Experimenting with the vmh3cs transmitter

[KingJL]

Tx current is probably regulated by feedback to the FPGA...

Actually I think (the operative word here is "think" as I do not know this for sure) the TX maximum current is controlled by the sampling network(s) under the 2 HV capacitors in conjunction with the DG442 on the botton side of the TX board providing feedback to the TL061 (furthest away from the HA14) which together with the TL061 right next to it control the 20V regulation by the FR9220. The timing for the DG442 sampling is controlled by the FPGA. I stopped tracing this portion for 2 reasons (1) tracing out the networks under the 2 HV caps introduced a high probability of irreparable damage to the TX PCB and (2) fear that I may destroy another digital board by an inadvertent probe slip (this had already happened once and Eric Foster saved the day with another digital board). My goal was to learn as much as I could and still come away with a working detector if possible. One thing I try to do is never lose sight of the original goal!

[green]

Hi Green, I am no expert. I just study the patents and do actual experiments.

Pure ferrites don't develop eddy currents, so once the current quits flowing their influence on an inductance balanced coil decrease. While conductive metals develop eddy currents and continue to affect the coil balance after the current becomes steady . Iron exhibits both a strong X and lessor R component, this affects the coil balance in the opposite direction. Sorry, I can't give a better explanation.

Hi Altra, I'm no expert either. I have read ITMD, this form and do actual experiments. Normally don't look at patents. I think iron effects coil balance in the opposite direction only when coil current is changing. Do you think the traces in the above patent JPG are during coil current change or after the current is constant?

[Altra]

@JLKing

I mentioned the Tx current because in their patent I recall reading they use some form of feed back control. By the way if you did not know the "HA14" is a 74HC14 inverting hex schmitt trigger. I started to trace out the power supply. Same as you I quit tracing before destroying the pcb. I did remove the wima 0.01uf under which were a string of zener diodes going back to the HA14. My thoughts on the series zeners was to reduce the high voltage to a low voltage divider network and then fed back to the power supply?

[Altra]

Hi Altra, I'm no expert either. I have read ITMD, this form and do actual experiments. Normally don't look at patents. I think iron effects coil balance in the opposite direction only when coil current is changing. Do you think the traces in the above patent JPG are during coil current change or after the current is constant?

I really can't explain anymore without putting my foot in my mouth. Three of the friendly inventors mentioned in the above patents are members here. May be one of them can answer your question.

[KingJL]

By the way if you did not know the "HA14" is a 74HC14 inverting hex schmitt trigger.

but with better and more robust specs than other variants both in voltage limits and propagation delay, but yes basically a 74HC14.

My thoughts on the series zeners was to reduce the high voltage to a low voltage divider network and then fed back to the power supply?

Probably correct.

[green]

I really can't explain anymore without putting my foot in my mouth. Three of the friendly inventors mentioned in the above patents are members here. May be one of them can answer your question.

I don't mind putting my foot in my mouth. Just interested in the correct answer. Not saying you are wrong. Just I see it different. I've been wrong and corrected more than a few times, how many times I've been wrong and not corrected is any ones guess. Trying to make sense of the drawings in
https://www.geotech1.com/forums/atta...7&d=1576345487 Was thinking maybe they were low pass filtered with the X signal bleeding into Tx off time but if that were true the penny would be opposite polarity when Tx is decaying. Maybe someone will straighten me out.

[Altra]

I don't mind putting my foot in my mouth. Just interested in the correct answer. Not saying you are wrong. Just I see it different. I've been wrong and corrected more than a few times, how many times I've been wrong and not corrected is any ones guess. Trying to make sense of the drawings in
https://www.geotech1.com/forums/atta...7&d=1576345487 Was thinking maybe they were low pass filtered with the X signal bleeding into Tx off time but if that were true the penny would be opposite polarity when Tx is decaying. Maybe someone will straighten me out.

