PI-1 ITMD2 Question

[cboyer1324]

Hello,

I recently built the PI-1 detector circuit from the 2nd edition ITMD book and used the exact part list from the book. I am able to get a detection depth of about 6-8in with a 400uH 7in diameter 20awg magnet wire coil. I have found that when I use a 9v (6AA battery pack) in combination with the 20awg wound coil the IRF9630 P-FET quickly overheats. I was wondering if changing the coil to 26awg wire and using a single 9v transistor radio battery would dramatically limit the current by introducing increased resistance in the path? I also was wondering if it would help to use a 12v supply instead of the 9v supply, driving the P-FET fully into saturation. This way RDS(on) can remain as low as possible. With the 9v I noticed only around 8v is getting to the gate which operates the FET in the high end of the linear region. What would be best for stabilizing the PFET temperature runaway?

Thanks

[Carl-NC]

The FET switch normally gets hot so usually it's heat-sinked. A higher R coil will help, a small 9V battery maybe not so much because C1 supplies most of the transient energy. The best way to limit FET heat is to reduce the TX pulse width and/or the TX frequency.

[cboyer1324]

Thanks Carl,

I will give this a try!

[cboyer1324]

Replacing the 20awg coil with a 26awg coil made the overheating issue noticeably better on the breadboard circuit. I then used the ZIP circuit board PCB GERBER from the forum to have my PCB board produced and shipped to me. When I soldered the first stage to the PCB board, I found that the FET signal looks fine on the scope but overheats both the coil and the FET. I then soldered the rest of the circuit and found that the audio does not output when connected to a speaker. I have gone through all the typical stages of debug with the scope and multi-meter. Please let me know if there are any known issues with the PCB GERBER layout from the forum. I was also wondering if there is a confirmed working version of the PCB GERBER for the PI-1 ITMD2 circuit or images of the finished product to confirm part orientation. Any help would be greatly appreciated!

Thanks

[cboyer1324]

This is where I got the GERBER from on the forum Gerber Files - Geotech Forums

[Carl-NC]

I think the PCB is correct except for potential transistor pinout problems, which can arise depending on where you source your parts. This thread talks about Q3 which would not cause your overheating problem:

https://www.geotech1.com/forums/foru...-pi1-pcb-error

But a Q1 pinout problem might. Make sure it is correct. If you still have the problem, disconnect the coil and post the waveform at Q2's gate.

Edit: AFAIK, the PCB was correct except maybe for the pinout of Q3. I know I built it on the PCB and it worked, but that was 14 years ago and I don't have the PCB any more.

[Qiaozhi]

Look at page 202 in ITMD2.
The problem with transistor pinouts is discussed there.

[cboyer1324]

I have read through that section of the book (page 202) but cannot see which orientation is stated that is used for this detector. I just measured all of the transistors in the circuit and found them as following: 2n3904/3906 (EBC), PFET (GDS), and JFET (DGS) I measured the gate in the middle. I am assuming this is the correct configurations.

[Qiaozhi]

I have read through that section of the book (page 202) but cannot see which orientation is stated that is used for this detector. I just measured all of the transistors in the circuit and found them as following: 2n3904/3906 (EBC), PFET (GDS), and JFET (DGS) I measured the gate in the middle. I am assuming this is the correct configurations.

There is a note in the errata that the silkscreen is incorrect for Q3.
I've also noticed that the base of Q1 is shown in the schematic as being connected to R20, but the PCB silkscreen shows it being connected to R21.
However, it is clear that the correct pinout for the TO-92 transistors is the USA version shown in Fig. 13-5.

Perhaps Carl can explain what's wrong with the Q3 silkscreen? I'm failing to see what's wrong, except that it looks like one of the connected resistors has the wrong reference designator.

[cboyer1324]

After some investigation with the oscilloscope this weekend I was able to trace the issue back to the 555-timer pinout. In the book R20=10k, R21=12k, and R22=220k. This created a pulse that turned the FET on 80% of the time not the other way around as it should have been, causing the overheating. The ITMD edition 2 should be noted for correction as the actual values should be R20=220k, R21=10k, and R22=12k in correlation with the GERBER file on this site forum. I was able to swap the resistors, and the PCB works great with good windowing depending on what R2 is set to for tuning.