BARACUDA-Legend-1

[gcause]

Hi All,

With Q5 in place I was getting a constant tone no matter how much I adjusted the offset / delay / threshold setup. These by the way are changing the tone coming out of the piezo going up and down in tone loudness and speed, so it appears the circuit itself is working to some extent.

However the tone never goes off completely, it doesnt even approach turning off the tone either.

I did find a problem on the first Baracuda board I had built I had a 270R in place of a 270K resistor and vice versa. I swapped these over and it still didnt work.

Removing Q5 didnt seem to help either, in fact the circuit stopped working altogether when it was removed. No tone no matter what I adjusted.

In the end I had to put a solder bridge across the three legs (posts) where it was in order to get the circuit to come to life again.

I tried both a commercial mono coil I have and a home built coil I have done following the instructions from the forums, this home made coil includes the shielding. Unfortunately I do not have the specs on the commercial coil.

I have now put both Baracuda kits I purchased together and both do not detect metal, well at least for me anyway.

I did add the SW1 to one of the PCB's and it did make a difference to the tone I got out of the piezo. It was still a constant tone though, it would never turn off or approach turning off.

I am a novice to electronics but have built other electronics kits succesfully in the recent past including etching my own PCB's and the regulated power supply I am using.

I have also tried running the circuits off a 9v battery and it made no difference.

On each test I have adjusted the voltage pin 6 on U2 and kept this monitored as I adjusted delay and offset values.

I have gone back over each board and checked all the components and they all now appear ok. As mentioned earlier I only found one pair of resistors that were wrong.

I must be doing something wrong but I dont know what it might be.

I have also purchased two Surf PI kits from Silverdog so I will now go off and try to build those and see if I have better luck.

Thank you everyone for your support and assistance it has been greatly appreciated.

Regards,

Grant

[porkluvr]

Hello Friends,

No problem I think (but I can be wrong..) the 40106 is connectet to 5volts, -5 on pin 7 and ground to pin 14 ..

You're right, no problem. My post was false alarm because I was stupidly looking at older schematic. Nice work on your PCB, by the way.

Hi ApBerg. Pin 14 of IC 40106 should be connected to "+5 V" and not "ground". It seems that the schematics of Alexismex not correct.

No. ApBerg and Alexismex are correct. I was looking at an OLD schematic copy - nicely printed but incorrect in many areas, the one posted by KT315. I should have been looking at the hand-drawn Alexismex (later) version. It does correctly show ground on U1-14.

Big mistake, my bad. Sorry, people. (Egg on me.)

[maikl]

Hi ApBerg. Pin 14 of IC 40106 should be connected to "+5 V" and not "ground". It seems that the schematics of Alexismex not correct.

[ApBerg]

Hi ApBerg,

Hope you are well.

The zip file seems to be missing some bitmap images, it only contains a LAY file.

At least when I load the layout viewer software it tells me there is a top board and bottom board bitmap image missing. I was hoping the layout file might show the component overvlay i.e. R35, Q5, SW1 etc.

Having trouble working out where SW1 and Q5 are located on the PCB.

I can see from this that it is one of the three MPF102's on the board.



Could you please advise where SW1 and Q5 are located on the PCB?

Thank you for your assistance with this it is greatly appreciated.

Regards,

Grant

Hello Grant,


The function from Q5+ switch and "his" parts is not clear... so do not use the switch
Best regards.

Ap

[ApBerg]

Hello Friends,

No problem I think (but I can be wrong..) the 40106 is connectet to 5volts, -5 on pin 7 and ground to pin 14 ..

[maikl]

You're right, Quiozhi. The Baracuda is fairly unique in that the primary sample is applied to the FET attached to the inverting input of the differential amplifier,
with the secondary sample being applied to the JFET at the non-inverting input.

On another note, I hate to make trouble but sometimes do it anyway...

