DEEPER PI DETECTION DEPTH

[Aziz]

May I add some comments to the induction heater principle coils?


With low operating frequency and very low TX coil inductance (5-25 µH), we would require a high tank capacitance.
Forget using only one capacitor. We would require a dozen or so and all of them need to be placed in the coil (or very close to it). Best operated with high current impulse capacitors like FKP-1 (or FKP-4, must be a metal foil capacitor). Would make the TX coil quite heavy.
Oh yes, the orientation of the capacitors does matter unless you do not want them to be melted (eddy current reduced orientation). And the connection of them does matter too.

Could be a challenge to realize it. Probably easier with big coils.

I think, it is much simpler to handle a high voltage rather than a high current.
What does matter again?
Oh yes, the TX coil current change (dI/dt).
Self induction voltage of the TX coil:
U = -L*dI/dt

Impedance matching: -> U gets smaller, I gets bigger, L is small
The standard variant: U is high, I is small, L is big

So where is the free lunch for me?

Aziz

PS: Merry Christmas and a happy new year to everyone.

[Tinkerer]

We have seen in the posts above, that it is well possible to pump a lot of power into a TX coil.

Why would we want a lot of power?

We are talking of detecting deep. We also want to detect small targets deep.

For deep detecting, a large coil gives best results. However, for a large coil to detect small targets, we need to boost the coil's magnetic field, to obtain a reasonable field density, needed for detecting small targets, deep.

A large coil, for example 1 meter x 2 meters covers a lot of ground. If we attach it to a vehicle, we could cover a lot of ground in a short time, for example in a dry salt lake.

How large are the average nuggets in that area?

Can we detect 1 gram nuggets with this coil? This is where the high power comes in. High magnetic field density.

A 1m x 2m coil would also make a good tow-able underwater detector, for example to search for artifacts like gold and silver coins, scattered over a large area by a sinking galleon, broken up in a hurricane.

Tinkerer

[Tinkerer]

Skin effect

Another matter to look at, for a deep search detector, is the skin effect.

How does the skin effect affect the target response?

We know that thin aluminum foil is a good PI test target. We can cut a one inch square of foil and it will give a target with a TC of about 10us.

If we cut this piece of foil in half, we get 2 targets with a TC of 5us. Only good PI detectors are capable to detect that at a reasonable distance.

If we take a thicker piece of aluminum, we get a target with a longer TC. In fact, the target response increases significantly up to the thickness of about 0.05mm to 0.1mm.

This is the skin depth of the TX switch off transient.

The skin depth means that the eddy currents generated in the target, only reach "skin deep" into the target.

A deeper skin depth of eddy currents will generate a target response with a longer TC or time constant. The target time constant is the amount of time it takes for the target response to diminish or decay to about one third of it's value.

A slower switch off, will give a deeper skin depth. A deeper skin depth gives a longer TC for the same metal.

Here we come to a new significant factor. The type of metal.

Aluminum and gold have nearly the same conductivity and therefore the same skin depth for the same frequency. However, most aluminum is alloyed and most gold is alloyed too.
Even virgin gold, like in nuggets often contains silver, platinum and other metals.

Nuggets also often contain stone fragments. This changes the TC of the nugget.

An uneven surface also greatly changes the TC of a target. It also changes the skin depth of the eddy currents.

Lead has much less conductivity than aluminum or gold, so it has also a lesser skin depth and therefore a shorter TC.

If we compare the density of aluminum with lead or gold, we see that if we have a sphere of gold of a certain weight, a sphere of lead of the same weight would be larger and a sphere of aluminum of the same weight, would be much larger.

The sphere of aluminum would therefore intercept more magnetic field line of the TX pulse and produce a larger response.

When we search deep, like 0.5m to 1m depth, for nuggets, we are not much interested in nuggets of milligram weight. We search for nuggets of at least 1 gram and upwards.

Therefore we want to design the TX pulse specifically for the size of nugget we are searching. We want a deep skin depth that generates a strong target response that lasts for a considerable amount of time.

