Help Setting up Surf 1.2 PI

[apulmax]

Hi,
I have been following lots of threads on this forum the last few weeks on the surf 1.2 PI Kit. I decided to buy a kit the other day and have completed the PCB. I couldn't find anywhere on what coil to make so I went for the coil in the Hammerhead Pulse Induction Metal Detector PDF. I formed 26 turns of 0.2mm hook up wire which I thought was a good starting point, powered everything up and adjusted the tuning pot. I get a faint buzz going in and out and when I pass a 2p coin over the coil it will detect it but only from about an 1". A gold ring has to touch the coil to detect it.

I know a average bit about electronics to build this and set it up if I had clear instructions but the kit didn't come with any.

I have a scope but never had a need to use it. I would appreciate very much if someone could talk me through the setting up with or without a scope and if I the coil I made was suitable.

[Lord Dracon]

I wrote to Silverdog some time back and he gave me the following advice regarding coils for the surf 1.2 PI i hope this helps:


Look at the following threads http://www.geotech1.com/forums/showthread.php?t=17121
and here http://www.geotech1.com/forums/showt...t=17240&page=2

There is some info about wire and turns in one of the threads - basically you need approx 300uH inductance
I would think 24swg will be ok, no thinner or resistance gets too high
In one of the threads you will see ApBerg using 7 strand equipment wire, nice and cheap - I have tried it and similar performance to copper magnet wire but cheaper
Coil does not have to be shielded
USB2 cable works well for the cable from coil to pcb


regards and good luck


i hope this helps you improve your return signal

i have noticed most people use 22 swg enamel coated copper wire in a regular bundle coil configuration.


Good luck with getting it working as you want it to

[apulmax]

Thanks for the reply. It seems the detector is working ok, probably just the coil that's causing problems. I made another coil out of 0.5mm tri-rated multi-strand wire. I wound it around the nail former and measured the resistance at 15 turns and so on, kept winding and measuring until I got 1.6 ohms, which was 36 turns. Results even worse. I am wondering if its to do with the thicker insulation?

How do you measure inductance of a coil? doesn't it need to be in circuit and with a scope?
anyone recommend a ready built coil that would work? Would save so much hassle

[SVEN1]

Thanks for the reply. It seems the detector is working ok, probably just the coil that's causing problems. I made another coil out of 0.5mm tri-rated multi-strand wire. I wound it around the nail former and measured the resistance at 15 turns and so on, kept winding and measuring until I got 1.6 ohms, which was 36 turns. Results even worse. I am wondering if its to do with the thicker insulation?

How do you measure inductance of a coil? doesn't it need to be in circuit and with a scope?
anyone recommend a ready built coil that would work? Would save so much hassle

I bought one of these
http://www.ebay.com/itm/NEW-LC100-A-...item3f0f54f5eb

[apulmax]

They don't have any of those on ebay in the Uk but do have this one that should work http://www.ebay.co.uk/itm/LCR-INDUCT...item5638cdd9f7

[Ism]

They don't have any of those on ebay in the Uk but do have this one that should work http://www.ebay.co.uk/itm/LCR-INDUCT...item5638cdd9f7

I have one of those, it gave me bad readings. May just have been my particular unit.
Ended up getting a better LCR meter but more expensive

[UFO]

I'll tell you how to tune that puppy.

You need to make a variable dampening pot to test different resistors. This will make a HUGE difference.
(The following assumes you have set the voltage to zero on that opamp and have the delay reasonably tuned to the metal(s) you desire.)

I got the idea from Qiazho, this is what he said in a thread:
"I use a 10k pot with a series 270 ohm resistor, both in parallel with a 1k2 resistor, which provides an adjustment range between 220 ohms and 1k1. Then you don't end up frying the mosfet. "

You can use slightly different values if needed. Just use an online calculator if your not familiar with parallel resistances etc. My range is very similar to the one Qiazho built, but I used a 5K pot.

Basically when the pot is turned all the way to the left (220 ohms) my detector rings constant, I turn it a bit higher, it starts to work, I turn a bit more and the range gets deeper, I turn more, depth disappears, so I turn back a bit.
Now that you have it dialed in, disconnect the pot and measure the resistance, this value will be your new dampening resistor.(Or something close)

Essentially I have found that the stock 330ohm that came with my kit is not suitable for the particular coils I built.

Each individual coil may have an optimum dampening resistor different then other coils.

[UFO]

Don't worry so much about measuring inductance right now, as those DMM meters are only so accurate.

Basically that "coil calculator" is pretty accurate. The calculator assumes enamel magnet type wire.
I use plastic insulated wire, I find that my coils end up about ~10%-15% lower then the calculator. (As you may no the thicker insulation makes a coil with less capacitance.)

