Tweet
Received the parts to try Moodz circuit, hope I have what I need. I have a standard H bridge on the bench now, similar to https://www.geotech1.com/forums/atta...9&d=1623413764. Want to compare the two circuits with the same coil(Rx133mm fig8, Tx328uh 3.4 ohms)repeating test in reply #53 https://www.geotech1.com/forums/atta...1&d=1623792682. Any thoughts what I might add or do different?
-
-
Scope pictures of amplifier out with standard H bridge, Rd adjusted for best signal. Unipolar circuit(reply #53)looked faster. Going to modify circuit to control Moodz CCPI circuit. Anyone think of something to try before I make the change?Originally posted by green View PostComment
-
I assume your tests with the unipolar & H-bridge were with a mono coil. With the CCPI circuit you will absolutely have to use a separate RX coil, IB or not. Once you figure out the CCPI coil you may want to go back & test the other circuits with it to keep comparisons on equal footing.Comment
-
ThanksOriginally posted by Carl-NC View Post
I did use separate receive and transmit coils, Rx_133mm figure8, Tx_oval over Rx.Comment
-
Yes, I had forgotten that.Comment
-
Finally got moodz CCPI circuit working after zapping a couple resistors and a few IC's, wiring errors. Loosing more energy than the spice simulation when switching polarity. Tried both polypropylene and NPO ceramics across the coil, didn't see a difference. Have some different series high voltage MOSFETS to try. Was hoping to use my existing coils, Tx resistance 3 to 4 ohms. Seems to work alright in spice. Any other suggestions what I might try?Comment
-
Lower resistance coil <1 Ohm, lower resistance Ron mosfets perhaps?Comment
-
Hmmm ... I am not treating it as a pulse induction style waveform for RX.
I did use a PI to generate the gate drive though.
I did not use an amp either ... CRO is connected directly to coil RX and TX.
The TX coil impedance is 3.4 mH and several ohms.
I am going to dig out a linear hall effect sensor to measure the mag field .... post later :-)
moodzComment
-
Are these pictures taken with the same circuit from reply#2? (https://www.geotech1.com/forums/atta...622687902)from reply#2Originally posted by moodz View PostComment
-
YES except ...Originally posted by green View Post
The same circuit but NOT the same coil .... since I am a PI man ( not Homer LOL ) and mostly mess with mono coils .. I had to find a coil more like the one I think you were using.
The coil in the picture has one winding that is 400 microHenry and another that is 3.4 milliHenry. I was going to connect the TX to the 400 coil but it would not work ( looked like there were diodes across it shorting the flyback ) ...
so I thought what the heck I will connect it across the 3.4 mH coil and thats what I measured at the 400 coil ... better result than I get with PI coils LOL.
So the peak current is not so high the flyback is lower and takes 6 microseconds ( prob because its 3.4 mH ) not 500 nanoseconds ( used 300 uH Pi coil ) as per the link above.
Its a commercial coil .. dunno whats inside ??
... looked everywhere and cant find my linear hall sensors .. ( if you dont need something you will trip over it ... if you do need something it will not be there )Comment
-
One of my goals lately has been to detect #9 lead shot with a PI. Appears I need to start sample 3.5 to 4us after Tx off or polarity switch(CCPI). Including some scope recordings with unipolar, bipolar and CC Tx that I have been playing with. The ferrite core has the least effect with the unipolar Tx. I have moodz CCPI circuit working with spice. I get more lose with my real circuit when polarity switches so my spice circuit won't work. I like moodz circuit, should be able to increase Tx pps 4times faster than unipolar(1000pps to 4000ps). Haven't given up on moodz circuit yet, need another approach to keeping Tx current constant during target and ground sample time or reducing lose when Tx polarity switches. Any suggestions appreciated other than reducing Tx resistance(would like to use my existing coils if possible)The unipolar Tx seems to work but I think moodz CCPI would be better if I could get it working.Comment
-
I have a GPZ7000 detector that uses a similiar transmit waveform and the way it does it is not simple .... but I do find alot of lead shot with it.Originally posted by green View Post
My Magpi project can detect very small shot and samples in the range you want .... https://www.geotech1.com/forums/show...462#post289462
The CCPI circuit is only a few weeks old ... I have not done enough bench time yet to meet your requirements.
moodzComment
-
Playing with moodz CC circuit on the bench again. Changed R2(polar2g)to 3.33 ohms. M1 and M2 not the same as simulation. Looks like it corrects CC like the simulation. Want to try some different MOSFETS for M1 and M2. Increase Tx coil current and frequency. Any other suggestions to try? Wondering why amplifier doesn't come out of saturation sooner(appears Tx pulse is about 1.5us).Comment
-
Tried some MOSFETS with lower Coss. Was thinking it might help. Not. Circuit works but can't get Tx constant current(A/sec)after reversal as low as I would like. Still not sure how low it should be. Maybe my coil is causing some of the problem. Back to playing with unipolar and standard bipolar.Comment
-
The transmit coil in the GPZ7000 is very low resistance ...
I think the designers of the GPZ would have had lots of trouble eliminating your issue which partly is due to coil resistance.
Comment
Comment