We have seen many different examples of TX pulses above. Which type of TX pulse generates the strongest response in thick targets, like gold nuggets?

Tinkerer

[Midas]

Hi Tinkerer,,

You raise some excellent points. People talk about maximizing dI/dT to increase performance but really what you want to do is maximize dI and optimize dT. Wiki has a really good write up on skin effect including some very useable formulas. Its even got an example calc on gold, 11.8mm at 50hz. Of course even though jamming energy in too fast is inefficient because of skin effect, too slow is just as bad since its constantly leaking out. There will have to be some happy medium in between. Oh btw according to the wiki formulas the reduced conductivity of lead actually results in a greater skin depth. I think its shorter TC is actually a direct effect of its reduced conductivity.

Midas

[Tinkerer]

Hi Tinkerer,,

You raise some excellent points. People talk about maximizing dI/dT to increase performance but really what you want to do is maximize dI and optimize dT. Wiki has a really good write up on skin effect including some very useable formulas. Its even got an example calc on gold, 11.8mm at 50hz. Of course even though jamming energy in too fast is inefficient because of skin effect, too slow is just as bad since its constantly leaking out. There will have to be some happy medium in between. Oh btw according to the wiki formulas the reduced conductivity of lead actually results in a greater skin depth. I think its shorter TC is actually a direct effect of its reduced conductivity.

Midas

thanks for the feedback. You are right, lead has a greater skin depth.

Lower conductivity gives a shorter TC. Why? Does it have to do with the skin depth?

And yes, the TX time needs to be within precise limits. Tests have shown that the target response for large targets increases with a longer TX time, but only up to a certain limit, after that the response decreases again.

There is hope that, with digital technology, we may one day be able to adjust the TX pulses precisely for maximum target response, for our preferred targets.

Tinkerer

[Sean_Goddard]

Funny enough, my laptop computer gives me more electrical pricks than the TX connector with 800V Flyback, when I plug and unplug it under power.

It's a DELL isn't it?

Their power supplies are illegal in the UK due to the floating voltages (called "Touch Voltages"). There is sufficient to kill someone with a weak heart. I believe they were forced to re-design them so all is well with current models - allegedly.

[golfnut]

I like the idea of utilising the electronic used in Induction heaters.

High volts, some C, small L, Low R, big thump - lots of magnetism right there.


This site is as good as any..


http://www.richieburnett.co.uk/indheat.html

[Tinkerer]

It's a DELL isn't it?

Their power supplies are illegal in the UK due to the floating voltages (called "Touch Voltages"). There is sufficient to kill someone with a weak heart. I believe they were forced to re-design them so all is well with current models - allegedly.

Hi Sean, thanks for coming back to this problem.

Yes it is a DELL. I suppose that the voltage is capacitivly coupled from the PSU. Whereas the alu housing is supposed to shield the EMI.

It is something to take a closer look at, so the same mistake does not creep in on a PI detector in an alu box.

Can anybody add some more information on this problem?

Tinkerer

[Midas]

Lower conductivity gives a shorter TC. Why? Does it have to do with the skin depth?

Lower conductivity = higher resistance = faster conversion of induced currents into heat. I'd say no, nothing to do with skin depth, in fact because your standard PI pulse can get currents to flow deeper in the lead (hence effectively through more conductor cross sectional area) in would counter the effect of its lower conductivity to some extent. Probably explains why people have no problems at all finding bullets.

[Carl-NC]

We know that thin aluminum foil is a good PI test target. We can cut a one inch square of foil and it will give a target with a TC of about 10us.

If we cut this piece of foil in half, we get 2 targets with a TC of 5us.

I think you have 2 targets still with a TC of 10us. The TC doesn't depend on surface area, surface area just affects the strength (amplitude) of the response.

If we take a thicker piece of aluminum, we get a target with a longer TC.

I have a set of 1" square targets made of household aluminum foil, with thicknesses of 1x, 2x, 4x, 8x, 16x, and 32x, each laminated with clear tape. Makes a good standard.

The skin depth means that the eddy currents generated in the target, only reach "skin deep" into the target.