Green, not trying to be difficult. I truly can't explain everything I do. But the diagram I posted from Earle's patent is worth a 1000 words. It contains Tx voltage, Tx current, no target response, ground response, copper response and steel response. That is all I need to know, that constant current in an IB coil is worth studying. I do not doubt John Earle's observations/diagrams, he is quit accomplished.

By the way while we are conversing. I been meaning to ask you if you have ever tested the TC of the foil emergency blankets? The one I have is made by Coleman. You can find them in the outdoor department in Walmart for $3 or $4. I cut two 24"x24" sheets and taped them to a piece of card board. Much like your foil squares test you posted in the past. I think it simulates saltwater. My non scientific test showed at 12 to 14uS it goes quite. The large surface area, I believe it simulate the beach better than a small piece of foil with the same tc. I also use a jug of concentrated saltwater which is not optimum. Something to investigate.

Have a good evening

https://www.coleman.com/emergency-bl...000016485.html

[green]

Green, not trying to be difficult. I truly can't explain everything I do. But the diagram I posted from Earle's patent is worth a 1000 words. It contains Tx voltage, Tx current, no target response, ground response, copper response and steel response. That is all I need to know, that constant current in an IB coil is worth studying. I do not doubt John Earle's observations/diagrams, he is quit accomplished.

By the way while we are conversing. I been meaning to ask you if you have ever tested the TC of the foil emergency blankets? The one I have is made by Coleman. You can find them in the outdoor department in Walmart for $3 or $4. I cut two 24"x24" sheets and taped them to a piece of card board. Much like your foil squares test you posted in the past. I think it simulates saltwater. My non scientific test showed at 12 to 14uS it goes quite. The large surface area, I believe it simulate the beach better than a small piece of foil with the same tc. I also use a jug of concentrated saltwater which is not optimum. Something to investigate.

Have a good evening

https://www.coleman.com/emergency-bl...000016485.html

I'll try to measure the TC of a 24x24 inch piece. Are you stacking the two pieces or are they side by side?

[Altra]

I'll try to measure the TC of a 24x24 inch piece. Are you stacking the two pieces or are they side by side?

Hi Green,
After posting I rechecked my set up and it's only one sheet. But you might find it interesting to stack more than one and see how it changes. Also notice one side is conductive and the other side is not. Maybe two conductive sides touching will be different than the opposite. One problem with the idea, we do not have a TC reference for wet salt sand. Some where on the forum it was stated to be in the low teens like 11-15us? On vlf detectors, depending on the frequency salt has a phase of about 5 to 7 degrees. In January after the holidays I'll try a beach test to establish some sort of reference.

[Ferric Toes]

'constant current in an IB coil is worth studying'. My thoughts are that there will be no response from iron mineralised ground only if the mineralisation is uniformly homogeneous. In reality this is rarely the case, particularly in the goldfields of Australia. Switching on a magnetic field has exactly the same effect as switching it off even if the current and field remains constant in between. Magnetic relaxation will occur with opposing polarities and will only cancel if the amplitudes are the same. If the coil is swept over pieces of ironstone rock sitting on, or in, gravel of all sizes down to fine dust then this method of GB does not work. The methods using sampling and subtraction would still be needed as, unlike metal targets, the decay time for iron mineralised ground is more or less the same whatever the quantity. Only the amplitude changes. At least that is my experience.

Eric.

[Altra]

'constant current in an IB coil is worth studying'. My thoughts are that there will be no response from iron mineralised ground only if the mineralisation is uniformly homogeneous. In reality this is rarely the case, particularly in the goldfields of Australia. Switching on a magnetic field has exactly the same effect as switching it off even if the current and field remains constant in between. Magnetic relaxation will occur with opposing polarities and will only cancel if the amplitudes are the same. If the coil is swept over pieces of ironstone rock sitting on, or in, gravel of all sizes down to fine dust then this method of GB does not work. The methods using sampling and subtraction would still be needed as, unlike metal targets, the decay time for iron mineralised ground is more or less the same whatever the quantity. Only the amplitude changes. At least that is my experience.

