Barracuda Legend Kit Problems

[ripsdevala]

I dont know how to work with oscilloscope and other measure tools, I use only analog multimeter to measure, the last broblem with my barracuda is not far from wrong part, reversed fet,short component and dead IC after short.
Very poor sensitivity = wrong resistor value
Unadjust 0 volt from 5534 is because of short
Burn smell = 4001 roasted

When I use HEF instead of CD40106 its have better sens but decrease in sound.

The only dead IC when i built is 7660, Q says that 40106 is kaput when using rev.2 but still working normal with beep on every swing, still waiting for rev.3
Hope other senior member can answer your problem.

[Tim]

Hi. I wonder if anyone can help with my Silverdog Bara. I had it working somewhat. Coil 1.8 ohms to the board.330R damping. I had no luck with several(new) CD40106s and ended up with HEF40106. But very poor sensitivity. I also had to try 3 different 7660s before I got one that worked. (Also new). Voltage is regulated to 10v(78L10). Originally had 0 volts adjustable on 5534. Changed 5485/102 for matched 5485s-sense better. Threshold tone was intermittent, and seemed to be there only when case top, with pots and wires, was in some positions. I thought it may have been loose jumper connections, but it may be due to a capacitance issue(?). Then I fiddled with the damping, replacing 330R with sockets. While trying different ohms there with machine turned on, I had a power surge/spark, then no coil sensitivity whatsoever. Thought I may have blown the 9640 so changed that. No difference , so changed the 2222. No difference. Now I have a constant +4.46 at pin6 on the 5534, with no adjustment via pot.
It beeps on startup, but no sensitivity whatsoever.
At no stage has the sensitivity been better than an OZ 10c piece right on the coil (435uH). I have also tried 350uH, 260uH and 200uH coils with similar results (before it frazzled).
Now I am starting to think I may have blown a 4148 or other component.? Maybe the offset pot wiper?
I have just bought a Hantek 6022BE scope, but have no idea how to use it yet. Because all my info is on my one computer, I am petrified I might fry it mis-using the scope.

Any suggestions where to go next? Perhaps I should just buy/etch another board and start again........???
There is 560Hz at the coil. Correct voltages at the 78/7905s. Correct voltages at the 40106 (1,3,5,7,14).
So close, and yet so far....
I am using PN2222 oriented with silkscreen, and double-checked on datasheet, but perhaps that is still wrong?
Also I need a 100R inline with LED otherwise speaker(32R) was barely audible. Could this also be an indication of a problem somewhere??
Perhaps I should buy another brand/batch of 5534s?
All resistors were measured and diodes orientation double-checked during build.
Wima caps used.
I tried a TL062 for LF412 at one stage, and it seemed a clearer signal .

[Koala]

http://www.geotech1.com/forums/showthread.php?22964-Barracuda-Legend-Kit-Problems&p=215486#post215486

Info here

[wakko]

Wow, those timings are a bit different to what I've read. I'll see how I get on with the in field tests and go from there I guess.

[Koala]

Qiaozhi had to modify a board to stop the beep. Here is his data


Earth field sample pulse width = 120us
Earth field sample pulse delay = 440us

[wakko]

Thanks Koala,

I was refering to the delay between the main sample pulse and the earth field pulse to reduce the beeping at the ends of the swing. Most likely I'll have to adjust in it the real world for best earth field rejection.

[Koala]

Yep depth sounds right. Delay setting is a compromise. Just set for best depth on targets your interested in.

[wakko]

Could someone please tell me the correct delay to use between the main sample pulse and the earth field sample pulse? I've seen a few figures on the forums from 100uS, 150uS and 180uS. Maybe it's best to trim the delay to whatever works best in the field?

Currently I have:

TX pulse: 100us width, 640pps.
Main sample pulse: 45uS width
Earth field sample pulse: 45uS width
Delay between Main sample and earth field sample: 150uS

Air depth results using an 11inch coil...

Thin gold ring - 10 inches faint, 8 inches solid.
1p coin - 8 inches
10p coin - 10 inches
£1 coin - 12 inches
£2 coin - 14 inches

I think these results are acceptable.... any thoughts?

[porkluvr]

whoops, I intended this for one of the other Barracuda threads.

Polymer posted a '7660 Do's and Don'ts around here somewhere, but now I can't find it. That's what I was referring to in the drawing.

[ivconic]

Yes

Best I can get is only an extra 1/2" but is not worth it with the loss of depth on other targets

It is a tough target. My VLF with a 6" coil only tests at the same depth.

It was only a couple of inches down when I found it. Still only gave an iffy signal.(IDX 10" DD poor ground condition)


Yes heat sink is not needed. FET only gets slightly warm.

