DEEPER PI DETECTION DEPTH

[Aziz]

Aziz is right mine did have a diode and the wave form is sinewave ... it is part of my new project D-IGSL ( Digital IGSL ) it rocks

Hi Moodz,

let's revival the good old ideas. They are prior art and the patent trolls can go to the hell.

Aziz

[Tinkerer]

Aziz is right mine did have a diode and the wave form is sinewave ... it is part of my new project D-IGSL ( Digital IGSL ) it rocks

Ups, sorry for the mistake.

Here is the corrected sim.

Again, the yellow trace is the coil current.

The other 5 traces are the responses of resistive targets as could represent gold nuggets of different size.

As it is, this Moodz version uses about 10% less power than the TEM sim as shown above.

Tinkerer

[Aziz]

Tinkerer,

this is now looking right.
The more efficiency is owed to the fact, that a lot of the energy is held in the resonant LC tank rather than moved in the whole TX circuit. Only losses will be injected into the LC tank.

Let's heat up the gold nuggets -> hot gold nuggets (induction heater principle).

Aziz

[Tinkerer]

Tinkerer,

this is now looking right.
The more efficiency is owed to the fact, that a lot of the energy is held in the resonant LC tank rather than moved in the whole TX circuit. Only losses will be injected into the LC tank.

Let's heat up the gold nuggets -> hot gold nuggets (induction heater principle).

Aziz

Maybe I should add, that the sim that I show, is not exactly optimal, I just tried quickly to adjust it for the same power consumption as the other sims, that is about 1A.
The method can be optimized for better results.

Tinkerer

[Tinkerer]

Below, again the traditional PI method, so that we can compare the amplitude of the target response.
About the same power consumption. (1A)

The yellow trace is again the coil current.

The PPS are very much less, about 1000PPS

The peak target response is 2.24A

About the amplitude of the target response: To be able to show the TX current and the target response on the same screen, so that one can see the relation, the scales had to be manipulated. Obviously the real target response is not 2.24 Amps, more like 224uA.

But the target response in itself is fairly accurate. This whole series of sims, use the same proportions so that they can be compared with each other.

Tinkerer

[Aziz]

More power to the people - ups - I mean targets!

Now split the coil L into two halves: upper and lower coil part with either equal or different winding turns (auto-transformer) and feed it at the center junction of the coils (capacitor over both coil parts). See below for more details.

This gives the target the freakin kicking at the expense of more battery power consumption of course. And you can crank up the freakin flyback voltage (kilo Volts region if you like), which gives more room the to limitted break down voltage of the mosfet and diode. The lower coil part (near the diode/mosfet) should have either equal winding turns or lesser.

You can find the correct inductance splitting formula in the spice model.
When Num1=1 and Num2=1, it's a center tapped coil.
When Num1=1 and Num2=2, the upper coil part has two times turns count of the lower coil part. Take care of splitting the resistance of the coils correct (LR1(lower) and LR2(upper)).

Beam me up Scotty!
Bzzzzzzzzzd.

Aziz

PS: I have added the zipped spice file for your convenience.
Cheers

PPS: Change the buffer caps on the right side to the same value (CB2 and CB4 = 10000µ).

[Midas]

The yellow trace is again the coil current.

Hi Tinkerer,

Your colour blind I assume? The coil current trace is actually green. From left to right you have: Green, Blue, Red, Turquoise, Magenta, Grey

Midas

[Midas]

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

This site has a lot of good information that is very applicable to what you guys are simulating. What maybe of particular interest is the different drive inverter designs. No need to limit yourself to the one switching component, half bridge, full bridge or even multiple parallel full-bridge is also achievable and quite possibly advantageous.

[Midas]

Reading through it again it also raises an extra complication that hasn't been mentioned. I think the ground and target effect will behave as what Richie describes as a 'lossy workpiece'. As well sapping energy and bringing the 95% efficiency dream southward it will also change the resonant frequency. So that's something that will need to be tracked and corrected for.

[Aziz]

Reading through it again it also raises an extra complication that hasn't been mentioned. I think the ground and target effect will behave as what Richie describes as a 'lossy workpiece'. As well sapping energy and bringing the 95% efficiency dream southward it will also change the resonant frequency. So that's something that will need to be tracked and corrected for.

Midas,
the modulation of the coil current and frequency due to ground and targets is intended! All the necessary information is there.
How to process all the information?
Well, this is the holly grail and all the patent trolls are just waiting for its reveal.

Aziz

[Tinkerer]

Hi Tinkerer,

Your colour blind I assume? The coil current trace is actually green. From left to right you have: Green, Blue, Red, Turquoise, Magenta, Grey

Midas

Sheeesh, now I know why the "green light" is always so short and the "other green" light and the red light take so long.

Tinkerer

[Midas]

Sheeesh, now I know why the "green light" is always so short and the "other green" light and the red light take so long.

Tinkerer

So you get 33.33% more green lights, you gotta be happy with that!
FYI, It was a genuine question, I wasn't being mean. Its more common that you might think.

Midas

[Aziz]

Hi all,

I have "tried" (simulated) the induction heater principle to the TX circuit with impedance matching circuit. Even you can achieve a very high ampere-turns or a high current power factor (10 - 20x), it's efficiency isn't the best however.

It offers other benefits:
- very low inducitivity TX coil (5-25 µH !)
- low voltage everywhere (coil, TX circuit) so you can use the common low voltage mosfets with less on resistance
- simple to drive (H-bridge)
- coil leads and TX driver stage drives less current (less losses)
- low EMI emission during the switching

But the TX tank (LC, coil + caps) need a very low resistance/impedance parts. A few TX coil turns but you need a thick litz cable. Otherwise, almost all of the energy will be burnt in these parts (generating much heat due to very high current).

To sum up: It works!

Aziz

[Tinkerer]

So you get 33.33% more green lights, you gotta be happy with that!
FYI, It was a genuine question, I wasn't being mean. Its more common that you might think.

Midas

OK now I understand why every September when I have my eye test, to renew my drivers licence, they show me this funny polka dot picture with Father Christmas on it and ask what I see. Then they ask for a "generous Christmas gift", which I always have ready in cash, after which they give me a 20/20 for the eyes and full #5 grade (all vehicles) drivers licence renewal.

Unfortunately, when I populate the SMD board, the eye sight deteriorated. The TSOT23-5 parts sometimes walk away before I get the solder iron to it, because it turns out they have 6 legs, not five and are more commonly called Ticks.

Anyway, of course you are right, it is a light yellowish green. About like a green lemon.

Tinkerer

[IBGold]

I have the same problem with small spiders and I thought I was the only one.

MERRY CHRISTMAS ALL.

Regards, Ian.