I was examining at the Baracuda circuit and made an observation that makes me think the timing circuit is drawn incorrectly.
Alexismex's drawing shows that U1 (CD40106) is powered from Vdd and Vss, +5 and -5V. I don't have a factory unit to look at but from looking at the photos on
page 1 of this thread it appears that U4-14 (positive power to U4) is connected to ground, not Vdd.
That would mean that the sample pulses swing from -5V to ground.

I wired the tx and timing circuit up in LTspice and it shows that excursions from -5V to ground would be plenty to drive the transmitter.

In the matter of the sample FET drive circuits, I never did like how the FETs were shown being pulled to Vdd (thinking that it would be more appropriate for them
to be pulled 'high' to ground, not Vdd) but then I came to accept it (mistake though it may be) because if you apply some 'noise' voltage on both terminals of a
differential amplifier, the noise voltage should be rejected because it is common to both inputs, and the output should still be an accurate product of differential
voltage amplification. So, that could explain the current crop of Baracuda clones functioning well enough, even if the RE schematic has the error I've indicated.(?!)

Maybe also the Tune Reset switch wiring needs be examined, as I have noticed some complaints about it being dysfunctional.

Am I right, or am I wrong?

Hi porkluvr. Yes you're right here something is wrong.

[gcause]

Hi Diddier,

It is confusing as I have seen some pictures of boards with bridges and some without. Its good to know that Silverdogs PCB has the bridges built in.

I didnt want to put the remaining IC chips in until I knew which way to go as some of the bridges are underneath the top two IC chips.

Thank you for doing up the PDF based on the schematic I had guessed correctly on Q5, must be learning something at last

I will go off and take Q5 off the first Barracuda I built and see if that helps.

I will also go off an solder up the remaining components on the second Barracuda board.

Thank you for your efforts it is greatly appreciated.

Regards,

Grant

[didier]

answers to Q5 and switch 1

Hello gcause,

I do not have much experience with metal detectors but I think I can answer your questions. The red stripes on the silverdog board is the copper layer on the top side of the board. You do not need to solder additonal bridges, copper wirer.

For switch 1 and Q5 see attachment.

[gcause]

Also noticed a difference between the lay file and the Silverdog kit PCB.

There is a small red strap (these appear to be called a link on the layout file) next to the delay 25k pot on the Silverdog kit PCB that isnt in the layout file contained in the zip file.

[gcause]

Hello Friends,

Made a new pcb, I do not like those large standing caps...
In the zip is the sprint layout file, hope there are no errors...

Best regards.

Ap

Hi ApBerg,

Hope you are well.

The zip file seems to be missing some bitmap images, it only contains a LAY file.

At least when I load the layout viewer software it tells me there is a top board and bottom board bitmap image missing. I was hoping the layout file might show the component overvlay i.e. R35, Q5, SW1 etc.

Having trouble working out where SW1 and Q5 are located on the PCB.

I can see from this that it is one of the three MPF102's on the board.



Could you please advise where SW1 and Q5 are located on the PCB?

Thank you for your assistance with this it is greatly appreciated.

Regards,

Grant

[porkluvr]

Ok, all the places where I said U4, what I really meant was U1.

U1 is what looks like it the positive power is supplied from ground. (doh.)

[porkluvr]

To be honest, I've not paid a lot of attention to the Barracuda circuit. But now I look at the schematic more closely, I think you are correct. Can you measure the delay time for each sample? i.e. monitor the gate signal for Q3 and Q4. The secondary sample (Q4) needs to be made significantly later than the main (Q3) sample.

You're right, Quiozhi. The Baracuda is fairly unique in that the primary sample is applied to the FET attached to the inverting input of the differential amplifier,
with the secondary sample being applied to the JFET at the non-inverting input.

On another note, I hate to make trouble but sometimes do it anyway...

I was examining at the Baracuda circuit and made an observation that makes me think the timing circuit is drawn incorrectly.
Alexismex's drawing shows that U1 (CD40106) is powered from Vdd and Vss, +5 and -5V. I don't have a factory unit to look at but from looking at the photos on
page 1 of this thread it appears that U4-14 (positive power to U4) is connected to ground, not Vdd.
That would mean that the sample pulses swing from -5V to ground.