Eric.

Hi Eric

Thank you for your valuable input. I was thinking if you took a later sample during the on time and subtracted it from the early sample it would remove enough of the ground signal. Allowing for ferrous / non-ferrous polarity test. This is not a commercial project so Australia's heavy minerals is not a factor, but it seems to be the benchmark.

Best regards

[green]

Hi Green,
After posting I rechecked my set up and it's only one sheet. But you might find it interesting to stack more than one and see how it changes. Also notice one side is conductive and the other side is not. Maybe two conductive sides touching will be different than the opposite. One problem with the idea, we do not have a TC reference for wet salt sand. Some where on the forum it was stated to be in the low teens like 11-15us? On vlf detectors, depending on the frequency salt has a phase of about 5 to 7 degrees. In January after the holidays I'll try a beach test to establish some sort of reference.

What I tried so far. Didn't get a usable signal with two mono coils or the 8inch figure8 Rx. Got a usable signal with 48mm figure8. Decay looks good but doesn't make sense. Including a chart. Tried to find aluminum thickness for a space blanket to calculate what TC should be without success. Just replied to show I tried, will try again. Might need to change something to measure a TC less than .8us.

[Altra]

Thanks Green, with the holidays I'll be Away Until January. Will try some more test then.

Have a nice Christmas and new year

[KingJL]

... I am leaning toward experimenting with the Analog Devices (ADUM3220) units sometime in the future.

@Altra, I believe the the ADuM4120 will work. I have a simulation of my TX working with a ADuM4122 which is very similar. The key is to use a diode between the coil and the high side MOSFET source. This isolates the capacitance of the bootstrap storage capacitor (in my case 1 uF).from the coil. My design is a full bridge design with the boost always provided by the high side, but everything can be configured to work with your TX+ and TX- also. Using the ADuM, the parts count is reduced from 9 down to 4. Of course, one of the parts eliminated is the toroid! Timing is accurate and extremely low propagation delay.

[green]

What I tried so far. Didn't get a usable signal with two mono coils or the 8inch figure8 Rx. Got a usable signal with 48mm figure8. Decay looks good but doesn't make sense. Including a chart. Tried to find aluminum thickness for a space blanket to calculate what TC should be without success. Just replied to show I tried, will try again. Might need to change something to measure a TC less than .8us.

More tests. I think the reason the decay didn't make sense(REPLY #57) is a large input causes the amplifier to overscale and the decay is the amplifier stabilizing. A large ferrite core toroid causes the same decay signal. Recorded the coil signal with the 20 and 140 layer targets reply #57. Differential Rx coil signal with the 140 layer target over 2V, 6mV over scales the amplifier. Didn't find the aluminum thickness for the space blanket but did find 3 to 50nm thickness for metalized Mylar. If I did my math correctly 50nm would give about .1us TC for a piece 24 inches square. Would have thought the 140 layer target might chart over 1us TC. Maybe to much space between layers? The layers were compressed between two layers of cardboard, think mostly Mylar spacing not air. Charted a 1, .5, and .25 inch piece of regular strength aluminum foil. Think the next smaller piece should have a TC half the larger. Think maybe my circuit response is causing the shorter TC targets to chart higher than expected. Think coil wire size, coil L/Rd and amplifier response effect the accuracy of he charted decay. Think their TC should be at least half(probably less)the desired target TC? To record a .1us TC, wire size AWG34 or less, amplifier response about 3MHz and coil L/Rd about .05us(circuit resonance=6MHz)? Having a salt target would be nice. Search salt gives 4 pages, search salt in the title gives 1 page. https://www.geotech1.com/forums/show...highlight=salt a thread that has some replies for salt response. The replies suggest a TC closer to 1us than .1us. I'm thinking if the 24 inch square piece of space blanket has a TC close to 1us I could measure it. Don't know what salt TC is and how much it varies with location? More questions than answers.