AMX TX

[Tinkerer]

Sure, not all applications require that speed but a PI detector does if you're after fast decaying targets.

The right choice of mosfet and mosfet drive is primordial when we want timing precision in the 10ns of nanoseconds.
Many components need to be chosen with great care.
Many, many datasheets need to be studied and understood.

[ivconic]

From the beginning of the first posts on these topics, on the subject of the AMX project, until today; I read maybe only 40% of what was written.
And that was enough for me to understand that there is no desire for simple solutions here, the ambitions are to create a detector that is unique in many ways.
Of course I agree with that attitude and of course I support such ambitions. Because the forum (and the market) is full of mediocre detectors.
I think the time has come to join forces on the forum and project something different, special and valuable.
Carl has already gone out of his way to complicate things a lot, for those of us who can't afford most of the modern components, much less work with them.
Although I criticized such things; that doesn't mean I don't support the project.
Because as I get to know the details better and as I slowly understand more and more how what works;
I realize that Carl's "megalomaniac over-ambition" is not really that, but a very realistic approach to problem solving.
In the last few days I have read many pages on the internet, trying to find alternative solutions, to show that it can be done in a different way... alas...
the more I read and understand; it is more clear to me that there is a very limited space and that there is no room for improvisations.
In particular, today I reviewed dozens of pages on the topic of drivers for those fets.
I analyzed from IR2103 to IRS2004. IRS2004 even has shoot-through prevention and "safe" dead time.
Alas... power supply is high (10v min.) for this purpose intended here.
I wanted to use 2x IRS2004 and only 4 mosfets. 2 pins from Atmega to control the whole switching.
But the IRS2004 supply is 10V and mosfets can't be supplied with only 2 volts.
etc.
These days I checked every possible aspect of Carl's schematic and tried to simplify it and adapt it for "poor mans solution".
Always something pops out as new obstacle.
So... finally I am giving up from such attempts.
As for audio; audio shall be "special" too, since the rest of detector will be also special and unique.
If it is not already saved on other things; why should you save on audio?
In words, descriptively, I have described how I would like the audio to be projected and how it should work.
And I think that's a very good idea. But if the majority does not accept it; then do a simple "bzzzzz" or "meow-meow"... I don't really care now.
I have wasted too much precious time on all these empty stories. Outside, the weather has improved, it's sunny, I have so many smarter responsibilities.
I'd rather be doing something useful and saving up to buy Deus II instead of wasting my time here.
I wish you success in this project. Only when you have a fully working version on your desk; let me know to look!
Because I can't contribute otherwise, except to describe what I like to see on the detector and not to write anything else that bothers others.
Carl, once you have the final version working, let me know the price of the finished pcb, I might decide to order one.

[moodz]

From the beginning of the first posts on these topics, on the subject of the AMX project, until today; I read maybe only 40% of what was written.
And that was enough for me to understand that there is no desire for simple solutions here, the ambitions are to create a detector that is unique in many ways.
Of course I agree with that attitude and of course I support such ambitions. Because the forum (and the market) is full of mediocre detectors.
I think the time has come to join forces on the forum and project something different, special and valuable.
Carl has already gone out of his way to complicate things a lot, for those of us who can't afford most of the modern components, much less work with them.
Although I criticized such things; that doesn't mean I don't support the project.
Because as I get to know the details better and as I slowly understand more and more how what works;
I realize that Carl's "megalomaniac over-ambition" is not really that, but a very realistic approach to problem solving.
In the last few days I have read many pages on the internet, trying to find alternative solutions, to show that it can be done in a different way... alas...
the more I read and understand; it is more clear to me that there is a very limited space and that there is no room for improvisations.
In particular, today I reviewed dozens of pages on the topic of drivers for those fets.
I analyzed from IR2103 to IRS2004. IRS2004 even has shoot-through prevention and "safe" dead time.
Alas... power supply is high (10v min.) for this purpose intended here.
I wanted to use 2x IRS2004 and only 4 mosfets. 2 pins from Atmega to control the whole switching.
But the IRS2004 supply is 10V and mosfets can't be supplied with only 2 volts.
etc.
These days I checked every possible aspect of Carl's schematic and tried to simplify it and adapt it for "poor mans solution".
Always something pops out as new obstacle.
So... finally I am giving up from such attempts.
As for audio; audio shall be "special" too, since the rest of detector will be also special and unique.
If it is not already saved on other things; why should you save on audio?
In words, descriptively, I have described how I would like the audio to be projected and how it should work.
And I think that's a very good idea. But if the majority does not accept it; then do a simple "bzzzzz" or "meow-meow"... I don't really care now.
I have wasted too much precious time on all these empty stories. Outside, the weather has improved, it's sunny, I have so many smarter responsibilities.
I'd rather be doing something useful and saving up to buy Deus II instead of wasting my time here.
I wish you success in this project. Only when you have a fully working version on your desk; let me know to look!
Because I can't contribute otherwise, except to describe what I like to see on the detector and not to write anything else that bothers others.
Carl, once you have the final version working, let me know the price of the finished pcb, I might decide to order one.

