Barracuda Legend Kit Problems

[Impulse]

PI coils tend to be in the range 300uH to 500uH, whereas VLF detector normally go from 1mH upwards. Also, PI coils tend to use thicker wire than VLFs.
You would be better off buying a PI mono coil such as a Garrett Seahunter MK2, which are relatively inexpensive.
http://regton.com/10-x-14-elliptical...r-mark-ii.html
http://regton.com/garrett-8-coil-seahunter-mk2.html

yep i am seeing my mistake thank you Qiaozhi i do apologise for the lack of knowledge or forward thinking, i saw the coil on offer and thought i would ask/before i do, the guys in the know as i am new to the arena.

[raygdunn]

Unwanted voltage spikes in 40106

Hi,

I've finally cleared this problem of unwanted voltage spikes. This is what it was like and finally I have . Swap the chip and the problem goes away, except the first time I tried, it gradually came back! It could be caused by setting the delay trim resistance too close to zero, or it could just be me bridging pins with a probe! May be even residual power because I didn't wait long enough after switching off. A small fixed resistor in series with the Delay pot. does seem a good idea.

Sadly this doesn't seem to improve the coil movement problem any. I'll try adding those extra capacitors when they arrive and work on a more rigid standard 10" coil.

I've added to my list of identified room interference sources . An old 32" analogue TV at 2 metres. Rotating the coil to vertical avoided most of this.

Ray

[Old cart]

Hi,

I've finally cleared this problem of unwanted voltage spikes. This is what it was like [ATTACH]35502[/ATTACH] and finally I have [ATTACH]35503[/ATTACH]. Swap the chip and the problem goes away, except the first time I tried, it gradually came back! It could be caused by setting the delay trim resistance too close to zero, or it could just be me bridging pins with a probe! May be even residual power because I didn't wait long enough after switching off. A small fixed resistor in series with the Delay pot. does seem a good idea.

Sadly this doesn't seem to improve the coil movement problem any. I'll try adding those extra capacitors when they arrive and work on a more rigid standard 10" coil.

I've added to my list of identified room interference sources . An old 32" analogue TV at 2 metres. Rotating the coil to vertical avoided most of this.

Ray

Ray now that is a proper sample pulse. Check the second pulse ( the earth filed pulse ), about 100 uS later too. It is good to hear it was just a bad 40106. I wonder if a lot of these are counterfeit or just very sensitive to (static) damage. This is common with CMOS parts. I would suggest buying parts only from reputable manufacturers and touching a grounded metal plate or other point with your hand before handling these parts and soldering them in. Static straps are also a good inexpensive solution.

https://en.m.wikipedia.org/wiki/Elec...nsitive_device

Impulse you might try changing this part with a brand name part. Ray has set his sample pulse just about perfectly.

Hopefully stabilizing the coil will eliminate the falsing. If it does not it may be interference, inaccurate earth field cancellation, or other system noise like a noisy 5534.

Keep us informed.

[ODM]

If I recall correctly the Barracuda timing circuit relies on timing capacitor discharge after rising edge from the driving stage. Unfortunately there's no path to discharge that timing capacitor after a falling edge except for the IC's static charge protection diode from VSS to input pin. This is something of a serious flaw in my book, though some manufacturers' 4000B and later series will withstand abuse like that better than others, with more extensive protection networks (multiple diodes/resistors/etc). The drive current is limited in a 4000 series logic IC which may at least hit the protection diodes less hard even if it's a repeated condition. There are many Barracudas that work just fine so some parts must be good for such "cheapskate" applications.

Driving any CMOS part out of its supply voltage range is just begging for latchup. Adding small signal schottky diodes, series resistors, etc for protection shouldn't be much to ask - then again that would rob the design of its attractiveness in simplicity

[FatBob]

But Bara's 40106 works off 5V. This is out of its supply voltage range?

And how would you change the schematic to cure this?



Oops on some schematics it shows a 10V supply which makes more sense as
the RC's are tied to +5V but still 10V is OK for 40106.

I drew my schematic from this original traced one found online;



It shows a 5V supply for the 40106 but some of the timing resistors tie to 5V above
the positive pin!

My Silverdog / ApBerg board is built like this and I have a picture of the original it is the same.
With a gnd to -5V supply how can you use +5V for the timing?

[kt315]

guys, i again must notice for you - original Bara works fine.
all problems you tell about are ONLY YOUR problems. NOT of the original schematic and of the original PCB.
you do something wrong and people are thinking that Bara has problems. NOT!
i especially copied original PCB and did see no one problem.

[jladre]

kt315 is right! There's nothing wrong in the original schematic and pcb.. I have build 5 barracuda on that schematic and all working fine. So the mistake her is in you only.

[kt315]

It shows a 5V supply for the 40106 but some of the timing resistors tie to 5V above
the positive pin!
--
yup. and what??? you want to say that is wrong?

[kt315]

important issue.

I've noticed a few people here are having problems with the 7660.
You need be aware that there are two versions of this IC - the 7660 and the 7660A.
The 7660 performs supply voltage conversions from positive to negative for the input range of +1.5V to +10V, and the 7660A has an input range of +1.5V to +12V.
The maximum supply voltages are also:
7660 = +10.5V
7660A = +13.0V

[kt315]

important issue.

7660CPAZ = 10 V
7660SCPAZ = 12 V

[FatBob]

If your timing resistor ties to +5V you will be outside the chips reference of -5 to 0.

CMOS chips do not like that.

So maybe that's why the latchup would occur?

[kt315]

If your timing resistor ties to +5V you will be outside the chips reference of -5 to 0.

CMOS chips do not like that.

So maybe that's why the latchup would occur?

guys, i again must notice for you - original Bara works fine.
all problems you tell about are ONLY YOUR problems. NOT of the original schematic and of the original PCB.
you do something wrong and people are thinking that Bara has problems. NOT!
i especially copied original PCB and did see no one problem.

[FatBob]

I agree it is per the original board and assume the original works fine (as they were sold for a long time).

I just wonder if the 40106 should be powered off +-5V instead of Gnd 5V.

Here is the Whites 1000 which also uses a CMOS inverter and it is powered from +- 5V;



These CMOS inverters are sometimes used in unorthodox ways as they are an array of FETs internally.
I was hoping an engineer would step in and say which way was more correct.



This is how our circuit would look;


The data sheet says input can be Gnd to Vdd (5V in this case), it doesn't say what happens when we exceed that....

[kt315]

max rating for cd4000b series is 20v. in 40106 -internal.gif the input connected NOT directly to +10v
but via a resistor. sometimes people do not want see evident things.

[FatBob]

Well I do not wish to argue here. I just want to say if you put a resistor to 10V between 2 caps the voltage (between the caps) will rise to 10v.



The power pin of the 40106 must be at the highest potential of the circuit (IE 20v , +5V).

I believe this is the way it should have been done;