I wired the tx and timing circuit up in LTspice and it shows that excursions from -5V to ground would be plenty to drive the transmitter.

In the matter of the sample FET drive circuits, I never did like how the FETs were shown being pulled to Vdd (thinking that it would be more appropriate for them
to be pulled 'high' to ground, not Vdd) but then I came to accept it (mistake though it may be) because if you apply some 'noise' voltage on both terminals of a
differential amplifier, the noise voltage should be rejected because it is common to both inputs, and the output should still be an accurate product of differential
voltage amplification. So, that could explain the current crop of Baracuda clones functioning well enough, even if the RE schematic has the error I've indicated.(?!)

Maybe also the Tune Reset switch wiring needs be examined, as I have noticed some complaints about it being dysfunctional.

Am I right, or am I wrong?

[gcause]

Hi All,

Thank you for the replies.

I will go and try it without Q5 in place and report back.

What do you do with the dark red strips on the silverdog baracuda PCB?

I soldered metal straps on these using the offcuts from the legs of other components I had already soldered to the PCB, was that correct?

I have seen photos on the forum with metal straps in place and without. I noticed the ones that didnt have the metal straps in place still had solder in the solderpads of these dark red strips. I could only see the top side of the board in the photos so I assumed they must have soldered something between them on the underside of the PCB.

I bought a second baracuda kit from silverdog (in case I stuffed up on the first one) and have put it together also but I have held off putting the metal straps on like I did on the first one and havent put the chip holders and chips in until I know what I should do about the straps.

Also what is the pinpoint mode modification people keep talking about?

Thank you for your help it is greatly appreciated.

Regards,

Grant

[dfbowers]

Hello, I ordered a board (not the kit) from Silverdog for my Baracuda project and mounted all components . It is my first detektor I build up and therefore I do not have any experience. I made a preliminary coil with 26 turns of 0.5 mm copper wire and about 22 cm in diameter. The baracuda works.
But now it is very unclear for me how to proceed. I could not find a instruction how to adjust step by step the trimmers in this thread. How to begin ?

There is a trimmer for the balance, what is the criteria for this one ?

There is a delay trimmer, on the schematic there is a trimmer on the board (25K) and a external one 10K) but on the siverdog board I see only the 25 K trimmer. Is there a need to connect a external trimmer with 50k for the delay ? I could not see how to connect this on the board. What is the criteria for the delay ?

What works fine is the 50 K poti for the treshold.

The baracuda is sensitive for moved objects, means when I move a metal next to the coil it makes a sound. The Fet Q5 has no function. When I remove it I have the same performance. When I connect the gate of the fet to ground, the baracude does not work any longer even when I tune the treshold.

I would be very happy if anyone could help me to proceed.

Didier

You can adjust the d.c. offset pot for 0v on pin 6 on the ne5534. Other than that, you can set the delay as low as possible for best performance. I generally set mine at 25uS.

[didier]

Baracuda adjustment

Hello, I ordered a board (not the kit) from Silverdog for my Baracuda project and mounted all components . It is my first detektor I build up and therefore I do not have any experience. I made a preliminary coil with 26 turns of 0.5 mm copper wire and about 22 cm in diameter. The baracuda works.
But now it is very unclear for me how to proceed. I could not find a instruction how to adjust step by step the trimmers in this thread. How to begin ?

There is a trimmer for the balance, what is the criteria for this one ?

There is a delay trimmer, on the schematic there is a trimmer on the board (25K) and a external one 10K) but on the siverdog board I see only the 25 K trimmer. Is there a need to connect a external trimmer with 50k for the delay ? I could not see how to connect this on the board. What is the criteria for the delay ?

What works fine is the 50 K poti for the treshold.

The baracuda is sensitive for moved objects, means when I move a metal next to the coil it makes a sound. The Fet Q5 has no function. When I remove it I have the same performance. When I connect the gate of the fet to ground, the baracude does not work any longer even when I tune the treshold.

I would be very happy if anyone could help me to proceed.

Didier