field test unit no 001 "model T"

[Ferric Toes]

No Eric, it is noisy during the avalanche mode (off-time flyback period) modulating the target response, timings (time constant modulation) and so on...
This is one of the secrets of ML (& the QED). Snubber circuits help to control the max. voltage to avoid the avalanche mode.
Aziz

How would you observe this noise? Does not the very target act as an integrator?

Eric.

[Davor]

Instead of a snubber, which is partly resistive and wastes some energy, we could go full monty and use TEM exciter instead. It would not need a charging period and it would conserve energy. Such TEM pulse would be duration limited by it's capacitor to the largest possible with a desired shortest target tau and sample delay.

[Aziz]

How would you observe this noise? Does not the very target act as an integrator?

Eric.

Well, if you put a secondary RX coil in induction balance condition, one can measure the avalanche noise. Some mosfets do have a very bad avalanche noise behavior. Some better.

This noise isn't much but if you want to beat the evil best, its getting very important to avoid it.

Aziz

[Ferric Toes]

OK, but in a practical PIB detector I would be sampling soon after the start of the TXon period when the Mosfet is fully turned on. On resistance is usually 0.5 ohms or less, which from the noise point of view would be in parallel with the coil resistance (say 1 ohm). Noise should be miniscule compared to input resistor and preamp noise. What happens during the avalanche period does not matter as no sampling is taking place until after a further delay. You could have noise superimposed on the switchoff ramp but I can't see that this amounts to anything significant. I have used snubbing circuits but found that the diode used to charge a capacitor introduced low level effects, while Mosfets gave a much cleaner linear ramp type of switchoff.

Eric.

[Aziz]

OK, but in a practical PIB detector I would be sampling soon after the start of the TXon period when the Mosfet is fully turned on. On resistance is usually 0.5 ohms or less, which from the noise point of view would be in parallel with the coil resistance (say 1 ohm). Noise should be miniscule compared to input resistor and preamp noise. What happens during the avalanche period does not matter as no sampling is taking place until after a further delay. You could have noise superimposed on the switchoff ramp but I can't see that this amounts to anything significant. I have used snubbing circuits but found that the diode used to charge a capacitor introduced low level effects, while Mosfets gave a much cleaner linear ramp type of switchoff.

Eric.

Hi Eric,

the avalanche noise can be particularly observed on high current pulses. I have a tuned pulse transmitter (we call it TEM TX here), where this can be observed in an induction balanced configuration easier. As long as the flyback voltage isn't above the avalance voltage, the mosfet operates quiet. But if the mosfet avalances at higher voltage levels, a horrible noise can be clearly observed.

BTW, the avalanche mode heats the mosfet and it won't be very much amused by that too. Maybe the excessive heat is contributing to the noise much.
Aziz

[Ferric Toes]

Methinks your conditions are not quite the same if your coil is tuned. You would see a burst of noise if the peak exceeds the avalanche voltage. A paper on Google dealing with Noise in Semiconductors says that low noise performance returns once you are out of the avalanche region.

Eric.

[Aziz]

Methinks your conditions are not quite the same if your coil is tuned. You would see a burst of noise if the peak exceeds the avalanche voltage. A paper on Google dealing with Noise in Semiconductors says that low noise performance returns once you are out of the avalanche region.

Eric.

Well, that's a good reason to compare the two states:
- with avalanche mode
- without avalanche mode
and the effects to the damping time variation and response.

Who is going to make the interesting comparison?
Ready to get ?
Aziz

[Ferric Toes]

Well, that's a good reason to compare the two states:
- with avalanche mode
- without avalanche mode
and the effects to the damping time variation and response.

Who is going to make the interesting comparison?
Ready to get ?
Aziz

Not me. I am satisfied that avalanche noise is not significant when using the delay + sample system. Of greater priority is better match between coil and preamp to reduce noise there.

Three years ago I designed a high power PI for underwater pipline detection. Peak pulse current was 20A in a 500uH coil. Power Mosfet on fan cooled heatsink. Temperature sensor on heatsink kicked fan on at 70C. Coil ran at 70C in air. Mosfet avalanche volts 250, on resistance 0.03 ohms. In the workshop I used a 20in square dual stacked differential coil to cancel external noise. No visible increase in noise at preamp o/p whether TX was on or off.

