Surf 1.2 PI help required please

[Stu]

Hi folks.

I built a Surf 1.2 a while ago for a friend, it was working well but I thought I could get more performance from it.
It was detecting a slim white gold wedding band at 8 1/2" and a £1 coin at 13" in air indoors, which I thought was very respectable.

Using a coil: Ø 300mm/12" L= 321.1 uH C= 63.43 uF R= 5.629Ω made from 22 swg enamel copper wire bundle, which was tied then wrapped with a thin coat of insulation tape, then spiral wrap, & finally notched aluminium foil leaving a gap of approx 6-8mm.

I was attempting to tune the damping resistor value using a 1K2Ω in parallel with a 4.7kΩ variable + 250Ω
For some reason the detector has stopped responding to targets and just sounds like its nulling / ringing all the time.
I'm afraid I've toasted or damaged a component but I don't know which one or how to tie down the problem.
I've tried several coils that have worked with the circuit. I've tried varying pulse delays from 8uS to 60uS but nothing seems to work.
The offset is bang on 0v
I've got a DMM & a PC based oscilloscope (I'm learning to use it though).
I also have a Megger B131 LCR meter at my disposal.

Below is a screen shot of the scope without a coil connected to the board:
[IMG][/IMG]

Below again is a screen shot of the scope with the afore mentioned coil connected:
[IMG][/IMG]

As mentioned before I'm very much beginner when it comes to using a scope, the same is true with electronics as a whole.

Any help would be very much appreciated as I only have a very basic grasp of the workings of a Pi machine.

[Carl-NC]

What is Ch1?

[Stu]

Hi Carl.
CH 1 = Testpoint 1 Pulse delay ?
CH 2 = Testpoint 2 ???
The other part of the oscilloscope probes was grounded to the board, not sure if that was correct?
Sorry for any confusion caused.

Just in case I was wrong I've attached a screen shot of the probes attached with no ground to the surf pi board.

[Waikiki_Sweep]

Did you checked +5 V and -5 V power supply? pin 8,4 of LM358.

Send +12 volts on the Base of transistor Q6 with using 10 K resistor. Can you hear permanent loud sound? If yes it is OK.

Also please check integrator output voltage - pin 7 of LM358. Most common failure of Surf PI is overload of integrator so op amp is saturated at high position 3.75 volts or so and detector stops responding on targets.

I can tell that your coil looks discharging much longer than pulse delay adjusted. If you telling it was working before with same setting so something has broken.
I suspect 4.7kΩ variable has been fried by high voltage. Check 4.7kΩ variable with multimeter.
You can try to remove all what you build for coil dumping and replace it with 390 Ohm 2W metalfilm.
If detector start working use variable only once to find proper dumping value and place quality 2W permanent resistor for dumping.

[Stu]

Hi & thanks for the information on how to fault find.
My results are as follows:
I replaced the Resistor with a new 2W 390R metal film type.
LM358 (U3 on the silkscreen) Pin 8 = +5.06V Pin 4 = -4.76v

Voltage = 10.70v (8x Sony AA Dry cells)
Q6 with the 10v battery pack +10K resistor gave a loud tone through headphones (12V supply buried in the shed somewhere)?

The detector seems to be working now & ready to be set up again. I think the root cause may have been the pulse delay trimmer, it may have stuck low from constant fiddling? I have changed it now for a new 100K pot to be box mounted.

Thank you very much indeed once again for your suggestions on how to fault find the circuit I hope this thread helps other people building this impressive project.

[Waikiki_Sweep]

Thanks. Also I can recommend you to use faster coil. Faster coils gives you more chance to find gold because gold decays fast.

According your scope screenshots your coil discharging 30 microseconds or more. It is too much. Take a look on lower flat line on CH2.

Surf PI was designed to work much faster starting 15 microsecond and PI PRO starting from 5 microsecond.
Every kilo-ohm of resistance in pulse delay is equal for one microsecond of delay.

You can see on different schematics a combination permanent resistor and pot:

15k + 10k pot for Surf PI
5.1k + 10k pot for Surf PI PRO
15k + 100k pot for some versions of very bad design.