It could be also because of possibility that fibula was made from alloy with larger percentage of Cu.
Not "good" bronze but with higher percentage of Cu.
I noticed that Cu is tough target for some detectors.
There is (sort of speaking) "male" and "female" copper alloys!?
That's local slang formed around prejudice at local diggers.
So, one of those two (can't exactly recall which one) is very easy detectable with PI detector and another one is not.
Another one is almost impossible to detect it with PI detector.
I guess it's just variations in alloy compounds and relation between it's compounds.
But i noticed that certain Cu alloys are indeed very tough targets.

[Koala]

Precisely! Good job!
Heat sink is HUGE! No need for that.
"...Can get nearly 14" on the gold ring but then the fibula is down to 3"..."
Obviously.
Have you tried different delay for fibula?

Yes

Best I can get is only an extra 1/2" but is not worth it with the loss of depth on other targets

It is a tough target. My VLF with a 6" coil only tests at the same depth.

It was only a couple of inches down when I found it. Still only gave an iffy signal.(IDX 10" DD poor ground condition)


Yes heat sink is not needed. FET only gets slightly warm.

[Koala]

That sounds normal if the fibula is small, say 1/4" or so in diameter.
Do you have a US nickel? Hard to tell from your pictures. That is a target I know well.

No put the bottom right is a Canadian dim if that's the same and next to it is a Latvian.

[godigit1]

Hi Old cart,
Thanks for the reply. I had damped the coil for the 410 amp and just plugged in the 412 So it was not damped properly. I keep blowing pots for some reason. Ive been using the same set up but with a 1.2 k in parallel. But only 1/2 watt pots so maybe that's it.
The two stage amp I am using was for the hammerhead so caps and resistors need tweaking . It worked well on the HH and quieted it down too but the wave form was not right there either. There is a picture in the link above. With proper design it could be a nice adaptation to the Barracuda or maybe the surf pi as well I never tried it on mine but I think I will put the smp back together and give it a try just for fun.
I do have a copy of ITMD it has helped but I haven't been able to spend the lab time with it I need and it deserves.
Thanks again

[Old cart]

Hi Old cart,
It is KT315s board but I am using the 74hc14 p instead of the 40106 and a bc517 in place of the mpsa13. Also the 410 amp but it did work fine 5534 as well.

http://www.geotech1.com/forums/showthread.php?22775-Plug-in-dual-front-end-gt-gt

It is not optimized for the barracuda (Or The HH1 for that
Digit

I noticed nice clean waveforms from KT315 and he uses 74hc14 as well. Maybe they are a better match. Getting the coil damping optimized is important whether you have a one or two stage preamp. I noticed on the trace you published earlier there was significant overshoot at the preamp. A little overshot can be good. It might be better to use the damping adjusting tool to try and optimum
y damp the coil/ cable/preamp combo.
The good damping tool is just a 10k 1watt pot in series with a 200 ohm resistor. Tack solder it across the board at the coil connections. Starting with the two set to max resistance slowly decrease the setting while observing the preamp out and the coil voltage as well and use a sweep speed of maybe 10uS per div. adjust until the pre output shows a smooth transition and the smallest pulse width.
A very useful reference is the book Inside the metal detector.

[godigit1]

Hi Old cart,
It is KT315s board but I am using the 74hc14 p instead of the 40106 and a bc517 in place of the mpsa13. Also the 410 amp but it did work fine 5534 as well.

Hi Kt 315,
Your welcome the build went nice. I realize I'm a little off on the pulse rate and my second sample one thing I was going to try was a couple different Ics and also order some 40106s to try as well. Most of my Emi problems I'm sure are environmental as my Mini pulse is also badly affected. I'm close to a airport where I'm at now and on the scope there is a wave that pulses backwards on the trace.

Thank you all for the help. I need all I can get . You may of hit it with R 15 and 16 though as I could not get any 24 ohm resistors at the time so I have parallel resistors there. I now have 24 ohm resistors but 5 % and was thinking of trying them. Im also only getting 240 volts of fly back but assume its due to only using a 300 uh coil when the circuit was designed for 450 uh. Or my pulse on time may need adjusting a little too.

The Dual Front end was made for the Hammerhead 1 as a plug in and is under this thread.

http://www.geotech1.com/forums/showthread.php?22775-Plug-in-dual-front-end-gt-gt

It is not optimized for the barracuda (Or The HH1 for that matter it was copied from the HH2 Schem) but does seem to really help the threshold and Emi. I don't have the engineering skills to do it myself I can only copy things at this point in my learning but I have good attention to detail and a good pcb background. I think the gain could be increased to it and it could be good mod for the barracuda.
Thanks Again
Digit