DEEPER PI DETECTION DEPTH

[Qiaozhi]

Ok, I'll bite...

I guess my first impression is that the spectral shape is very sensitive to target time constant but not too sensitive to RX/TX coil inductance, so you can use that information to differentiate targets.

However, you have pretty radical differences in target resistance. Is that realistic?

For fun, I tried substituting actual metal conductivity resistances for the target resistor values. I have no idea if that is valid. So let me know if that is meaningless or not.

Below is a comparison of the RX response for those targets.

Does it mean anything, or just BS?

-SB

P.S. Gee, the phases look nice and differentiated above 100 kHz -- maybe some sort of phase detector would be useful... what does that remind me of...

Simon - In AZIZ's simulation he use targets with time constants of 1us, 10us, 100us and 1ms. These all seem very reasonable values to use, except perhaps the 1ms target.
However, in your version you have the following:
silver = 625ns
copper = 588ns
gold = 435ns
aluminium = 370ns
zinc = 167ns

These all seem to be far too short.
Remember that TC = L/R.

[NC-Dave]

My intention in posting what I did was attempt to kick start what has, as far as I can see, become a stalled process by getting people to analyse what they were doing in more depth (pardon the pun). Using Dave's name to imply that something was imminent when in fact this was NOT the case. It was wrong of me and I admit full liability.

Dave has expressed his anger regarding this matter, and I have apologised for the misuse of his good name. Dave in NO WAY endorses any of my work (and nor should he) and contrary to what was implied has NO intention of working on ANY project (commercial or otherwise) with me now or in the future!

One a personal level I have made NO advances worthy of note regarding PI Development, but have made a few discoveries which I shall be keeping to myself.

In light of the offence I have cause, this will be the LAST post I shall me making on this forum!

Thanks for all the great stuff I've learned and keep up the good work folks!

Sean, there is no need to quit the forum. Your post was dumb but you obviously did not intend to cause any problems. I enjoy reading all the stuff from people like yourself and Aziz as well as countless others on this forum. I would hate to have you simply fade away. To back this up I am going to ask the rest of the readers on the forum to all post a message requesting Sean to keep on posting. On a final note, Sean is a lot closer to a vastly improved beach machine than he realizes. His work is excellent. His only problem is that he tries to work too fast causing him to make a few mistakes. Keep up the good work, Dave. * * *

Dam it all, life is way too short and this is only a hobby.

[simonbaker]

My intention in posting what I did was attempt to kick start what has, as far as I can see, become a stalled process by getting people to analyse what they were doing in more depth (pardon the pun). Using Dave's name to imply that something was imminent when in fact this was NOT the case. It was wrong of me and I admit full liability.

Dave has expressed his anger regarding this matter, and I have apologised for the misuse of his good name. Dave in NO WAY endorses any of my work (and nor should he) and contrary to what was implied has NO intention of working on ANY project (commercial or otherwise) with me now or in the future!

One a personal level I have made NO advances worthy of note regarding PI Development, but have made a few discoveries which I shall be keeping to myself.

In light of the offence I have cause, this will be the LAST post I shall me making on this forum!

Thanks for all the great stuff I've learned and keep up the good work folks!

Hot FETs or hot people... which is tougher problem?...

Well, didn't seem like such a big offense, but not me to say I guess...anyway seems like you made it right...

Look forward to more hot discoveries you come up with in the future...

-SB

[simonbaker]

Ok guys,

I have put the simplest form of an AC analysis (TX -> Target -> RX). Regardless of small signal (1 V) or full signal (battery voltage), BTW this is totally irrelevant in the provided simple spice simulation, what can we obtain from it?

Is there a free lunch for us?


Comeon, feel free to comment it, what you see. We will get very valuable information from it.

Aziz

Ok, I'll bite...

I guess my first impression is that the spectral shape is very sensitive to target time constant but not too sensitive to RX/TX coil inductance, so you can use that information to differentiate targets.

However, you have pretty radical differences in target resistance. Is that realistic?

For fun, I tried substituting actual metal conductivity resistances for the target resistor values. I have no idea if that is valid. So let me know if that is meaningless or not.

Below is a comparison of the RX response for those targets.

Does it mean anything, or just BS?

-SB

P.S. Gee, the phases look nice and differentiated above 100 kHz -- maybe some sort of phase detector would be useful... what does that remind me of...

[Sean_Goddard]

My intention in posting what I did was attempt to kick start what has, as far as I can see, become a stalled process by getting people to analyse what they were doing in more depth (pardon the pun). Using Dave's name to imply that something was imminent when in fact this was NOT the case. It was wrong of me and I admit full liability.

Dave has expressed his anger regarding this matter, and I have apologised for the misuse of his good name. Dave in NO WAY endorses any of my work (and nor should he) and contrary to what was implied has NO intention of working on ANY project (commercial or otherwise) with me now or in the future!

One a personal level I have made NO advances worthy of note regarding PI Development, but have made a few discoveries which I shall be keeping to myself.

In light of the offence I have cause, this will be the LAST post I shall me making on this forum!

