Voodoo Main Discussion

[useus]

do you have the full schematic?

[SilverSurfer17]

I took it down to 3 batteries 11.1 volts nothing else inline and same thing no change

[surfdetector]

I took it down to 3 batteries 11.1 volts nothing else inline and same thing no change

You need to troubleshoot. Have you tried what I wrote to you in #1271?

[Sean_Goddard]

Re alleged lack of depth:

I would have done things a different way George.

The silver sabre micro was rubbish because the design used the ADC (12 bit) in the PIC to determine beep or no.

I would have used all analogue detection channels and merely used the PIC to determine a tonal frequency for the audio. Deep beep but no disc reading? Dig it!

As Mark Rowan at Whites once said, "We decided not to let the micro decide whether to beep or not, but sound off on everything and let the users ears decide", or very similar. Look how deep the Eagle Spectrum goes as a result.

I'm going to rework the Voodoo design using this method and see how much depth I can get.

BTW, the above is the approach the Pulse Devil uses, and that is DEEEEEEEEEP.

[SilverSurfer17]

No i didn't see that you posted that i don't have a damping resistor yet i thought thats what the jig was for was to find out what resistor you need by tuning it to be damped then ohm out the jig to see what resistor you need and I'm pretty sure it kinda works already i can see the wave bounce when i run metal past the coil but how can i see what resistor if the wave isn't correct to start with also im starting to think those higher wattage resistors maybe a factor because i noticed when i turned the voltage down to almost 6 volts the wave looked a lot closer

[SilverSurfer17]

Im going to just order 1/4 watt resistors and replace those monsters i put in can i use just the jig or do i need to order a certain size like 10.5 or some starting point resistor ?

[surfdetector]

im starting to think those higher wattage resistors maybe a factor because i noticed when i turned the voltage down to almost 6 volts the wave looked a lot closer

The higher wattage rating is not going to effect the working of the circuit. Ohm ratings are a different matter. You need to check the ohm ratings of everything that you have landed on the PCB so far. You should also prove your transmit pulse is correct as I told you previously prior to continuing.

[dfbowers]

I wouldn't say that it's a "matter of concern".

The intention of Voodoo was to produce a PI detector that was capable of rejecting iron, and was able to find non-ferrous targets hiding among ferrous trash.
A standard PI (one that detects all metal) will always have greater depth when compared to Voodoo because it uses a self-adjusting threshold (SAT). i.e. a slow single-differentiating filter rather than double-differentiating.

You can easily determine if a detector is non-motion (NM), single-differentiating (SDiff), or double-differentiating (DDiff) with a simple test. Hold a coin stationary a few inches from the coil. If the audio tone does not die away then the detector is NM. If it does, then it's either SDiff or DDiff. Now move the coin to one side away from the coil. If you don't get a beep it's SDiff, but if it beeps it's DDiff.

A detector with SDiff filters (or SAT) will start to detect a target as it approaches the side of the coil, whereas a detector with DDiff filters will start to detect the target as it crosses the centre of the coil. The bottom line is that the detection area is smaller for DDiff, which is why it has far better target separation than SDiff. For a standard PI, target separation is not an issue. However, for Voodoo (and for the majority of VLF detectors) DDiff is necessary to be able reduce the effects of iron masking.

In my test garden (which is littered with ferrous trash) any detector with SDiff filters, or extremely slow DDiff filters, will fail miserably when attempting to find the buried non-ferrous targets. If you use a standard PI in the test garden the audio beeping sounds like a machine gun as it detects everything.

This is why I suspect Don's variant will have difficulties with target separation due to the use of a self-adjusting threshold, and ground effect in the discriminating channel.

Yes, absolutely. We are comparing oranges to apples when it comes to target separation.
Mine shows no promise of that at all. It?s only good as an aid for identifying iron targets in open ground, or making a PI useful in parks or lawns. The places I have tested mine were relatively free of iron clutter and works wonderfully there.. Just rejecting Bobby pins, barbed wire, nails and horseshoes is about it..

[eclipse]

The silver sabre micro was rubbish because the design used the ADC (12 bit) in the PIC to determine beep or no.

What's wrong with the 12 Bit ADC beeper? If the signal gets amplified high enough /not a direct sampling/. Sampling speed is more important.
I've got such design in the box waiting to be soldered and tested. I have made both digital and analog disc options /via pluggable mini boards that plug into the main board/ so I can compare both solutions.

[Matt_Rowe]

Im going to just order 1/4 watt resistors and replace those monsters i put in can i use just the jig or do i need to order a certain size like 10.5 or some starting point resistor ?

did you check every part manually before you installed it, very good point the surfdetector made a video on, and in making the video proved a resister, or a part was bad right out of the package.
its a little chore, but he proved well worth the effort. all new parts are not good from the manufacturere in electrical parts, same with autos and our electrical parts for us electricians

d

[SilverSurfer17]

Can i use the jig to check that for the damping resistor or does the tx use 10.5 and the jig is for rx i dont understand the purpose of the jig if there is a calculation to figure out the damping resistor and the book says trial and error is best

[SilverSurfer17]

Here is what the output of tx is with jig installed and the prob on pos and negative on neg and channel 2 on tp2

[Sean_Goddard]

Two words; Dynamic range!

ADC's give a theoretical 6dB per bit. Amplification has nothing to do with this issue.

To make a rally SENSITIVE, and deep machine, you would require some 124dB of dynamic range. You can do this 2 ways, switched gain or more bits.

Robert Hoolko proposed switched gains (3dB = gain of 2X ) so 60dB is gain of 1x10^6 or 1 million. In emails to him, I proposed more bits specifically 24 with a 192khz sample rate. After we analysed the merits of both methods, more bits won because switched gain introduces response lag (like an auto ranging DMM). I would use an I2C 18bit ADC minimum for a detector, anything less is false economy.

[KingJL]

Two words; Dynamic range!... 18bit ADC minimum for a detector, anything less is false economy.

Totally agree... I have been preaching that for 3+ years. Twelve bits simply just do not cut it!

[SilverSurfer17]

I don't want to have to buy 15 different ohm values for it if there is a way with the jig to know the exact ohm value of what it needs from what ive seen those type resistors aren't cheap i knew i would have problems with the tuning part i didn't think as much building the board ill check all the values later on the resistors to make sure they are what they are suppose to be