gold nugget simulation?

[dbanner]

Just out of curiosity, is IACS standard related to annealed copper(pure) or pure copper. I would have expected closer to 100℅ for the .999 copper bar.

[Carl-NC]

So at 60kHz, your nickel sample is appearing too thin to register.

Yes, I forgot about it being ferromagnetic. My steel block shows the same error.

[Skippy]

Is it 999 fine steel?

[Ferric Toes]

I found my Hocking AutoSigma, it's a 3000 handheld model. It is dual frequency, 60kHz & 200kHz, but the 200kHz doesn't seem to work. I get an overload error even when trying to calibrate.
Hi Carl,
The manual for the 3000 states that it is 60kHz and 500kHz. The latter for even thinner materials.

You didn't state what reading you got on your calibration piece at 60kHz. Your .999 copper bar may not be annealed, which makes a difference. Copper can be hardened without the addition of other alloyed metals, and hardening reduces the IACS%.

I once used nickel loaded paint for coil shielding until I realised that I was getting a long low level decay curve from it due to its ferromagnetic properties.

Eric.

[green]

I had a quick go at sanding a 'zinc' cent, the copper disappears in seconds, it's very thin. But it will only be good for targets 1.1mm thick or less, I think. Hence my interest in the old WW2 coins, which were more substantial, close to 2mm thick.

Trying to revive my test detector is proving troublesome. The stock coil is not working, looks like the TX is dead, so I've got to dismantle the cable plug. I tried with the small 5" coil, but it's not giving the same readings as the large one, it's more picky about target sweep speed.
I did do some provisional tests, which showed the difference in time-constant between square and circular targets was much less than the rough estimates I made earlier in this thread. It looks like the difference is 95%, that is to say a circular target has 95% of the TC of a square one ( same diameter / edge length). It seems the current doesn't flow into the corners much. This is for thin targets - in this case 0.102mm aluminium foil.
It's looking like the multiplier was a bit high in our T-C formula, perhaps more like 0.065 for square, 0.061 for circular.
I'll have to dig out the original targets ( alu drinks can sheet ) that I used when devising the '0.072' figure ... I think one potential error may have been the use of a different can for the IACS measurement. I have some newer cans 'in stock' that are from the same batch, so I can get matching IACS samples and targets.

I had round and square closer the last time I tried, tried again today with a larger target. Like .061 for round and .065 for square. Tried a zinc cent, plain and sanded to see if it made a difference(about the same). The sanded thickness measurement works better when calculating TC. Carl had a 99.9% nickel test target, did a test with some 5 cent targets including one with a 99.9% nickel Canada 5 cent.

Included linear-linear charts to show difference in charting. Looks like straight line decay linear-log after one to two times target TC. Don't know if the 2008 Canada 5 cent ever straightens(steel). Didn't know if the 99.9% nickel would?

Sanded penny. No copper on outer diameter, couple small spots towards middle.


Used Tx_4msec constant rate.

[Carl-NC]

I found my Hocking AutoSigma, it's a 3000 handheld model. It is dual frequency, 60kHz & 200kHz, but the 200kHz doesn't seem to work. I get an overload error even when trying to calibrate.
Hi Carl,
The manual for the 3000 states that it is 60kHz and 500kHz. The latter for even thinner materials.

You didn't state what reading you got on your calibration piece at 60kHz. Your .999 copper bar may not be annealed, which makes a difference. Copper can be hardened without the addition of other alloyed metals, and hardening reduces the IACS%.

I once used nickel loaded paint for coil shielding until I realised that I was getting a long low level decay curve from it due to its ferromagnetic properties.

Eric.

Yes, 500kHz, was going from memory last night. The 500k gives an error even on thin foil. On the pure nickel bar, I get readings when the probe is very slightly off the surface. I've had this meter for years but never messed with it.

[Ferric Toes]

Yes, 500kHz, was going from memory last night. The 500k gives an error even on thin foil. On the pure nickel bar, I get readings when the probe is very slightly off the surface. I've had this meter for years but never messed with it.

Hi Carl,
Do you have one of the Whites 1oz silver medallions? I was given one on one of my early visits to Whites and it is .999 silver. This reads 101.4 on my meter when probe is on the older Ken Whites head. Slightly less on other side, maybe uplift on the Whites logo, but still 98 - 100. Should be the same at 60kHz.

