Houston...I have a problem! (bandido coil)

[Qiaozhi]

Hi all,
anyone knows if bandido sounds with bricks/ceramics too sometimes ?

I had a response using disc mode and a piece of brick! And that shocked me !

I'm worried....

Qiaozhi, do you know if ceramics can be totally cancelled out by disc in Bandido ???

Maybe is just my circuit problem - not present in bought bandidos.

Best regards,
Max

Certain bricks seem to have some metallic content. In particular I noticed that blue engineering bricks can give a response. I also tested this with my Garrett GTAx1000, and this will detect blue bricks as well. So it's not a fault with the detector.

[Max]

...do it again...

Hi all,
anyone knows if bandido sounds with bricks/ceramics too sometimes ?

I had a response using disc mode and a piece of brick! And that shocked me !

I'm worried....

Qiaozhi, do you know if ceramics can be totally cancelled out by disc in Bandido ???

Maybe is just my circuit problem - not present in bought bandidos.

Best regards,
Max

[Carl-NC]

I now remember that I was using an HP LCR meter, and great big expensive bench unit. Now I have a smaller, cheaper BK 878 handheld LCR meter, which is much more stable. So maybe I need to remeasure. Or, just use Gary's values.

- Carl

[chemelec]

Hi Carl,
yes not -5v ...one must consider also that there is a drop on last transistor, ,the Vcesat voltage, that will be about 1V, capacitive losses...too low frequency in charge pump...etc so maximum -4Volts on negative rail as best.
I think that with right components it will be fine anything from -2.8 to
-4volts.
MC33178 works and is fully characterized at, according to datasheet, +2.5/-2.5 volts (and so also for MC33078 but much current required).

"So I make no guarantee they are right."

I know - it's not easy to reverse eng. sometimes. I know from other experiences , expecially if it was an smt assembly.
I found second set of values is working better than first, so I think could be ok.

Best regards,
Max

On my Revised design, I'm using the 7660 and the Minus rail is now -4.5 Volts.
But I also Agree that Even -3 volts is OK.

Yes, Carl said at the time that "He Did Remove the Caps to test them". He Also mentioned he was having a Bit of Problem getting "Stable Readings".
** Too Bad I don't have one of those Boards Here, So I could do some Very accurate Tests.
I have Three LCR Meters, One is a Sencore LC102, Capacitor/Inductor Analyzer.

I have now decided to put my Split board into a case with Proper controls and finish it off.
Maybe it will help me Solve some of the Problems mentioned here. (Trimpots are not easy to adjust when on both sides of the PCB and the board laying loose on the workbench.)

Take care....Gary

[Max]

As a reminder, Tesoro didn't design the -V chargepump for -5V. I think it puts out about -3V or so. This should be sufficient for the circuit, i.e., -5V really isn't needed. Most opamps will run happily on unbalanced supplies.

On the caps, all I remember is having trouble getting stable readings on the LCR meter. I think I had pulled the caps out-of-circuit. So I make no guarantee they are right.

- Carl

Hi Carl,
yes not -5v ...one must consider also that there is a drop on last transistor, ,the Vcesat voltage, that will be about 1V, capacitive losses...too low frequency in charge pump...etc so maximum -4Volts on negative rail as best.
I think that with right components it will be fine anything from -2.8 to
-4volts.
MC33178 works and is fully characterized at, according to datasheet, +2.5/-2.5 volts (and so also for MC33078 but much current required).

"So I make no guarantee they are right."

I know - it's not easy to reverse eng. sometimes. I know from other experiences , expecially if it was an smt assembly.
I found second set of values is working better than first, so I think could be ok.

Best regards,
Max

[Carl-NC]

To get -5V I made some corrections: ...

As a reminder, Tesoro didn't design the -V chargepump for -5V. I think it puts out about -3V or so. This should be sufficient for the circuit, i.e., -5V really isn't needed. Most opamps will run happily on unbalanced supplies.

On the caps, all I remember is having trouble getting stable readings on the LCR meter. I think I had pulled the caps out-of-circuit. So I make no guarantee they are right.

- Carl

[Max]

Hi Max

I too started to built Tesoro Bandido and visit the same problems how You (-V, poor discriminate). I use original Tesoro concentric coil for this detector.
To get -5V I made some corrections: for IC1 and transistor T3 (in some circuits Q4, 2N3906) I give power supply 9V, then I increase R67 value to 2K and in emitter of transistor T7 (Q5) I get -5V.
In next days I try to solve discrimate problems.Please, can You give me Yours capacitance values of next components:C16, C10, C14, C13 and C18.
Thanks.

Hi Satel,
first I've used same Carl's values as in original schematic. Things were not so bad but I don't know if they were right.

Now I've substituted with these (as in Gary's revised schematic):

C16 : 39pF
C10 : 22pF
C14 : 8.2 pF
C13 : 10pF
C18 : 22pF

They are from Carl too, but measured OUT OF CIRCUIT. I think they are right.

And seems things are going well. No major differences in All-metal mode (just a bit faster response I think), but seems that in disc mode they work better than previous values.
I haven't now double pulsing or too distortion but sometimes it detects iron masses too.