One feature of diode/capacitor snubber that might make a small difference is that the rate of field switch off speeds up as it approaches zero. 1/4 cycle of the LC frequency. This might result in a better signal from small objects where the switch off with an avalanche circuit is less than 5X object TC.

Eric.

[Davor]

The same accelerating effect happens with TEM exciter. Point is that a capacitor gets reverse polarised, and in fact a reverse voltage accelerates the transition. The main difference between a snubber and a bare capacitor is that a circuit with a snubber has a Q well below 0.5

Perhaps a simple snubber could be a solution for accelerating the simple rigs such as Surf or Baracuda?

[Tinkerer]

Well, that's a good reason to compare the two states:
- with avalanche mode
- without avalanche mode
and the effects to the damping time variation and response.

Who is going to make the interesting comparison?
Ready to get ?
Aziz

The problem with the switching noise, is the coil, cable, shielding capacitance. This spreads or integrates the noise spikes so that the resulting effect is prolonged or persists some time. This is clearly visible when using an IB coil arrangement that allows observation of the RX waveform during the whole PI cycle.

[Tinkerer]

Instead of a snubber, which is partly resistive and wastes some energy, we could go full monty and use TEM exciter instead. It would not need a charging period and it would conserve energy. Such TEM pulse would be duration limited by it's capacitor to the largest possible with a desired shortest target tau and sample delay.

Have a look at this simulation.
The green trace is the coil current.
The blue trace is the coin voltage.
The red trace is the RX coil current with IB configuration.

Adding more capacitance to C1, reduces the coil voltage and increases the Flyback time. You need the increase the OFF time.

The full cycle is 100us, for a 10,000 PPS

Try the target response with different target TC's

I add the LTSpice sim in zip

[Ferric Toes]

It would be interesting to see a scope shot.

Eric.

[Tinkerer]

It would be interesting to see a scope shot.

Eric.

I have some old scope shots, not exactly the same Flyback voltage and time, but if you find it interesting I can fire up the circuit and take pictures.

Attached picture of coil current wave form, measured across 0.1Ohm resistor in series with coil.

Flyback wave form

Input filter and preamp.

Output of preamp, with no target but ground, full cycle.

The coil is a 45cm diameter TX about 300uH and 22.5cm diameter RX + Bucking coil. RX is center tapped about 300uH

[crane]

The proof is redundant as the the assumptions for sampling an exponential decay curve are wrong.

Any chance you could explain what you mean here? I.E. Do you mean a metal target's exponential decay curve?

There seems to be some aggressive moves in the industry to claim any form of sampling ground balance as proprietary.

Are you suggesting there is another way to achieve PI ground balance without "sampling"?

[Ferric Toes]

I have some old scope shots, not exactly the same Flyback voltage and time, but if you find it interesting I can fire up the circuit and take pictures.

Attached picture of coil current wave form, measured across 0.1Ohm resistor in series with coil.

Flyback wave form

Input filter and preamp.

Output of preamp, with no target but ground, full cycle.

The coil is a 45cm diameter TX about 300uH and 22.5cm diameter RX + Bucking coil. RX is center tapped about 300uH

I have never succeeded in making a balanced coil system for a PI that didn't show a transient during the flyback time. Your flyback appears to be without any avalanche phenomena, so it is a cleaner transient and therefore your resulting RX trace is different to what I have observed. However, I do not get the long tail which stretches out to mid screen in your RX scope picture. Could this be due to the RC time constants that you have pre the clipping diodes - are they matched?

I have tried DD arrangement and coaxial stacked coils on a PI, and the coaxial stacked gave the best balance as the geometry has the best symmetry. However, each coil has to be individually shielded, taken via separate cables and carefully grounded at the electronics end. Shielding and where it is grounded is very critical and I dont know if this can be replicated on a simulator. Most commercial DD coils seem to have minimal overall shielding for both coils and is no good at minimising the transient. Having three cables is a pain and I suspect that it is possible to get it down to two by having screened twin and a good differential front end.

I would expect to see a transient during the switch off period but for the RX to flatline as least as quick as that obtainable with a mono coil.

By the way another thing to look out for is eddy currents generated in the coil connector, if it is metal bodied and has parallel pins. This gives a long tail as in your waveform due to the TX pins acting like a small one turn loop close to the metal body. I got rid of it by using separate connectors for RX and TX. For a mono coil a coaxial connector is best.

Eric.