Thanks for all the great stuff I've learned and keep up the good work folks!

[Sean_Goddard]

Re: The PI I am working on. I need to clarify a few points.

Dave Emery kindly helped me improve my design after I sent him the
schematics for his opinion. I mentioned Dave in my previous post so as to
give him the credit for this. I should have explained that this was the
only involvement Dave had and that we are not working together on any kind
of a commercial venture (I didn't proof read what I had typed and implied something which was NOT true).

Dave kindly helped me even though he was fully aware that I plan(ned) to sell my detectors in one way or another. One last thing, my opinion of Eric Foster's detectors is mine alone and not Dave's, it was meant in fun and in NO WAY was it a slight at Eric's incredible achievements over the years which FAR outweight anything I have done.

SORRY ERIC, no offence meant .

You can also note that against Daves advice I made further modifications to his suggested layout (which I have since lost as the only mods which existed were Dave's mark-up on MY design) and managed to burn up a whole batch of FETs. Depth is off the map, but the FETs just keep blowing after 20 hours or so. I've tried backing off various things which SHOULD, in theory, solve the probelm, but nothing seems to be helping, not even changing the FET for one with different parameters. I've NO IDEA what is causing the problem.

Dave has rightly (after reading my comments) refused to help further so I'm not sure I will be doing ANYTHING with this design.

Sorry if I gave people the impression that something was happening which was NOT! And an even BIGGER apology to Dave for implying that you were involved with a "project" when you were not.

[Qiaozhi]

I have put the simplest form of an AC analysis (TX -> Target -> RX). Regardless of small signal (1 V) or full signal (battery voltage), BTW this is totally irrelevant in the provided simple spice simulation, what can we obtain from it?

The use of a 1V signal does not imply "small-signal". Since the models are linearized around the operating point, prior to the AC simulation starting, means that only a "small-signal" variation will yield correct results. Any non-linear effects that may occur in practice are removed by the simplified models. In other words, some circuits will give different results during an AC simulation for different operating points. The "trick" of setting the AC amplitude to unity, is simply used to display the gain of the circuit directly in the Bode Plot. In fact, 1V may be an excessively high input voltage for some circuits, but this does not affect the results of an AC analysis.

Just thought I need to make this clear.

[Aziz]

Ok guys,

I have put the simplest form of an AC analysis (TX -> Target -> RX). Regardless of small signal (1 V) or full signal (battery voltage), BTW this is totally irrelevant in the provided simple spice simulation, what can we obtain from it?

Is there a free lunch for us?


Comeon, feel free to comment it, what you see. We will get very valuable information from it.

Aziz

[Qiaozhi]

One important point, when using SPICE's AC analysis, is that it's a "small-signal analysis". This is why you can set the voltage source parameters (AC amplitude and AC Phase) as 1 and 0, so that the Bode Plot shows the gain of the circuit directly. You need to be aware that the SPICE models are first linearized around the operating point before the simulation starts. In many cases there can be a huge difference between the large-signal and small signal results. It may not cause any problems in this instance, but it's something to bear in mind.

By the way, whatever independent source you use in an AC simulation, the frequency you define on that source (SINE, for example) is not used. Instead, the simulation uses the frequencies defined in the .AC statement.

[Aziz]

Yeah, I probably missed your point. Maybe you are saying: do an FFT on the response signal and look at that for patterns? How do we sample enough points to do that, isn't it tough to get even a few, or not? What are you proposing?

-SB

Hi SB,

no I don't say do the FFT on the response signal. I'm just saying look at the AC response in the posting ( http://www.geotech1.com/forums/showp...&postcount=221 ).
I'm just saying to do more AC response spice simulations. It answers some fundamental (but interesting) questions.

So, what's the simple graph is telling us?

Put the following command line somewhere and change the transmitter inductance (don't remove the L=300µ command):

.step param L list 150µ 300µ 600µ

And look what happens.

Report all here, what you have found.

Cheers,
Aziz

[Midas]

The Drain-Gate shoot through with real mosfets, when using high A coil currents, is exactly the problem that I have. To me, used to real circuits, the simulations look quite good.
Of course, I still have a lot to learn about simulations, but most of the time when there are differences, I can find the cause in my own mistakes.

Tinkerer

Lol yeah I'm sure I've got ten times as much to learn. I think you might be right. It may be a better model as it is. Its still seems too sensitive to DG shoot through but the alternative doesn't respond at all, which is worse. I've also lost confidence that's its got anything to do with the ESR of the parasitic capacitance having made a discrete model also with no ESR in the capacitance that just as insensitive.

Anyway I'm starting to realize the danger of simulations... you can spend hours tweaking them never really learning anything with confidence.

However, we have to be careful about making conclusions based on tweaking models. Although tweaking some ESR into the inductor model may make our simulation look better, it doesn't prove we are tweaking the right component. Maybe the inductor is working quite close to a real one, but our MOSFET model is too simple and we see an effect that we wrongly blame on the inductor. Or perhaps we need to put a little inductance and resistance into our circuit wires.