So when I use the calculator, I just calculate a coil with 10% more inductance then I need. Works good enough for this application.

Of course go ahead and get an inductance meter, as it would be handy, I'm just saying that you don't "need" one to make awesome coils.

I don't have a meter, but I was able to half-azz measure inductance by feeding a 10kHz sound from my computer ran through my stereo into the coil with a pot attached to it.
Basically you turn the pot until the voltage is the same across the pot and the coil, then just some quick math. If you are interested I can fink the thread and post it here.

[Lord Dracon]

Thanks for the reply. It seems the detector is working ok, probably just the coil that's causing problems. I made another coil out of 0.5mm tri-rated multi-strand wire. I wound it around the nail former and measured the resistance at 15 turns and so on, kept winding and measuring until I got 1.6 ohms, which was 36 turns. Results even worse. I am wondering if its to do with the thicker insulation?

How do you measure inductance of a coil? doesn't it need to be in circuit and with a scope?
anyone recommend a ready built coil that would work? Would save so much hassle

I assume that you are building a bundle coil and not a flat or planar coil.

Whats the inner diameter of your coil?

36 turns??? damn, no wonder it don't work very well. the inductance in the coil is gonna be huge, which, in turn will cause the induced decay curve to be out of sync with the delay on the measurement points of the circuit... so no wonder it don't work very well

if its a bundle coil of 36 turns, 0.5mm wire with a internal diameter of 200mm your inductance is gonna be like 1670uH.. the circuit is designed for around 300uH !!!

try this .. i suppose you still got some .5mm copper wire. (the thinner copper wire gives much greater resistance and slows down the coil feedback decay curve, yet multi-strand does help reduce the resistance (resistors in parallel) ) ..(a better wire is .62mm (22swg) single strand as this has lower resistance than the thinner wire.. However, i was informed 24swg (.56mm) is about the thinnest you can go with this circuit (and 22swg the thickest) without having to get an oscilloscope to test the decay curves and altering some of the resistors in the circuit - namely R6 to balance the coil....but anyhow i'm digressing)

the resistance in the coil needs to be as small as possible, but not so small as to overload the i/c's. so what you need to do is to use a coil calculator:

http://www.geotech1.com/forums/showthread.php?t=17909

which works very well to calculate as near as possible to the 300uH inductance you need (but should be ok if your coil is between 250 and 300uH).. then once you have this you may need to add a resistor in series with the coil to get to the 1.6ohms you are looking for. but some people don't really bother with this .. only if ic's start to get warm on the circuit board.

if your using .5mm i would suggest trying the following coils for the surf 1.2 PI

inner radius - 280mm ...................or................ inner radius - 215mm
wire thickness - 0.5mm ................./.................. wire thickness - 0.5mm
Turns - 12 ................................./.................. Turns - 14
Inductance - 303uH ...................../.................. Inductance - 300uH

but a better coil for this would be (as it has a lower resistance and therefore is faster acting)

inner radius - 200mm
wire thickness - 0.62mm single strand
Turns - 15
Inductance - 302uH


Once you have your coil made measure its resistance and if needed add a resistor in series if the resistance is too low.




Also don't forget, the detector requires motion to detect, so the object has to be moving for the unit to detect it! ... also the detector is susceptible to outside interference (electrical) so your coil may need to be shielded, but this depends on the environment your working in.

here is a link to some help i was getting when i was gonna build my surf 1.2 .. i hope this extra info will help (you will also find other help on this site about the surf 1.2 PI)

http://www.thunting.com/smf/metal_de...t23200.25.html

Good luck .. i hope you start to get the results you are looking for.

[apulmax]

Thanks for all the advice.
I have ordered some silver plated PTFE INSULATED EQUIPMENT WIRE, 7/0.15 A BLACK 25M
I will wind this around a 10" former with 21 turns as suggested.
As not to fry anything I am using a bench PSU with the current turned down.
How do I set up the trimmers on the board? I guess I need to have these set right before attempting to play around with the coil?

[UFO]

Yes .5mm is large, you can't make a coil without a resistor as mentioned by the poster above. Any coil I attempted to make was too low of resistance for the desired inductance range.

You should be able to find lots of instructions for the two onboard trimmers, but basically your adjusting one so that Pin 6 of the opamp (5534) reads 0v. Easy enough.

The other one is the delay. Basically if you had a piece of gold, say a ring, you would keep advancing the delay until the ring is no longer detectable, then back a bit. You will have to figure out which way to turn it to advance the delay. If you were turning it the wrong way, iron and stuff would be detectable but gold will not. Easy enough to figure out, maybe someone will tell you which way to turn.