Well, actually, eddy currents reach about 5 skin depths deep.

A slower switch off, will give a deeper skin depth. A deeper skin depth gives a longer TC for the same metal.

I don't understand the first half of this.

Lead has much less conductivity than aluminum or gold, so it has also a lesser skin depth and therefore a shorter TC.

While the characteristic skin depth of a metal affects TC, so does thickness. So saying "lesser skin depth" == "shorter TC" isn't necessarily true, unless the targets are otherwise identical. Ferinstance, a 58-cal lead Minie has a longer TC than a 1 grain pure gold nugget.

Therefore we want to design the TX pulse specifically for the size of nugget we are searching. We want a deep skin depth that generates a strong target response that lasts for a considerable amount of time.

"Skin depth" is a characteristic value that doesn't change with pulse width; I think what you want is a deep "penetration depth."

- Carl

[Aziz]

What is better?
A higher or a lower stimulation frequency in respect to skin depth?


A lower stimulation frequency will increase the skin depth, which will increase the targets time constant due to lower current path resistance.
If we double the frequency, the skin depth will decrease by sqrt(2) and the resistance will increase by sqrt(2). The eddy current will decrease by sqrt(2).

On the other hand, if we double the stimulation frequency, the induced voltage in the target will double (direct proportional to frequency), which the eddy current doubles at the end if we don't take the skin effect into account.

Putting both together, if we double the stimulation frequency, the induced target voltage will double, the resistance will increase by sqrt(2):
I=U/R, f=2*f -> I = (2U)/(sqrt(2)*R)
Eddy current I will increase by 2/sqrt(2) = sqrt(2) = 1.41

The target can be energized with more eddy current by higher stimulation frequency.

So increasing the stimulation frequency wins. That's fine.

Aziz

[Tinkerer]

Hi Carl, thanks for the feedback. I will add my comments between your lines.

I think you have 2 targets still with a TC of 10us. The TC doesn't depend on surface area, surface area just affects the strength (amplitude) of the response.
It has been some time since I made this test, but as I remember, each of the 1/2" squares had a TC of about 5us. Of course, there is the possibility that the lower target amplitude made the response fade into the noise sooner and that I mistook this for the TC. I did not actually calculate the TC from the decay curve.

I have a set of 1" square targets made of household aluminum foil, with thicknesses of 1x, 2x, 4x, 8x, 16x, and 32x, each laminated with clear tape. Makes a good standard.

With a thicker target I was thinking of a solid target of a thickness that corresponds to the skin depth. Aluminum foil has a coating of aluminum oxide, which is an excellent insulator. Laminating 32 layers of aluminum foil might have strange effects due the the the inter-laminate capacitance.


Well, actually, eddy currents reach about 5 skin depths deep.
Agreed, but the skin depth follows the classical exponential curve, where the 1TC is roughly 2/3 and the other 4 skins have rapidly diminishing little influence.

I don't understand the first half of this.

With a slower switch off, I meant that the same di/dt applied over a longer time will generate a stronger target response in long TC targets.


While the characteristic skin depth of a metal affects TC, so does thickness. So saying "lesser skin depth" == "shorter TC" isn't necessarily true, unless the targets are otherwise identical. Ferinstance, a 58-cal lead Minie has a longer TC than a 1 grain pure gold nugget.

"Skin depth" is a characteristic value that doesn't change with pulse width; I think what you want is a deep "penetration depth."

Yes, I mean eddy current penetration depth into the target. In general, the pulse width corresponds to the amount of energy stored in the coil magnetic field. With little energy, or short TX pulse, we can obtain a high di/dt for a short time. With a lot of energy stored in the coil magnetic field, we can obtain the same di/dt for a longer time. A longer time for a long TC target, gives a higher response amplitude.


- Carl

[Carl-NC]

I think you have 2 targets still with a TC of 10us. The TC doesn't depend on surface area, surface area just affects the strength (amplitude) of the response.

