Bipolar PI design

[Carl-NC]

Just turn the GB pot fully to the side of the target channel.

[pechkata]

The GB in this design is done differently than the TDI but the results are the same, except for where the target hole ends up. Any time you have trouble detecting a target turn off the GB and try it.

Hello Carl NC
How do we do this - with a key?

[Carl-NC]

See this post:

https://www.geotech1.com/forums/forum/inside-the-metal-detector-aa/itmd3/pi-designs/432900-pi-gb-design?p=436095#post436095

[algan]

Hi Carl. Are these changes enough to change the width to 100us? I just don't know which ones I have circled in red and whether they should be 100us. Thanks.

[eduardo1979]

Thank you very much, Carl and Altra, for your answers. I'll try changing the sample widths of the GNDN and GNDP channels and matching them to the samples of the TGTP and TGTN channels. I'll also change the sample times and set them to the same as the TDI. In short, I'll use the same time base as the TDI, and I'll also use a 3 kHz frequency and a 100 µS pulse width. I hope this works.

[Carl-NC]

The TDI is a hotter design, everything else equal (sample timings & GB) the TDI will always go deeper. At a minimum the TX is 100us and 3.2kHz which burns a lot of power. The BPI is a circuit whose purpose is to demonstrate basic GB so I chose a conservative setup to keep people out of thermal trouble. However, the GB mechanics between the TDI & BPI are the same, just done in different ways. You should be able to alter the BPI to mimic the TDI in terms of GB operation, with less overall depth. The preamp BW is part of the GB equation so it may also need changing.

When changing the TX width/frequency, probably what needs to be done is to also change the ground sample width to maintain the same overall GB range. I did not do that in the code but it should not be difficult.

[Altra]

Ok, maybe the same problem TDI tx pulse width 100uS and yours is 50uS. That is less power going to coil assuming the same Tx frequency and less sensitivity.

[eduardo1979]

Hello Altra.
Let me explain my problem. The Bipolar PI with its original GB, just as Carl designed it, works perfectly and rejects any hot stones very well. My complaint is that this type of GB, I think, significantly reduces the depth of the detector. I've tried replacing the GB circuit in the Bipolar Pulse with the TDI's GB circuit, and it doesn't work. The TDI's GB detects all hot stones. I don't know what's going on there, because in theory it should work. I've used the TDI's GB in many detectors and it's worked well the first time, but it doesn't work in the Bipolar PI, and I don't know why. I like the TDI's GB because it doesn't reduce the depth of small gold objects like the Bipolar PI's GB does.

[Altra]

Hello Altra.
The pulse width is 50uS and the frequency is 2000Hz.

I have found with BP you need a Tx pw of greater than 70us to eliminate certain ground. The ground signal at 50uS tx does not have enough power during the the GB sample to cancel the ground signal during the target sample.

[eduardo1979]

Hello Altra.
The pulse width is 50uS and the frequency is 2000Hz.

[Altra]

Eduardo what is your Tx pulse width?

[ivconic]

This type of subtractive GB always results in a target hole and it's possible the Ottoman coin is in this hole. Turn off the GB and see if it can be detected. If so, it is in the hole. You can move the hole around by changing the timing parameters but then something else will end up in the hole. The book describes how to eliminate the hole.

Sure will do!
First to find the damn 0805 caps and fit them too.
The new ones I ordered will reach me God knows when.
Then to wind a better coil and to "jig" an adequate damper.
And only then can I approach any serious tuning and testing.
By the way, Carl, one "stupid" question; considering this is an H-Bridge TX... what cable to coil do you recommend I use?
Currently on the video is a coaxial cable, the middle one is PVC coated and braided around it.
In DD configuration and in mono configuration, do you have any special advice?

[eduardo1979]

The TDI with the GB ON (at number 8 ) detects the nugget at 10 cm. The Bipolar PI with the GB off detects it at 3 cm, and with the GB potentiometer halfway through the range, right where the hot stone is rejected, the nugget is detected at 1 cm. I'm surprised that the TDI, with the GB in any position, doesn't lose depth for the 0.5 g gold nugget, and the Bipolar PI loses depth when changing the GB value. I understand that both detectors have subtractive GB, but in different configurations. I've tried modifying the GB on the Bipolar PI to match the TDI, and it doesn't work. There's no way to reject the stones. I've put a 1 Mohm potentiometer on the GB to get a good adjustment range, and it doesn't work either.
Any ideas, Carl?
I've attached the schematic of my circuit, the one shown in the video, in case it helps anyone.

[Carl-NC]

The GB in this design is done differently than the TDI but the results are the same, except for where the target hole ends up. Any time you have trouble detecting a target turn off the GB and try it.

[eduardo1979]

I just did a test with a 0.5g gold nugget, and it detects it very faintly. The TDI detects this nugget about 10cm in the air. The TDI also has a different GEB circuit. I assume that could also influence the depth.

https://www.youtube.com/watch?v=3jIz7-Rm__I