Voodoo Main Discussion

[Qiaozhi]

G'Day Qiaozhi
Thanks for the files REV-C.zip file only contains Gerber files to make the pcb. The zip doesn't contain the schematic that were used to create the Gerber files
Ivconics sch file is not compatible with KICad it is either Protel or Eagle

The original EasyPC project, schematic and PCB are definitely contained in the zip file.
They have the extensions: prj, sch and pcb. Of course, you will need EasyPC to read them.

If you want to create your own version, then I would use KiCad.

[phred01]

Having studied the circuit noticed the whole circuit uses positive as ground like old British cars. Is there a logical reason for this approach?

[Qiaozhi]

Having studied the circuit noticed the whole circuit uses positive as ground like old British cars. Is there a logical reason for this approach?

Carl's Hammerhead design also uses battery positive as ground (0V).
In Chapter 2 it explains that the Voodoo Project started as an experiment to combine Hammerhead and Raptor (a VLF design from ITMD) as a means of providing ferrous/non-ferrous discrimination.
It's just a different power supply topology. The Arduino Nano PI Project (ANPI), for example, uses a more conventional approach.

[Carl-NC]

Having studied the circuit noticed the whole circuit uses positive as ground like old British cars. Is there a logical reason for this approach?

In a typical PI, the battery voltage is applied to the coil and then, when it's removed, you want the coil to decay to ground potential. When using a low-side switch, if the coil is connected to ground then the switch must be connected to -V. The easiest solution is to just relabel the battery so the + side is ground.

[phred01]

Happy New yr to all.
There are 2 a ways to control the current in the TX coil. In the Voodoo case, the case it is source side of the power FET. What is across after turn on pulse is a ringing coil with a loading 1k and 2 diodes. In the MPS PI - Iron discriminating metal detector design the coil is on the drain side of the FET with 300 ohm permanently across it. The difference with the Voodoo case TX & RX coils are both used to receive echo signals from the target. Where as in the MPS-PI only the RX coil is used;. The MineLab metal detectors use a complex cct capacitor diodes to quench the back EMF to about 185 v The inputs of the receive op amps have 2 FETs to prevent of being fried. The only downside of the Voodoo design I see is the digital side is a little bit harder to debug in case of a problem. So far in my view the Voodoo approach is novel.

[kraimov]

Good work thanks

[Voyager]

Hey kraimov, why are you using my signature?

[scs-bg]

I made a hybrid pulse detector 20 years ago, long before Voodoo. It turned out that due to the influence of the ground, incorrect discrimination of deep-lying objects occurs.

[scs-bg]

The best working hybrid detector is with a DD coil. Sound from a pulse induction channel. Discriminator with a meter with an arrow. Separate channel, signal when transmitting. Both channels without movement, with a reset button. Both channels work with a signal from the receiving coil. In this way, the influence of the ground is weak and a great depth is achieved. Frequency 220 hertz.

[zmaj od nocaja]

Hello. I made a Vudu detector, but I have a problem. I got the wrong picture from China instead of 18f4520, I got 18F452. The programming went fine and everything works as far as the program is concerned, but in any mode there is no detection sound. At test point 20 I see a voltage change when metal passes over the coil in pulse mode, but at test point 19 nothing happens. Could this be a problem with the PIC 18F452?

[Qiaozhi]

Most likely it's a problem caused by using the PIC18452.
The PIC18F4520 is backwards compatible with the PIC18F452, but not the other way round.

[zmaj od nocaja]

Thank you very much. I'll have to get a pic18f4520 and try it out.

[zmaj od nocaja]

Hello! I received the original pic 18f4520, it was successfully programmed and now everything works perfectly in hybrid and pulse mode, but the 79l05 heats up quite a bit and with a supply voltage of 12v, the total consumption is about 180ma. Is this some component drawing too much current or is it normal consumption?

[Qiaozhi]

180mA seems too high, especially as the 790L5 is rated at 100mA.
Are all the voltages at the correct level, or are they being dragged down?

[zmaj od nocaja]

The total current drawn by the detector from the battery is 180ma. The current flowing through the 79l05 is only 3.4ma but it still gets quite hot. The voltage on the test pin TP14 is -5.01v and on TP15 it is +5v.