2. If we sample from this short period with a suitable divider on this additional resistor, we might, maybe, detect remnants of the active component of the signal, as a DC will be filtered by the CT and will not be present in the samples.. This would help solve the ferrous/non-ferrous discrimination problem in monocoil mode.

Paul, please help.


EDIT: It might be more reasonable to investigate the interval t0-t1 up to t0+t2, where t0 is the moment the pulse is turned off, and t1 and t2 are intervals on the order of short microseconds,in which interval the current from the current transformer passes through zero point .

EDIT: My mistake - the current will not pass through the zero.

LAST EDIT: Actually, it will pass if the current transformer is connected to the damping circuit, but then there is no point in investigate the signal before the stop of pulse.

REALLY LAST EDIT: Forget about point 1 - the the decay time will not be reduced by an additional resistor due to the reverse action of the CT in the primary circuit, but early samples can be taken.



NO MORE EDITS, that was really a strange post.
BTW where is Aziz?

Use a current sink as a non linear resistor and you will damp alot faster. ( eg mosfet and diode )

[boilcoil]

OK, thank you.

When I have more free time, I will try 'HV' samples using a small CT in the 0-5us after the pulse.
I might also try with a saturating current transformer, if the changes in the samples are in too wide a range.
I am curious to see what will come out.