I think biggest problem of your coil is an aluminum foil shield. That foil keeps eddy currents longer than probably many small gold targets you would like to find.
Use graphite for shield.
Also cooper conductor in the coil keeps eddy currents too. Detecting itself.
Use insulated stranded wires.
Put Teflon or polyethylene spacer between shield and coil to reduce coil - shield capacitance.
Use shorter cable.
Check that your coaxial cable is made with stranded core and Mylar shield or stranded mesh shield but NOT FOIL. Because your detector will detect your cable.

Rule is simple. Build your coil and cable only from materials that you can not detect with PI detector. (graphite, Mylar, stranded wires)

Also MOSFET with low Coss will help or use HER208 fast diode in series with MOSFET to reduce internal capacitance.

Once you will see that lower flat line CH2 on scope gets shorter than 15 microsecond (5 microsecond in ideal) than you have build really great coil and ready to find gold:
Big gold chains, small gold rings, some moderate size nuggets.

[Stu]

Thanks once again for all your help & I'm sure this thread will help others too.
When I first looked at this site I noticed that projects are catagorised but I couldn't find one place with all the Surf 1.2 PI info.
Great site all keep up the excellent work

[Qiaozhi]

Thanks once again for all your help & I'm sure this thread will help others too.
When I first looked at this site I noticed that projects are catagorised but I couldn't find one place with all the Surf 1.2 PI info.
Great site all keep up the excellent work

If anyone has the time to put together a complete set of information on the Surf-PI (including debug info), I will create a dedicated area (with a sticky) in the Projects section. Have a look at the Baracuda PI section as an example.

[Waikiki_Sweep]

Everything about Surf PI is here:

http://www.geotech1.com/forums/showt...f-PI-schematic
http://www.geotech1.com/forums/showt...55-Surf-PI-Pro
http://www.geotech1.com/forums/showt...ock-NEW-Prices

[amousetoi]

Thanks. Also I can recommend you to use faster coil. Faster coils gives you more chance to find gold because gold decays fast.

According your scope screenshots your coil discharging 30 microseconds or more. It is too much. Take a look on lower flat line on CH2.

Surf PI was designed to work much faster starting 15 microsecond and PI PRO starting from 5 microsecond.
Every kilo-ohm of resistance in pulse delay is equal for one microsecond of delay.

You can see on different schematics a combination permanent resistor and pot:

15k + 10k pot for Surf PI
5.1k + 10k pot for Surf PI PRO
15k + 100k pot for some versions of very bad design.

I think biggest problem of your coil is an aluminum foil shield. That foil keeps eddy currents longer than probably many small gold targets you would like to find.
Use graphite for shield.
Also cooper conductor in the coil keeps eddy currents too. Detecting itself.
Use insulated stranded wires.
Put Teflon or polyethylene spacer between shield and coil to reduce coil - shield capacitance.
Use shorter cable.
Check that your coaxial cable is made with stranded core and Mylar shield or stranded mesh shield but NOT FOIL. Because your detector will detect your cable.

Rule is simple. Build your coil and cable only from materials that you can not detect with PI detector. (graphite, Mylar, stranded wires)

Also MOSFET with low Coss will help or use HER208 fast diode in series with MOSFET to reduce internal capacitance.

Once you will see that lower flat line CH2 on scope gets shorter than 15 microsecond (5 microsecond in ideal) than you have build really great coil and ready to find gold:
Big gold chains, small gold rings, some moderate size nuggets.

Hello .. I am very interested in your answer, but with the translator and my lack of competence of electronics I do not 'clear -


(You can see on different schematics in combination permanent resistor and pot: 15k + 10k pot for Surf PI) what are the resistors to replace?
thanks

[Waikiki_Sweep]

Please check that schematics for Pulse Delay is same for many clones of Surf PI but resistors values are different.



Smaller value gives shorter delay and more sensitivity to small gold target but not every coil can work as fast as schematic required.
Barracuda and Surf PI PRO have 5.1 k + 10 k resistors so they designed for delay from 5 to 15 microsecond because for time constant formula
if you using 0.001 microfarad capacitor with 1 k resistor you will have 1 microsecond RC time constant. As well as 15 k will give you 15 microsecond of delay.
But if you made bad coil that discharging in 20 microsecond or more (solid wires, foil shield, no insulation, long cable, e,t.c.) it will not work in that schematic at all.
So you will need 50-100 k pot to move sample into zone where your coil works - 30-40-50 microsecond pulse delay. And forget about sensitivity to small gold items.