Persistent bug in PI circuit design.

[Teleno]

There are a lot of things being tossed into the mix that have nothing to do with the OP.

First, let's talk about the resistor-diode clamp. No, it doesn't play a big role in sampling time; a 1N4148 has a depletion C of about 5pF, so 2 diodes are 10pF; with a 1k resistor this is a 10ns time constant.

The diode's recovery time is 10 x that when the input signal goes into the millivolts.

For example, a 500V pulse takes 130ns to reach 0.5V at the diode and 200ns to reach 0.1mV.



Another problem is that most of the signal's power is wasted in the damping resistor as a current. Instead of using a high-impedance amplifier on the signal's voltage, the signal's own current can do the job faster if properly configured.

In the example below the traditional approach (bottom graph) produces 1.7mV while the signal's own current, when transferrred to a 4K resistor, produces 17mV 250ns earlier (top graph) :

[Teleno]

Second, damping is another separate issue. Yes, the R used in the clamp reduces the damping R during clamp but you can account for it in the design.

I think you mean increasing the damping resistor to account for the presence of R in parallel. But then when the signal goes below 0.5V the diode is blocked and the circuit goes underdamped (oscillates) .

Yes, you can switch in the damping R during decay but it's really squirrely. I did this many years ago, it worked great in Spice but not so well in practice. Too much decay variation over ground.

Inductance changes over ground, which changes the transient curve. A fixed delay will not work, but a voltage dependent delay can fix this. It's easy to implement using the Id/Vds curve of a MOSFET. For a given Vgs, when Vds > Vp (saturation) the resistance is infinity, for Vds < Vp (ohmic region) the resistance is a fixed value. Vp is low (about 0.2-5V) and the transition is abrupt.

Wth this technque, a variation of 10% in the inductance causes a mere 60ns difference in the delay.

[ivconic]

Made a small o0o coil for charting short time constant targets. Two Rx coils side by side(wired inverted to induction balance) with a Tx coil wound around them. I can adjust the coils for zero out during coil on but not during coil discharge. Should I be able to get zero out during discharge? So far using a diff amp and induction balance seems to work best for me. No diodes or series resistors. Back to back diodes across the coil don't seem to work, probably because I can't get zero volts during discharge.

Are those two RX coil identical? Assume they are.
Would be interesting to experiment with two different RX coils in such setup?

[green]

Are those two RX coil identical? Assume they are.
Would be interesting to experiment with two different RX coils in such setup?

I tried to make them the same. The inductance of the Rx coils were close. I have a larger coil made with two 200mm dia. Rx coils side by side with a Tx coil around them. A 200mm DD coil. Used two 200mm coils to make a OO coil overlapped for induction balance. All can be adjusted for amplifier zero out during coil on but not during discharge.

[Teleno]

All can be adjusted for amplifier zero out during coil on but not during discharge.

Any target away fron the cenbterpoint will slightly alter the flux balance, which is most noticeable during discharge.

[Chet]

Made a small o0o coil for charting short time constant targets. Two Rx coils side by side(wired inverted to induction balance) with a Tx coil wound around them. I can adjust the coils for zero out during coil on but not during coil discharge. Should I be able to get zero out during discharge? So far using a diff amp and induction balance seems to work best for me. No diodes or series resistors. Back to back diodes across the coil don't seem to work, probably because I can't get zero volts during discharge.

Hi Green


In your coil arrangement both receiver coils are tightly coupled to the transmit coil. Both coils are receiving high voltage signal levels during the coil discharge/flyback time. So the smallest imbalance will produce a very large undesirable signal level to the input of the op-amp during the discharge time/flyback pulse. I think both Rx coils need to see the same small amount of transmit signal. To do that most effectively I think they need to be individually zero balanced with the transmitter coil. Otherwise there will be considerably more transmit/flyback signal which will unbalance the target signal response between the receiver coils.


I think the oOo coil arrangement that MikeBG proposed would be a better choice. Or a DOD coil arrangement. By careful induction-balance nulling each receive coil across opposite sides of the transmit coil it reduces the signal levels at the input to the op-amp to very low levels.


I think there may be the following possible advantages with the oOo coil arrangement:


Shorter sample time delay is possible since there are no high signal levels to be handled by the receiver op-amp.
EMI noise is reduced since it is received equally in both coils and cancelled by opposing polarities.
Ground balance and variations in ground may be smoothed out somewhat.
Noise from the coil bumping into rocks, shrubs or uneven ground may be reduced.
Some signal gain is possible by having more turns in each of the receiver coils.
It may eliminate the earth field effect.


In the attached patent address the DOD receiver coils are in series instead of opposing polarities as in MikeBG’s proposed arrangement. It would be interesting to compare the two forms. The MikeBG method might require more turns in the receiver coils.


Have a good day,
Chet


https://www.google.com/patents/US20130154649

[ivconic]

[green]

Hi Green


In your coil arrangement both receiver coils are tightly coupled to the transmit coil. Both coils are receiving high voltage signal levels during the coil discharge/flyback time. So the smallest imbalance will produce a very large undesirable signal level to the input of the op-amp during the discharge time/flyback pulse. I think both Rx coils need to see the same small amount of transmit signal. To do that most effectively I think they need to be individually zero balanced with the transmitter coil. Otherwise there will be considerably more transmit/flyback signal which will unbalance the target signal response between the receiver coils.


I think the oOo coil arrangement that MikeBG proposed would be a better choice. Or a DOD coil arrangement. By careful induction-balance nulling each receive coil across opposite sides of the transmit coil it reduces the signal levels at the input to the op-amp to very low levels.


I think there may be the following possible advantages with the oOo coil arrangement:


Shorter sample time delay is possible since there are no high signal levels to be handled by the receiver op-amp.
EMI noise is reduced since it is received equally in both coils and cancelled by opposing polarities.
Ground balance and variations in ground may be smoothed out somewhat.
Noise from the coil bumping into rocks, shrubs or uneven ground may be reduced.
Some signal gain is possible by having more turns in each of the receiver coils.
It may eliminate the earth field effect.


In the attached patent address the DOD receiver coils are in series instead of opposing polarities as in MikeBG’s proposed arrangement. It would be interesting to compare the two forms. The MikeBG method might require more turns in the receiver coils.


Have a good day,
Chet


https://www.google.com/patents/US20130154649

Hi Chethttp://www.geotech1.com/forums/showt...528#post205528
Thanks for the reply. Reminded me of some things I've done in the past. Reply #63 in above thread has a comparison of DOD coil Rx adding and subtracting. The coil looks similar to ML's coil, don't know if it acts the same. I keep putting off making a detector thinking I might learn something better. Probably time to try one and see what needs improving. I still need to try some more with a mono coil and a switch to disconnect during discharge. The o0o coil looks like it has enough advantages to give it a try while playing with the mono coil. The DD chart at 12 inches looks like I didn't take enough care taking the data.

[Chet]

Hi Green

I had forgot about your earlier tests. They say that memory is the first thing to go. I think?

Thank you,
Chet