You only need 2 volts for the tx circuit the drivers can run from 10 volts no problem.
you don't have to use fancy parts I already showed that. The Tilt control can be done with a analogue loop and if you sample and integrate you don't need fast ADC.
Of course if we want to investigate just how good this detector can be you have to lift the bar parts wise and Carl is doing this. Carl's approach will allow more flexibility regarding DSP and Audio.
There is no reason why in the meantime you can't test some of the concepts ... If you want

[green]

To generate audio you need a target signal. We haven't solved all the issues with generating the target signal so it's not a priority to resolve the audio. A CPU is programmable and can generate any audio scheme you like so it's a non problem without a target signal.

(target signal) Auto-Mation-Assist has been testing his AGD analog detector using a .23grain nugget. Are we trying to detect a smaller nugget?

[moodz]

(target signal) Auto-Mation-Assist has been testing his AGD analog detector using a .23grain nugget. Are we trying to detect a smaller nugget?

Small nuggets are more dependant on the coil size and timings in my opinion and I guess we are trying to detect any nuggets.

[Carl-NC]

(target signal) Auto-Mation-Assist has been testing his AGD analog detector using a .23grain nugget. Are we trying to detect a smaller nugget?

On a Good Day the SDC2300 will detect a 1/2 grain nugget. That's my personal goal: as good or better than the best on the market. A good VLF (like the Goldbug 2) can probably detect 0.1 grains, but won't work in harsh ground. I think that's a bit too much. If gold is $1920/ounce (it's actually higher today) then that's $4/grain, assuming 24k purity. $2 nuggets have appeal, maybe even $1 nuggets. At 40¢ and almost invisible, it's less appealing.

[ivconic]

Have I made any mistake here?
Of course ; dead time will be deifned as pause between Pin x and Pin y in code.
But in essence this should work... or not?

[Carl-NC]

What you have drawn is a VLF transmitter. There are no flyback-blocking elements so the TX current will always be a ramp. You either need blocking diodes on Q7,Q9 or you need to add the flyback switches on either side of L1.

[ivconic]

Yes, but that's a side issue right now, I brought it into Proteus and have much bigger headaches at this moment.

I'll redo it... this was just a rehearsal.
Me and simulator!? Can you believe it!

[Carl-NC]

A day on the simulator is worth a week on the bench.

[ivconic]

Without blocking diodes:



With blocking diodes:

[ivconic]

Dead time...
20uS when both signals are at 50%. I get the impression that this is not a good solution.
What is the "damage" if the duty cycle is not 50% at both pulses?
It seems to me that the "safe zone" is wider when one of the pulses has a duty cycle slightly below 50% ... or am I wrong?

[ivconic]

I was intended to use IR2103 but this version of Proteus does not have definition for it.
(Murphy is killing me whole my life, IR2103 is the only one I can obtain at local suppliers, the only one missing from Proteus!!!)
It seems that those IR drivers differs a lot.
I expected dead time in scale of tens of nano seconds.
20uS doesnt sounds right to me?!
I will input now original Moodz schematic and see what happens...

[ivconic]

I got same 20uS as proper timing difference for dead time.
I can't measure the TX amplitude, it is way over the virtual oscilloscope range, I tried with attenuation, voltage splitter, first with 1M/9M and then 10M/90M, no results though.
Amplitude is still over the roof!
Also it was the doubt where to connect the bootstraped VS from the driver. It works this way though:

[ivconic]

Signal at coil is rock solid and super clean ... ONLY with both pulses set with 50% duty cycle.
But when I set the pulse width lower then 50%; at one or both the pulses, the same: signal at coil is "tilting" in a sequences!?
Carl... this will be the problem if you plan adjustable pulse width... or not?
Too many things still not clear to me...