Eric.

[Carl-NC]

I think I do, but don't know where it's at.

[Skippy]

Quote:"Tried a 'zinc' cent, plain and sanded ..... The sanded thickness measurement works better when calculating TC."
That was one of the problems I had when trying out the TC formula on coins - determining the thickness. You can't use the 'official' figure, because it's the maximum thickness. And if you use a micrometer on the face, you get a variety of readings, of course. So I ended up using the weight, the density, and the diameter to calculate the mean thickness of any given coin. It seemed to work well. Sanding down the coin makes thickness much easier, as well as removing the thin copper layer. I haven't tried it, but I assume ferric chloride ( PCB etchant) would remove the copper without affecting the zinc ? ... should this be of any scientific use.

I was just looking at your foil decay chart ( round and square shape) and the figures don't seem to match up with the 'TC formula' . As you're fond of foil, I assume you've measured its %IACS figure ? It should be fairly easy, as cooking foil comes in large sheets. I haven't done it yet, only for the thicker foil used in food trays ( 55 / 56% ).

[Ferric Toes]

From Skippy - And speaking of Zinc coins: Eric, do you have any zinc coins among your numismatic collection? Like the WW2-era German, Austrian coins ? I was curious what Mr.Hocking made of them.

To Skippy/ A local friend has some, '3 Reich issue coins, 2 for Germany and 1 for France when under German control'. I will have these on Monday for conductivity test.

I also have on order some 10mm diameter 99.95% zinc rod which is just sufficient diameter to get a conductivity reading. I can then use my lathe to part off different thicknesses and diameters to test for time constants. I would also suspect that a cylindrical piece with a diameter equal to the length would be close to a sphere in it's response. If successful, I could send some to yourself and Green for further tests on your equipments.

Interesting thing about nickel being ferromagnetic. A US nickel is 75%Cu and 25%Ni and yet has no magnetic susceptibility as measured on the Bartington MS2. I read somewhere that there has to be more than 56% nickel in an alloy for ferromagnetic properties to appear. I don't have any more details on this at present.

Eric.

[green]

Quote:"Tried a 'zinc' cent, plain and sanded ..... The sanded thickness measurement works better when calculating TC."
That was one of the problems I had when trying out the TC formula on coins - determining the thickness. You can't use the 'official' figure, because it's the maximum thickness. And if you use a micrometer on the face, you get a variety of readings, of course. So I ended up using the weight, the density, and the diameter to calculate the mean thickness of any given coin. It seemed to work well. Sanding down the coin makes thickness much easier, as well as removing the thin copper layer. I haven't tried it, but I assume ferric chloride ( PCB etchant) would remove the copper without affecting the zinc ? ... should this be of any scientific use.

I was just looking at your foil decay chart ( round and square shape) and the figures don't seem to match up with the 'TC formula' . As you're fond of foil, I assume you've measured its %IACS figure ? It should be fairly easy, as cooking foil comes in large sheets. I haven't done it yet, only for the thicker foil used in food trays ( 55 / 56% ).

Measurement today: Aluminum foil. (Wiki) regular .016mm thick, heavy duty .024mm thick. My micrometer has .001mm resolution, need at least a reading of 100 so I folded a piece 3times for 8 layers. Regular strength 138/8=17.3mm. Heavy duty 195/8=24.4mm.Is there a more accurate way to measure foil thickness? Haven't measured foil DC resistance. My multimeter, .01mv resolution .06%reading+-2 count. .01ohm resolution .3%reading +9 digits. Thinking at least 5mv reading across foil sample. If math is correct, regular foil 1cm wide, 1m long about .166 ohms. Need 30ma. 400ohm resistor in series with foil across 12v. R=E/I, R=V across 1m foil/(V across resistor/resistor ohms). Aluminum resistance increases 1% every 2.3deg C. Should I correct for temperature? Haven't been with most of the charts. How to cut foil 1cm width to +-.1mm? Need to measure across 1m length, how much longer does foil need to be to connect to circuit. I've made a lot of measurements strain, torque, temperature, speed, force are some of them. Just guessing the best procedure to use for these measurements. Any thoughts or suggestions?

Didn't measure foil thickness in reply #260. Cut targets from a piece I thought was heavy duty, the charted data matched regular strength so I used Reynolds aluminum foil thickness.

