New FelezjooPI o PulseMaster

[sabios]

I am working on a new version of the Felezjoo that includes some improvements. It is still in the testing phase—as you can see, it is in the early stages—but it is coming along well.

I have already conducted preliminary tests; the unit features switchable Ground Tracking and a numerical adjustment setting for use on different soil types.

It also now includes discrimination based on the standard Threshold method, as well as Virtual Conductivity discrimination, which utilizes seven sampling windows.

And, of course, it incorporates the best features of the original Felezjoo.

I plan to change the name, however, to avoid confusion with the Felezjoo.

[pigeon]

Very good job my friend

[haylaz0036]

You're amazing, Engineer; I'm following your work.

[Ruben2000]

I am working on a new version of the Felezjoo that includes some improvements. It is still in the testing phase—as you can see, it is in the early stages—but it is coming along well.

I have already conducted preliminary tests; the unit features switchable Ground Tracking and a numerical adjustment setting for use on different soil types.

It also now includes discrimination based on the standard Threshold method, as well as Virtual Conductivity discrimination, which utilizes seven sampling windows.

And, of course, it incorporates the best features of the original Felezjoo.

I plan to change the name, however, to avoid confusion with the Felezjoo.

Excellent work; you're all very wise!
I'll be following your progress; keep it up!

[Xtrem]

In terms of functionality, the software is flawless; it includes some extra features that aren't really necessary—specifically the contrast and brightness controls. There is one problem area, however: the audio output level cannot be adjusted. Everything else is fine. I am referring to version Full 3.35.
As for the hardware, there is plenty of room for improvement.That guy, Hamid, is talented; I hope he's doing well.

[sabios]

This is a demo of the new program I’m working on; the GT function is already working, along with other improvements I’ve implemented. I want you to test it on various types of terrain—I’m from Veracruz, Mexico, where the ground conditions differ, so the goal is to see how it performs in different environments. I still need to work on the discrimination feature, but otherwise, it seems to be functioning well; give it a try.

It works with the original Felezjoo board, so no additional modifications are required.

Finally, there are two different menu modes: the first is accessed by simply pressing "OK," while the second requires a long press (holding the button for about two seconds); the same long-press action is used to exit the menu.

One final detail is that a 20x4 LCD is used for this.

[pigeon]

OK dear brother and thank you so much

[sabios]

Sorry, I made a mistake; the version I uploaded is flawed. I had that version left over from when I uploaded the VolcanoPI program—and the Volcano itself is actually flawed, with much lower sensitivity. That’s why I set out to build one that works with the 20x4 display I already owned (since I originally bought the parts for the Volcano PI). I removed the 16MHz crystal requirement and stuck with the 20MHz one to avoid confusion, as it uses the same fuse settings.
Feel free to use it however you like, but keep in mind this is a demo version—though all features are unlocked.

If you'd like, I can create an operation manual, since it works completely differently from the Felezjoo and Volcano models. Alternatively, you can experiment with the settings yourself; this version is designed for people who are knowledgeable about the subject and want full access to all adjustments.

If you have any questions, just email me or message me on WhatsApp
at +55 296 105 38 52. Please remember that I speak Spanish, as I know very little English.

I almost forgot to ask: does anyone here know a way to get resources or donations so I can keep buying parts and experimenting? As you know, I had the idea of using real windows for discrimination, but that requires some money since it involves a lot of trial and error to get it just right. Right now, I'm in the red—meaning I'm out of money.

[sabios]

Could any of you help me out? I’m close to achieving great results, but I’m stuck on the discrimination aspect. I’m struggling with the ID values—it’s a bit confusing, and as I mentioned, I lack an oscilloscope to visualize the full sampling stages; the timing windows used for the Felezjoo method differ here, so the readings don't quite align.

I need the new sample data to calibrate the detector and achieve peak performance.

That said, I’m liking how it’s shaping up so far. I’ve already implemented low and high "Iron Reject" settings, which make it easy to filter out unwanted objects in the detection range. However, this feature is still in the early stages because, as I noted, I’m having trouble with that specific phase.

Right now, it’s capable of isolating coins, copper, and similar items while ignoring iron and other metals—including aluminum. But since it’s still in the early stages, I need to figure out a way to get the IDs to align correctly; that would complete the picture.

[samir3]

Good

[sabios]

I’ve just uploaded a new video to showcase the new program I’m finishing up; I should clarify that it’s still in the testing phase, but it’s coming along very well.

This version differs from the Felezjoo and VolcanoPI; it now includes improvements found in very few pulse induction detectors. It features a discrimination function with adjustable high and low settings, allowing you to lock in specific parameters if you’re targeting only certain types of items. However, as I mentioned, I’m still working out the kinks with the target IDs—I hope to have that sorted soon, and once that’s done, the program will be complete.

It also includes GT (Ground Tracking)—an automatic ground adjustment feature that adapts to changing terrain so you never miss a thing.


https://www.youtube.com/watch?v=g4xGaM6geII

[sabios]

I was working on this new version, but I accidentally dropped it, and now there’s an annoying problem: it powers up fine, but as soon as I connect the coil, the voltage drops from 12V to 4.94V. I’ve already checked the IRF840 and the 4148 diodes, and they’re fine; also, the 7812 regulator gets very hot when I connect the coil. What could the problem be?

[eduardo1979]

Check with an oscilloscope if you have a signal at the gate of the IRF840. Have you checked the BS170 transistor that's connected to the IRF840? The problem might be there.

[nickel_n]

I’ve just uploaded a new video to showcase the new program I’m finishing up; I should clarify that it’s still in the testing phase, but it’s coming along very well.
This version differs from the Felezjoo and VolcanoPI; it now includes improvements found in very few pulse induction detectors. It features a discrimination function with adjustable high and low settings, allowing you to lock in specific parameters if you’re targeting only certain types of items. However, as I mentioned, I’m still working out the kinks with the target IDs—I hope to have that sorted soon, and once that’s done, the program will be complete.
It also includes GT (Ground Tracking)—an automatic ground adjustment feature that adapts to changing terrain so you never miss a thing.
https://www.youtube.com/watch?v=g4xGaM6geII

Sabios just a thought.
You should name your version by a different name than “new version of the Felezjoo “
it will only create problems later on with builders that they will not be able to distinguish what is what.
Perhaps name it as Sabios-Felezjoo or what ever but please Don't use the same name.
Just my 2 Cents worth

[sabios]

Regarding the name, I’ve already changed it—in fact, I use "Felezjoo" or "PulseMaster" in the title, which I’ve already done—and finally, I fixed the error I was having; it was indeed the BS170.

And if you watch the video, you can see it’s called PulseMaster, and as I said, it’s a new system.