Is this a major advance in PI sensitivity ??

[WM6]

I urge those with suitable test / measure equipment to build this circuit and confirm / repeat the results.

I have measurement equipment, but those parts (as THAT1510, TC4421, FQPIN60, SF18G) are totaly unknown on my components stock list. I have to see first where to purchase those parts - if possible.

What are adequate FQPIN60 replacement? From memory I have IRFs 740, 640, 840 in my lab stock.

[moodz]

What mean this limitations regarding to use your front end in connection with classical PI detectors?

Happy and successful New Year to you.

The aquisition window of a conventional sample and hold in classical design would be too wide and timing accuracy might be an issue also. So it would either swamp the signal or miss it timing wise.

I am sure this is not an impossible problem ... if the CRO can capture it ...so can we.

The timing can be generated from the signal itself ... this will provide the capture pulse and phase shift information ... a circuit begins to form in my head now ....

best wishes.

moodz.

[WM6]

The BAD news is that this effect is in a 500 ns window at the base of the flyback response at the amp output

]

What mean this limitations regarding to use your front end in connection with classical PI detectors?

Happy and successful New Year to you.

[moodz]

Forgot to mention ...

My bad ... forgot to mention that I changed the switching mosfet to an FQP1N60 device to limit current draw during experiments .... so I am getting these results with less peak currents in the TX coil than I normally get with the IRF device.

Regards

Moodz.

[moodz]

More experimentation has revealed the circuit configuration that gives high level signal changes for small targets in the coil field.

The signal is stable in time and amplitude and the configuration is a actually a simplified version of my original circuit which was "over protected" with diodes. The coil can be moved around etc and returned to same position so it is not some weird resonance effect etc.

There is now only a single clamping diode which has been beefed up to a SF18G ... see specs. The input balance resistors have been dropped to 500 ohms. A dollar coin placed in the centre of the coil generates a volt of wave form amplitude displacement in addition to an obvious phase / freq delta. The gain of the amp is 100.

I urge those with suitable test / measure equipment to build this circuit and confirm / repeat the results.

The GOOD news is that this is a significant signal to noise improvement for PI technology.

The BAD news is that this effect is in a 500 ns window at the base of the flyback response at the amp output so a dspic will probably not be able to extract this artifact and I will probably be looking at developing this further in my FPGA project with a fast ADC. Although you could probably do something with a peal hold circuit etc. Those with CROs should be able to verify on screen though.

Some very quick tests show that the 22cm 32 turn diff coil is much more sensitive .... it shows nearly 100 mv of signal at 1 meter for large targets like the metal side cover off my computer.

We need now for others to confirm the results, find the optimal circuit ( I only spent 10 minutes trying this ) and investigate if target descrimination info is also present.

happy new year from moodz.

The circuit .... simpler yet better. I have removed all the protective diodes except one and changed the balance resistors to 500 ohms.

The flyback pulse at ouput of amp. Note the stepped response now ....

Purple = NO target .... White = Australian $1 dollar coin placed in centre of coil. 1 volt of signal + phase shifts