Help with new gold detector design

[dom44]

here you go this part i know works.


http://img7.imageshack.us/img7/9241/pictureun.jpg

[ivconic]

This circuit was copyed ........ I will use the other 2 gates to make a VCO for target signal to vary amplitude as well as frequency on a target.

This is most interesting part for me! I would like to see more advance solution that suits those demands. How to design "high resolution" audio vco, which will vary amplitude and frequency in wider range.. even if input signal is short and momentary, as it usually is...

[dom44]

Hi ivconic the one i have posted is a simple design which should work with most of the tesoro detectors presented, I came up with the design my self by tinkering arround, I agree the sound part of a detector is neglected but minelab seam have done some work in that direction, amazingly this design sounds very much like a sd2000 but a completely diffrent circuit but you can split the input signal through the diodes and add other usefull information, to the target signal I would think.

[Qiaozhi]

Audio sample of the VCO

http://hotfile.com/dl/121494312/8255...32900.amr.html

here you go this part i know works.


http://img7.imageshack.us/img7/9241/pictureun.jpg

If you upload these files directly to Geotech as attachments, it would be a lot easier for other members to access. They will also remain part of the thread, and not disappear over time as happens with remote file sharing. If the file is unsupported, such as the ".amr", then simply attach it as a zip file.

[ivconic]

I just hear audio file you attached; yes it is interesting. I asked for "high res" audio vco because i've been struggling for years to solve that on more better way, using analogue approach. Relating to your first post here; here is probably one of the possible way to solve it:

[dom44]

thanks Qiaozhi I have done that, And thanks Ivconic I take it one pot is for nulling and the other for phase?

[ivconic]

Offset and gain. Directly affects phase, of course. C1 should be few picofarads. Other values are debatable. Someone more skilled with spice should check this...

[porkluvr]

Looks like everything is under control. Hmmm, interesting VCO.

I only have one important thing to add:

I believe there is a pull-up resistor missing from the output of the phase reference comparator in the schematic I posted. The demodulator will not work without it. That's the LM392 pin-1.
Maybe 20k~24k ??

edit:
Just about any adjustment you make to component values will have unintended consequences. When you jump from 20 to 32kHz you will have a wider adjustment range from your ground pot which may make the ground cancelling more difficult. You can change the 24k resistor to about 33k and reduce the adjustment range but that may throw off your phase 'zero' point. You could also play with the 100pF value to change the range and zero point of your ground pot. You get the idea. Many possibilities.

[dom44]

Thanks porkluvr I will be working on that a bit later still have to work out the values for 32khz one step at a time,
as for the VCO I built it first and it works, as the sample audio file shows I only now had a think about how it works its actually a fixed oscillator in the second stage modulated by the first stage which is voltage controlled by the signal. there is a point in the tuning with the trimmer pot where the 2 signals BFO against each other and cause that distinctive sound. and change in amplitude is the same as the tesoro by the diode at the output of the vco , well thats my thinking anyway.

[dom44]

Thanks again porkluvr just tested the lm392 and you are right used a 22k pull-up I will update this post as I go.

[porkluvr]

Here I have used a transformer input and swept the receiver input over a 10kHz to 30kHz range. It doesn't really prove anything but it may be useful to show how strange the receiver circuit behaves (in its current form) without having a tuning capacitor across the rx coil.

I have included the necessary models and symbols so that anybody should be able to open this simulation and run it even though the MC33178 is not included with the (free) LTspice download.

Here I use MC33178 because it is the closest substitute for LS204 that I know of and I could not find LS204 spice model.

If anybody knows what I could do differently to make this simulation more meaningful, let me know and maybe I can accommodate.

[simonbaker]

Here I have used a transformer input and swept the receiver input over a 10kHz to 30kHz range. It doesn't really prove anything but it may be useful to show how strange the receiver circuit behaves (in its current form) without having a tuning capacitor across the rx coil.

I have included the necessary models and symbols so that anybody should be able to open this simulation and run it even though the MC33178 is not included with the (free) LTspice download.

Here I use MC33178 because it is the closest substitute for LS204 that I know of and I could not find LS204 spice model.

If anybody knows what I could do differently to make this simulation more meaningful, let me know and maybe I can accommodate.

For fun, I played with the LTSpice sim.

It seems that the response is a combination of perhaps increased coupling of the "transformer" with frequency, offset by the TX coil reactance and mild bandpass op amp filter (which peaks at about 14 kHz), resulting in a total frequency response that peaks at about 35 kHz. (Some coil resonance is included, and there is a little "tank" resonance with the coupling cap C1).

So as you increase your TX signal frequency, the response increases up to about 35 kHz. After that, it would decrease with frequency.

-SB

[Derx]

I have included the necessary models and symbols so that anybody should be able to open this simulation and run it even though the MC33178 is not included with the (free) LTspice download.

Here I use MC33178 because it is the closest substitute for LS204 that I know of and I could not find LS204 spice model.

All models http://www.gunthard-kraus.de/Spice_M...r%20ICs/Spice/

[Derx]

http://tech.groups.yahoo.com/group/L...spice%20Files/

[porkluvr]

Thank you for the list, Derx.

In a GUI driven spice simulator (such as LTspice) you need to have a symbol that has pins with labels corresponding to the appropriate nodes in the model. Many of the models from ST Microelectronics (such as the listing in post 24) have misleading notes that might cause an unwary user to make an error when creating a symbol.

Here is an example (taken from the LS204 model from ST Microelectronics):

** CONNECTIONS :
* 1 INVERTING INPUT
* 2 NON-INVERTING INPUT
* 3 OUTPUT
* 4 POSITIVE POWER SUPPLY
* 5 NEGATIVE POWER SUPPLY
.SUBCKT LS204 1 3 2 4 5 (analog)
#######################################

It would make more sense to read as follows:

** CONNECTIONS :
* 1 INVERTING INPUT
* 2 OUTPUT
* 3 NON-INVERTING INPUT
* 4 POSITIVE POWER SUPPLY
* 5 NEGATIVE POWER SUPPLY
.SUBCKT LS204 1 3 2 4 5 (analog)

Note that pins 2 and 3 have been reversed.
Beware, there other ST models that exhibit the same anomaly.