It has been some time since I made this test, but as I remember, each of the 1/2" squares had a TC of about 5us. Of course, there is the possibility that the lower target amplitude made the response fade into the noise sooner and that I mistook this for the TC. I did not actually calculate the TC from the decay curve.

That's probably the case.

I have a set of 1" square targets made of household aluminum foil, with thicknesses of 1x, 2x, 4x, 8x, 16x, and 32x, each laminated with clear tape. Makes a good standard.

With a thicker target I was thinking of a solid target of a thickness that corresponds to the skin depth. Aluminum foil has a coating of aluminum oxide, which is an excellent insulator. Laminating 32 layers of aluminum foil might have strange effects due the the the inter-laminate capacitance.

Foil layers work fine as long as they are tightly pressed together. There is no vertical current flow so the oxide doesn't matter. Gaps between layers do matter which you can see if you let the stack relax, but the oxide thickness isn't enough affect this.

Well, actually, eddy currents reach about 5 skin depths deep.
Agreed, but the skin depth follows the classical exponential curve, where the 1TC is roughly 2/3 and the other 4 skins have rapidly diminishing little influence.

Be careful equating TC with skin depth. I think in this case you are simply comparing exponentials. In any case, a metal with a thickness of 5 skin depths will have a fairly different response than a same metal with a thickness of 1 skin depth, but not different than one with 10 skin depths.

With a slower switch off, I meant that the same di/dt applied over a longer time will generate a stronger target response in long TC targets.

A same di/dt with a "slower switch-off" (higher dt) implies a proportionally higher current, and I think this would be the only reason you would get a stronger response. This would also produce a stronger response in short TC targets, but the slower switch-off might prevent the early sampling required to see the target.

Yes, I mean eddy current penetration depth into the target. In general, the pulse width corresponds to the amount of energy stored in the coil magnetic field. With little energy, or short TX pulse, we can obtain a high di/dt for a short time. With a lot of energy stored in the coil magnetic field, we can obtain the same di/dt for a longer time. A longer time for a long TC target, gives a higher response amplitude.

Be careful! TX pulse width only corresponds to the amount of energy stored in the magnetic field if the current is still on the rise at the turn-off point. If my coil current flat-tops at 50us, then a pulse width of 100us produces the same TX field.

As above, I think you are looking at this backward, and your conclusion is only valid because a long dt flyback precludes the use of early sampling required for short TC targets. That is, if we could use a high current TX pulse and still use early sampling (perhaps by using an IB coil) then short TC targets would also have a higher response amplitude.

[Tinkerer]

[/QUOTE]
As above, I think you are looking at this backward, and your conclusion is only valid because a long dt flyback precludes the use of early sampling required for short TC targets. That is, if we could use a high current TX pulse and still use early sampling (perhaps by using an IB coil) then short TC targets would also have a higher response amplitude.


[/QUOTE]

You are right, I look at things from a different angle, so I perfectly agree with your comments.
I like to use a saw-tooth TX wave form that is not commonly used with traditional PI and sample at, what would be called extremely early, which is not possible with traditional PI.

And then I forget to mention all that and just talk about PI. So I cause all this confusion. Sorry about that.

Tinkerer

[Midas]

Hi All,

Well this has been somewhat enlightening for me. It would seem that a faster switch off is in fact allround beneficial. Intuitively equating inductance with momentum it seemed to make sense that superfast pulses might not be the best way to transfer a given quantity of energy. Kind of like trying to get a car moving by shooting bullets at it. At some very high frequency I'm sure that would prove to be true. ie. if you tried to pulse it faster than the resonant frequency of the target nugget. But since thats probably somewhere in the MHz (anyone ever tried to work it out?) it can probably be completely ignored. And since Aziz has quite rightly pointed out that increased skin effect is less detrimental than the benefits of higher frequency, bring on the near vertical current ramps...

Carl, I'm not sure alumimium foil is much good as a 'standard'. 0.022 millimetres is apparently typical thickness but I highly doubt its universal across all brands. Coins are an obvious choice for consistancy though I guess there not small enough for low TC nugget simulation.

Midas