"I give power supply 9V, "
I was thinking too using 9v line but I give up on this because it's not stabilized. When battery discharge you have voltage variations on V- rail that I think it's not good. Best to get mc33178 if you can; I'm still awaiting for delivery of these parts...not so easy.
Think that in worse case, one could use also a 12volt battery and an LDO regulator for 9v line.

Best regards,
Max

[Satel]

Hi Max

I too started to built Tesoro Bandido and visit the same problems how You (-V, poor discriminate). I use original Tesoro concentric coil for this detector.
To get -5V I made some corrections: for IC1 and transistor T3 (in some circuits Q4, 2N3906) I give power supply 9V, then I increase R67 value to 2K and in emitter of transistor T7 (Q5) I get -5V.
In next days I try to solve discrimate problems.Please, can You give me Yours capacitance values of next components:C16, C10, C14, C13 and C18.
Thanks.

[Max]

Which caps do you think are incorrect?

Hi Qiaozhi,

c9,c10,c14,c13,c16

They are all small pF values and I think there were problems in measures.
Can't say exactly what could be good values. Don't know what phase differences I have to expect exactly (is -37deg to +50deg correct at pin2 ic5 (goodrat values) ? because I think I haven't turning gnd pot.)

Best regards,
Max

[Max]

Those Values were given to me by "Carl Moreland" in an email to me.
He said he measured them Out of Circuit.

On the Other hand "Goodrat" emailed me and Said he measured a Tesordo Detector.
C10, 13, 14 and 18 were all 5pF.
C16 and C29 were 10nF.

So take your choice.

Hi Gary,
many thanks for the interest.
"Those Values were given to me by "Carl Moreland" in an email to me."
do you mean the values in your revised schematic(s) ? Why they are different from original Carl's schematic ? Maybe he can't measure them well first time due to circuit assembly or something ?

Goodrat values are different too. There are big differences !
Why ?

Is the original pcb using SMT ?

Gary, I've found another little mistake on my circuit:
when disc pot, or gnd pot or both are close to zero ohms (extrem) and big load is on tx signal from oscillator I have a peak-doubling...means that too load on oscillator cause signal distortion visible also after analogue gates.
I think I can solve increasing TX osc. current modifing , so reducing, R1 from 1.2K to 1K or something lower.

What do you think about?

Best regards,
Max

[chemelec]

Which caps do you think are incorrect?

Those Values were given to me by "Carl Moreland" in an email to me.
He said he measured them Out of Circuit.

On the Other hand "Goodrat" emailed me and Said he measured a Tesordo Detector.
C10, 13, 14 and 18 were all 5pF.
C16 and C29 were 10nF.

So take your choice.

[Qiaozhi]

Hi Qiaozhi,
I've tested with half crown of 1967 (queen Elizabeth II) and gives distance of
more than 35cm.

I can test too with other coins...thanks for the infos.
I have to solve too the slight iron detection. I think caps are involved.
I'm using Carl's vaues at now with tollerances of less than 2%.

Do you think I have to use Gary's values ?

Best regards,
Max

Which caps do you think are incorrect?

[Max]

Victorian Penny = 30mm
Liberty Dollar = 24mm
Dime = 18mm

Hi Qiaozhi,
I've tested with half crown of 1967 (queen Elizabeth II) and gives distance of
more than 35cm.

I can test too with other coins...thanks for the infos.
I have to solve too the slight iron detection. I think caps are involved.
I'm using Carl's vaues at now with tollerances of less than 2%.

Do you think I have to use Gary's values ?

Best regards,
Max

[Max]

MAX, That 50 Hz sounds like a Line Frequency Being Picked up "by the COIL".

Gary

Hi Gary,
yes I was thinking too that coils picks up too much noise...but I disconnected RX...then measured noise voltages at inputs of IC8 and get same noise levels (before shielding circuit). So I realized that small noise variations were in the circuit , starting at c20/c21.
To prove that I've used my fingertip:
touching c20 I can see increase in noise in its channel ; touching c21 the same for the other channel.

Concluded that small noise asymm creates huge differential at IC8 inputs.

Then shielded and solved the problem.

Maybe coils gets too big noise...I think so. But it's nulled very well...then if the noise source is far away noise is nulled too.

Best regards,
Max

[Max]

Hi Ivconic . Why to make the bandido. It is about 90% the same with CL3. It is a little better at discrimination and cl3 is a little better at sensitivity.
Total weight .....the same
I say this because you are professional at CL3 construct.
Regards

Hi Geo,
I know CL3...used a bought one sometimes in past...not so good with disc...I think that bandido is better (when right tuned in disc mode). Performances seems to me even better (in air) than CL3.

Anyway, we do these circuits also for curiosity...I think.
I have GS4-proto and goes well...very deep...but doesn't really disc...just ignore ground mineralization. It sounds with bricks...so I decided to make a try with (unknown for me) bandido II, as Reg suggested.
I do not expect miracles here...from vlf, but circuit is really good from an engineering point of view except maybe the voltage converter.

Best regards,
Max