minimum sample delay

[green]

And... not all ferrites are created equal!!

When I recorded the imbalance caused with the ferrite core reply#47 I recorded imbalance caused by offsetting the Rx coils for the same imbalance during Tx on with constant rate(about 3600A/sec). The signal for the two methods during Tx off looked the same. Thought this might mean something, not sure what.

[KingJL]

... The ferrite bead could be giving a signal because of magnetic viscosity...

And... not all ferrites are created equal!!

[Bill512]

http://www.geotech1.com/forums/attac...4&d=1527807206 attachment from another thread.

Rx(two 1.5inch round coils connected figure eight) Tx(oval surrounding Rx). What is required to reduce sample delay to 2usec maybe 1usec? What measurements from existing circuit are needed if any? Maybe forget about existing circuit and define what is required to have sample delay at 1 or 2usec.

What is required to get the no target signal near zero volts at 1 or 2usec?

why not measure the "height" of the flyback voltage?
After all, in most PI designs, a significant percent of the "coil signal" is present in sampling times and the signals from targets are superimposed on the coil decay curve.
It is all there at zero time, coil, targets , ground - just the distribution of these, changes with time.
Of course, it will need a different circuit, but simpler than multiple sampling.
Done some quick experiments a decade ago on this subject (sensing the flyback voltage with a second coil ) and if I remember correctly, it worked and had discrimination as a bonus.
And where exists discrimination also there is Ground balance.
It was a sort range metal detector , something better than a pinpointer.
Technically not a genuine PI, but who cares.

[Ferric Toes]

Another scope picture, same coil as reply #47. Reduced peak current from 1A to .5A peak and adjusted damping on Rx coil. Holding a ferrite bead near coil still causes a change out to 4usec or greater. Is Tx coil current causing this or some other reason? Could sampling where the ferrite bead is detectable cause a problem? Scope picture of IB coil.

I see the trigger is one division different on todays scope trace, maybe 1.5usec delay today vs 3usec delay yesterday.

What I did was to reduce the coil current to 0.2A, then put up the repetition rate proportionally and adjust the integrator TC.

The ferrite bead could be giving a signal because of magnetic viscosity.

Eric.

[green]

Still working on a coil for 1usec delay to chart AWG28 solid wire TC. I tried another coil, better but not 1usec. Tried another, not much different than last one(maybe 3usec delay). Some scope pictures of the last coil. Observations from scope pictures? Ideas on what to try next? More information about present coil? I'm lost.

Another scope picture, same coil as reply #47. Reduced peak current from 1A to .5A peak and adjusted damping on Rx coil. Holding a ferrite bead near coil still causes a change out to 4usec or greater. Is Tx coil current causing this or some other reason? Could sampling where the ferrite bead is detectable cause a problem? Scope picture of IB coil.

I see the trigger is one division different on todays scope trace, maybe 1.5usec delay today vs 3usec delay yesterday.

[green]

Still working on a coil for 1usec delay to chart AWG28 solid wire TC. I tried another coil, better but not 1usec. Tried another, not much different than last one(maybe 3usec delay). Some scope pictures of the last coil. Observations from scope pictures? Ideas on what to try next? More information about present coil? I'm lost.

[KingJL]

What the H#!!, bought it anyway. If anyone needs some, let me know!

Maybe , would like to see a picture of the wire when you receive it thanks

Received my order of 5000' wire today.
Specs: 24AWG 19 conductor, tinned, 1 mm outside diameter, polyalkene primary jacket, polyvinylidene fluoride secondary jacket, 600 volt insulation rating.

I am super happy with it's performance. I wound a 1.5 mH (actually 1495 uH) coil for my Vallon test bench set-up... 10" mean diameter, 49 turns. With a 39 pf capacitor to isolate the injection source and scope probe, the coil resonates at 375 kHz. With an L of 1495 uH and and the 39 pf isolation cap, that results for ~80 - 90 pf for the coil. If anyone would have told me that I could make a coil of 1.5 mH, 49 turns @ 10" diameter and achieve less than 100 pf for the coil, I would have said they are crazy! I estimate that for a 10" 500 uH coil, I could achieve about 30 - 35 pf. With a 1 mm wire diameter, 49 turns make for a relatively thick bundle. But it fits well in my coil form.

I really do not have use for 4000' (1000' already spoken for)... anyone interested, PM me.

[KingJL]

Thanks for the reply, think I'm starting to figure it out. Still working on a coil for 1usec delay. Attempted to chart signal decay with fast decay and decay I think is similar to the impulse. Wondering if it makes sense. Target distance was adjusted to give a near full scale signal for fast decay with coins and foil(couldn't get full scale with ground). I'll have to try a full bridge circuit.

Thanks KingJL for posting it.

As Carl has said... "There is much fun to be had with this circuit".

[green]

The zeners aren't zeners, they're Schottkys. The Schottkys short the flyback to the rails and recharge the tank caps, thereby recycling most of the TX energy. You can run the sim without the tank caps and it will work the same as long as the supply and coil return are ideal batteries. You can replace the BJT switches with MOSFETs, then you can use the body diodes and eliminate the Schottkys. If you reduce the tank caps, you can tune the circuit and make a half-sine transmitter. You can then chop the trailing edge and make a truncated half-sine. You can also full-bridge the thing and double the TX current "for free." There is much fun to be had with this circuit.

Thanks for the reply, think I'm starting to figure it out. Still working on a coil for 1usec delay. Attempted to chart signal decay with fast decay and decay I think is similar to the impulse. Wondering if it makes sense. Target distance was adjusted to give a near full scale signal for fast decay with coins and foil(couldn't get full scale with ground). I'll have to try a full bridge circuit.

Thanks KingJL for posting it.

[6666]

In most designs I'm using a separate RX coil. In the Fisher PP I use a differential preamp and pick off both sides of the TX.

Thanks for reply

[Carl-NC]

In most designs I'm using a separate RX coil. In the Fisher PP I use a differential preamp and pick off both sides of the TX.

[6666]

Hi Carl in your experiments when you use full-bridge, where do you take the target signal from ? thanks.

[Carl-NC]

The zeners aren't zeners, they're Schottkys. The Schottkys short the flyback to the rails and recharge the tank caps, thereby recycling most of the TX energy. You can run the sim without the tank caps and it will work the same as long as the supply and coil return are ideal batteries. You can replace the BJT switches with MOSFETs, then you can use the body diodes and eliminate the Schottkys. If you reduce the tank caps, you can tune the circuit and make a half-sine transmitter. You can then chop the trailing edge and make a truncated half-sine. You can also full-bridge the thing and double the TX current "for free." There is much fun to be had with this circuit.

[green]

Not really! Current rate and shape is affected greatly by the value of the capacitances (and inductance)... not sure that I can help your understanding of the concept. You can only gain that understanding when you have analyzed it's performance with varying values of L and C for tens of hours... then a light switch goes on and you say... "OH I SEE"!

Try clicking on L1, D1 and D2 to display current.

[KingJL]

I think the discharge current is going thru the zener diodes like in my simulation...

Not really! Current rate and shape is affected greatly by the value of the capacitances (and inductance)... not sure that I can help your understanding of the concept. You can only gain that understanding when you have analyzed it's performance with varying values of L and C for tens of hours... then a light switch goes on and you say... "OH I SEE"!