Very true statement. The lack of ESR may not be the problem at all.. I honestly have no idea anymore.

[simonbaker]

Hi SB,

the simple but fundamental AC analysis tells a lot. You should think about it.
BTW, you all should do more AC analysis rather than transient analysis.

Cheers,
Aziz

Yeah, I probably missed your point. Maybe you are saying: do an FFT on the response signal and look at that for patterns? How do we sample enough points to do that, isn't it tough to get even a few, or not? What are you proposing?

-SB

[Aziz]

Don't forget Aziz, the information in the TX pulse is not what we are trying to recover. We're really trying to estimate the system parameters defining a target. The reponse has that information even though the bandwidth is radically different from the TX pulse.

We are ringing a bell and trying to figure out what kind of bell it is. We don't expect the sound of the bell to be anything like the hammer blow we hit it with. But it will hopefully tell us all about the bell.

Cheers,

-SB

Hi SB,

the simple but fundamental AC analysis tells a lot. You should think about it.
BTW, you all should do more AC analysis rather than transient analysis.

Cheers,
Aziz

[simonbaker]

Hi guys,

the Dark Pulse Theory™ ( nice name - isn't it? ) states, that there is no free lunch for you! I'm so sorry for this bad news. But it is very overdue now.

The EM pulse has a spectral energy, which is "seen" by the inductively coupled target and is responding to this pulse accordingly. The TX -> Target -> RX system has a frequency response as well (well, we can see it as a multi-stage filter response).

The nature offers you the maximum possible available information by physics law. All you can do is making either worse or loosing/throwing away that amount of available information.

You can easily see the proof of this statement by making a simple AC analysis (frequency response analysis) for different time constant (TC) targets. I have made this for four different TC targets for you. For your convenience, I'll put the spice file and you can play with it.

Now looking forward to the interesting discussion.

Cheers,
The Dark Aziz

Don't forget Aziz, the information in the TX pulse is not what we are trying to recover. We're really trying to estimate the system parameters defining a target. The reponse has that information even though the bandwidth is radically different from the TX pulse.

We are ringing a bell and trying to figure out what kind of bell it is. We don't expect the sound of the bell to be anything like the hammer blow we hit it with. But it will hopefully tell us all about the bell.

Cheers,

-SB

[simonbaker]

Indeed, I wouldn't have expected it to be so fundamentally flawed either. Thinking about it a bit more the neglected resistance isn't so small either, since the parasitic capacitance is coupling to the mosfet drain through the coil it has go through at least some of the coils resistance. Also there should be resistance between the coil and its parasitic capacitance, otherwise the coil could potentially oscillate forever, which of course isn't possible.

With those thoughts in mind I made the following discrete model. Its hard to get your head around how much resistance there should be between all the connect points for a 15 ohm coil but its a start. It still has some weird current spike during switch on but the switch off is a lot better. Also removing the damping resistor doesn't result in the kind of oscillations I would expect so its obviously still not perfect.

Edit: Yeah actually.. its electrically identical to one of Simons earlier post. Lol I think we can burn this extra wheel.. never mind its been fun.
The question over what proportion of the total resistance is taken by each component remains though. It may not be 50/50.

That's an interesting point about a pure capacitor in parallel with a pure coil with no series or parallel resistance causing infinite oscillations, if the model does in fact allow that. I have not been able to actually get LTSpice to do that. Can you come up with a case, using the LTSpice inductor, setting all the parasitic resistances to zero and some finite parasitic capacitance?

General thoughts:

It may be true that our PI coils with their unique flyback pulse are a special case where the LTSpice inductor model is too simple. In switching power supplies, I guess there are usually some hefty capacitors nearby that would mask the lack of ESR in the coil parasitic capacitor.

However, we have to be careful about making conclusions based on tweaking models. Although tweaking some ESR into the inductor model may make our simulation look better, it doesn't prove we are tweaking the right component. Maybe the inductor is working quite close to a real one, but our MOSFET model is too simple and we see an effect that we wrongly blame on the inductor. Or perhaps we need to put a little inductance and resistance into our circuit wires.

I see your model with what looks like 30 ohms of ESR in the parasitic capacitance. That may fix the simulation, but it just doesn't seem realistic to me. However, I really don't know. I guess my physical interpretation of the parasitic capacitance would be a current that jumps directly across from winding to winding through the insulation and does not really travel down the wire like the inductor current. It is plausible to me that the capacitive current could have extremly low ESR.

But I just don't know. It would be great if we could actually set up a test rig to measure these parameters directly, but given the tiny parameters, how would we do that? Even oscilloscope probes would probably be a big factor.

In any case, we may in fact need a more realistic model for the inductor for PI simulations, and it would be good to have one. Your model looks like a good start, and I believe there is a way to package it as a component we can easily use in LTSpice -- I hope.

I still wonder if there is some virtually pure capacitance in parallel with the coil that should be modeled also. Hard to know when to stop with a lumped approximation of a distributed real-life component.

Regards,

-SB