Haven't been listing ambient temperature on charts. Probably should since my ambient in the garage can change some. A few % change in target TC would make a difference when calculating multiplier or %IACS. Probably not to important for detecting, just when trying to understand why.

[Skippy]

That's partly why I asked about the foil measurement, I know it's not easy, and wondered whether you had struggled to get a good reading. I'm fortunate that at work, I have access to a fancy HP meter, with too many digits and 4-wire resistance capability, so that will be my choice for difficult jobs. But at home, I have to think harder. The tempco of aluminium is something I had considered, not the ambient temp, but any temp rise caused by passing too high a current through the sample. And what is too high? All the usual sources give data for round copper wire, not wafer-thin alu. So halve any copper current, due to about 50%IACS conductivity figure. But a flat sheet should be able to dissipate the heat much better than round wire ...?
I agree with your calcs, I made it roughly 180 milliOhm for 1m of 'regular' foil, and 50mA test current should be OK. You only need the foil sample to be 5cm longer at each end than the measured length; ie 110cm length, with monitoring points 100cm apart.
Thickness measurements? Lots of layers, I used 20 I recall. It's important to avoid creases, wrinkles, I tried folding a long strip in a /\/\/\/\ Z-fashion, to avoid the problems of repeatedly folding in the same direction.
Cutting 1cm wide strip accurately is tricky. I would use scissors, I think a sharp craft knife is likely to cause tearing. Apply tape to it, I would use decorators paper 'masking tape', but Scotch tape or whatever you have, will no doubt be fine. Is there logic to making a 2cm wide strip, and using twice the test current?

In the past, I've knocked up DC amplifiers from precision opamps like OP27's, to add resolution to basic meters. I have a couple of fancy-azz chopper-stabilised opamps somewhere, DC offset voltages in the microvolt range, probably intended for strain-gauge work. I never have found a need for them .....

[Ferric Toes]

I checked a piece of my 12mm D brass rod and got an IACS% of 25.5, which is not far short of what zinc is. I then machined down to smaller diameters and parted off 2mm thick discs. I ended up with a disc 4mm D x 2mm thick which gave a plotted decay curve that matched my 0.87gm nugget. TC measured 5.8uS and detectable at 5" with the 7" coil on a Vallon electronics. I will try a 3mm D x 2mm next in brass and when the zinc rods come I will do a comparison, although I expect the sizes will be very similar. As Carl said, any metal will do, although lower conductivity metals, for the same TC, will result in larger and less easily lost test pieces. The detection ranges may be then be over optimistic though. I am going to embed my brass ones in the ends of acrylic rods as the mine detector test pieces are.

Eric.

[green]

More tests
Measure DC resistance: Used .7inch(17.8mm) wide masking tape across 12inch width of regular foil and trimmed foil width with scissors(thanks Skippy, good way to do it). Marked foil at 250mm about 1inch from each end. Connected foil, (5)500 ohm 1/4W 1%resistors in parallel and 12V battery in series. Volts across resistor 12.30, volts across 250mm foil 3.02mv, ambient 16degC. Calculates .025 ohms at 20degC. Measured thickness with micrometer. Reading didn't repeat each time so I recorded the lower number that repeated. .138mm/8layers=.01725mm, .278mm/16layers=.01738mm. More layers read higher. Cut a piece 170mm x 139mm and weighed it, 16.1grains. Calculated thickness=.0162mm. (.025ohms*.0178*.0000162*4=2.884ohm-m) IACS=59.8%. Thinking thickness by weight is more accurate, cutting a strip of foil 2 or 3 times longer could increase the accuracy, not sure it's needed. Recorded decay curve and added a best fit decay line, added 2 more lines(1.03 and .97 best fit)either could work so my slope could have a 2 or 3% error just from drawing the slope.

[Polymer]

... Interesting thing about nickel being ferromagnetic. A US nickel is 75%Cu and 25%Ni and yet has no magnetic susceptibility as measured on the Bartington MS2. I read somewhere that there has to be more than 56% nickel in an alloy for ferromagnetic properties to appear. I don't have any more details on this at present.

Eric.

Thought this might be of interest to you regarding CuNi alloys.



from https://www.copper.org/applications/...I_booklet.html