As a general guideline, any half decent coil is going to detect a 14k gold ring to a depth equal to diameter of the coil. Any more would be a bonus. (Of coarse they will detect larger things much deeper then the coil diameter.)

[apulmax]

I assume that you are building a bundle coil and not a flat or planar coil.

Whats the inner diameter of your coil?

36 turns??? damn, no wonder it don't work very well. the inductance in the coil is gonna be huge, which, in turn will cause the induced decay curve to be out of sync with the delay on the measurement points of the circuit... so no wonder it don't work very well

if its a bundle coil of 36 turns, 0.5mm wire with a internal diameter of 200mm your inductance is gonna be like 1670uH.. the circuit is designed for around 300uH !!!

try this .. i suppose you still got some .5mm copper wire. (the thinner copper wire gives much greater resistance and slows down the coil feedback decay curve, yet multi-strand does help reduce the resistance (resistors in parallel) ) ..(a better wire is .62mm (22swg) single strand as this has lower resistance than the thinner wire.. However, i was informed 24swg (.56mm) is about the thinnest you can go with this circuit (and 22swg the thickest) without having to get an oscilloscope to test the decay curves and altering some of the resistors in the circuit - namely R6 to balance the coil....but anyhow i'm digressing)

the resistance in the coil needs to be as small as possible, but not so small as to overload the i/c's. so what you need to do is to use a coil calculator:

http://www.geotech1.com/forums/showthread.php?t=17909

which works very well to calculate as near as possible to the 300uH inductance you need (but should be ok if your coil is between 250 and 300uH).. then once you have this you may need to add a resistor in series with the coil to get to the 1.6ohms you are looking for. but some people don't really bother with this .. only if ic's start to get warm on the circuit board.

if your using .5mm i would suggest trying the following coils for the surf 1.2 PI

inner radius - 280mm ...................or................ inner radius - 215mm
wire thickness - 0.5mm ................./.................. wire thickness - 0.5mm
Turns - 12 ................................./.................. Turns - 14
Inductance - 303uH ...................../.................. Inductance - 300uH

but a better coil for this would be (as it has a lower resistance and therefore is faster acting)

inner radius - 200mm
wire thickness - 0.62mm single strand
Turns - 15
Inductance - 302uH


Once you have your coil made measure its resistance and if needed add a resistor in series if the resistance is too low.




Also don't forget, the detector requires motion to detect, so the object has to be moving for the unit to detect it! ... also the detector is susceptible to outside interference (electrical) so your coil may need to be shielded, but this depends on the environment your working in.

here is a link to some help i was getting when i was gonna build my surf 1.2 .. i hope this extra info will help (you will also find other help on this site about the surf 1.2 PI)

http://www.thunting.com/smf/metal_de...t23200.25.html

Good luck .. i hope you start to get the results you are looking for.

Thanks for all the info, I will try and digest it all before continuing.
The coils I made were bundle coils with a 250mm internal diameter.

I have ordered 25m of teflon silver plated 7 strand 0.15 and will give that a go.
The multi-strand .5mm has a 400v rating and has a thick pvc sheathing which I understand is not good capacitance wise.

I still have the .5mm 36 turn coil so I can remove some turns to get to 12 turns, would 12 turns work on 250mm?

[UFO]

Your welcome.

[arkaba]

Hi Apulmax

Like you I am new to this game and currently at the same point with winding a Surf PI coil. I fully appreciate the excellent assistance offered by contributors to this Forum and the discussions greatly stimulate the grey matter.

Be careful however with the Coil Calculator - it can be a little confusing.

The way I understand the Calculator is that it requires the Inner Radius of the proposed coil NOT the Diameter - it is easy to enter the wrong dimension and end up with the wrong winding info as (perhaps) the previous poster may have done (a 280 mm Inner Radius will end up being ~560mm or 22 inch in Diameter).

I would recheck your coil figures with the Calculator again being careful to enter the Radius in lieu of Diameter - I would suggest removing ~16 turns from your existing 36 turn coil would be very close to that required ie 125mm Inner Radius (a 250mm Diameter coil), 0.5mm wire, with 20 turns should give you 0.312 mH.

Good luck with the winding - this is a great hobby.

[Lord Dracon]

OOps sorry .. missed that little radius thing ... i admit i mis-read and thought diameter (i don't know why .. getting old i suppose ).. so i appologise for the confusion .. also my calcs will be out because of this ... its been a long time since i was doing mine i wondered why it seemed a few less turns than what i made .. i was sure it was 20 ish (thanks arkaba - you have proved me right and my gut feeling .. maybe i should slow down a bit and read it more carefully)

but, yes using this calculator will help an awful lot