how to simulate a PI metal detection in the software such as ltspice

[Tinkerer]

I haven't tried capacitor coupling to preamp in, hardware or simulation. Added input from PI-1 to circuit above(reply #7). Input capacitor causes an offset but maybe no different than amplifier input offset?

About capacitor C1:
C1 and R5 make a high pass filter at fo about 16Hz, using + input to Preamp.

[green]

About capacitor C1:
C1 and R5 make a high pass filter at fo about 16Hz, using + input to Preamp. If you use the - input, you can put R5 in series with C1. Either way, this attenuates the coil signal by about 50% or more, depending on the sweep speed and coil diameter.
The solution would be a large C1, like 100uf/R5/1k Not very practical. Or large R5/100k/1uF, where the 100k resistor adds considerably to the noise. Like 0.16uV noise at the coil signal frequency of 16Hz.

I've been direct coupling the coil to the amplifier. Is there some advantage to capacitor coupling? Tried in simulation to learn something. See now I should have made simulation 1ms long and 100 cycles instead of one cycle. Don't see an advantage.

[Carl-NC]

Yeah I just do not know why the C2 needs 30ms to be set. In addition, I built the circuit in ltspice. The output of ne5532 is shown and i use a voltage source to produce the delay. However, the result from ne5532 shows that it keeps changing which is strange and after the 1k resistor the signal has an additional square part and the integrator circuit does not output a certain DC level and the output of TLO72 keeps raising. Could you please help me to solve it.

First, your opamp output looks like it might be working OK. Yes, there is a long-tern linear "droop" due to the charging tail on C2*. It doesn't matter as long as the initial portion of the opamp signal is around 2.5V when it gets sampled. What it does after it gets sampled makes no difference, unless you try to add a late sample for Earth field cancelation. This circuit won't work for that.

Second, the opamp output looks to be drifting upwards. Maybe it's still not settled, can't tell from your results. You need to run it a while (try 100ms), and then adjust R6 to get the initial output point close to 2.5V.

Third, you're running the TX off a simulated (I assume) 555 timer, but driving the demod from an ideal clock. Likely they are not synchronous. Run the TX from an ideal clock. Bonus: it'll make the sims run faster.


*I'll reference all numbers from the original schematic, not your LTSpice schematic.

[Tinkerer]

I've been direct coupling the coil to the amplifier. Is there some advantage to capacitor coupling? Tried in simulation to learn something. See now I should have made simulation 1ms long and 100 cycles instead of one cycle. Don't see an advantage.

Somebody once said that solderless breadboards are a royal waste of time. I really have to agree with that.
I just slapped together a capacitor coupled PI input to look at the offset level you mentioned. Different capacitors give different offset levels. I think it has to do with the leakage current of the capacitor. Then things started to go wild and it took me some time to realize that bad contacts were at fault.
Traditional PI uses the + battery rail as ground. some designs use AC coupling to place the ground at a different level.
It has been said that the minus Battery ground gives less capacitance coil to earth problems. PI coil shielding is supposed to cure this problem, but there is no consensus of what the resistance of the shielding should be.

[Carl-NC]

I've been direct coupling the coil to the amplifier. Is there some advantage to capacitor coupling? Tried in simulation to learn something. See now I should have made simulation 1ms long and 100 cycles instead of one cycle. Don't see an advantage.

The cap offsets the TX signal going into the opamp so you don't have to use two batteries or a charge pump to create a new rail voltage. If you direct couple, then it likely won't work, although there are a few opamps where inputs can run all the way to -V.

[Carl-NC]

Somebody once said that solderless breadboards are a royal waste of time. I really have to agree with that.

Heh, they can be tempermental but I love 'em. I'm a big fan of rapid prototyping.

[DJSUOL]

Thanks for your reply. Could you please tell me the advantages of the design such as surf PI or Hammerhead which are the popular ones in this forum?

[DJSUOL]

In addition, have you designed some PI type metal detectors? Would you mind showing it to me? I do not know how to design a metal detector even though I know the basic component stages.

[DJSUOL]

And why the output from ne5532 should be 2.5V?

[DJSUOL]

Would someone also tell me if I use two coils with diameter 12cm for the surf PI, will it work well and detect the metal such as coil at sufficient distance?

[kt315]

And why the output from ne5532 should be 2.5V?

what schematic you mean?

[DJSUOL]

卡尔 said the opamp signal should be around 2.5V by adjusting R6. I do not know why it should be this value.

[Carl-NC]

Thanks for your reply. Could you please tell me the advantages of the design such as surf PI or Hammerhead which are the popular ones in this forum?

The PI circuits in the book are very simple, just to illustrate PI basics. Hammerhead was designed as a parametric demonstrator, it gives you a lot more control. Other designs are closer to decent commercial units.

In addition, have you designed some PI type metal detectors? Would you mind showing it to me? I do not know how to design a metal detector even though I know the basic component stages.

Yes, I designed the Hammerhead, the TDI-SL, the Fisher F-Pulse, and a couple of security walk-throughs. The best thing for you to do is build some circuits and play with them. See what happens when you make changes. That was the purpose of Hammerhead.

And why the output from ne5532 should be 2.5V?

The input to the integrator needs to be ~2.5V. See IC1-pin3.

Would someone also tell me if I use two coils with diameter 12cm for the surf PI, will it work well and detect the metal such as coil at sufficient distance?

Not sure what you mean.

[eclipse]

Perhaps you should focus more on reading through the forum and doing more experimenting.

[DJSUOL]

I think the designs of surf PI and Hammerhead are similar. What I mean is that can I add an Rx coil for the surf PI and if I use a 12m diameter coil, will it work well? Finally, can you tell me about your WT design? Have you designed some multiple zone type ones? For PI-1 design in the book, why it creates a ground and adds a circuit for NE5532 IC1A pin- to compensate the offset?