felezjoo PI(the best pulse induction metal detector that I made until now)

[Searchy]

Would you mind me asking which schematic and project you made ?, and was it error free ?, as some people have had problems with error's, thanks.

I used the clonePI AVR schematic from this webpage:

https://montazhtv.ru/de/metalloiskatel-impulsnyi-clone-pi-avr-otzyvy-blog-ot-said-t-vystavlyaem-minimalnoe/

On this webpage you can find a ZIP-file called "LINKS-3037.zip" with a complete clonePI collection (scm, firmware versions etc.) as download.

https://320volt.com/en/mikrodenetleyici-kontrollu-metal-dedektoru-projeleri/

I am using the latest firmware version 1.8.1 and did not notice any issue yet.

[6666]

Yes, works well.

Would you mind me asking which schematic and project you made ?, and was it error free ?, as some people have had problems with error's, thanks.

[Searchy]

Did you build a working clonePI ?

Yes, works well.

[6666]

The frontend of the clonePI is well thought through.
First stage acts as differential low-gain amplifier who's input and output ranges are well within the opams specs at any time.
Second stage is a pure gain stage which is further switchabel between 1 and 57 to prevent opamp saturation during TX decay.
What could be improved is the automatic offset compensation, which is done with a simple active low-pass filter.

Did you build a working clonePI ?

[Searchy]

What would you suggest to replace it ?

The frontend of the clonePI is well thought through.
First stage acts as differential low-gain amplifier who's input and output ranges are well within the opams specs at any time.
Second stage is a pure gain stage which is further switchabel between 1 and 57 to prevent opamp saturation during TX decay.
What could be improved is the automatic offset compensation, which is done with a simple active low-pass filter.

[Altra]

Long time ago Dave Jones has posted on youtube great video about "Cascading Opamps For Increased Bandwidth"

Great video, Thanks

https://www.youtube.com/watch?v=ZvT9hHG17tQ

[6666]

Long time ago Dave Jones has posted on youtube great video about "Cascading Opamps For Increased Bandwidth"

Interesting videos

[6666]

That's exactly what i was talking about.
The way the opamp is used (misused) is not the way it's properly done.
It's a totally unstable/unpredictable walk at the limits of the silicon/input rails offsets etc.
It's simply a bad design and this also reflects in the circumstance, that you have to follow all the values as precisely as possible to hopefully get it work (and best get a opamp produced in the same week in the same year ).
Really... the opamp stage is a nightmare.

What would you suggest to replace it ?

[Vadim18]

Long time ago Dave Jones has posted on youtube great video about "Cascading Opamps For Increased Bandwidth"

[Qiaozhi]

I haven't heard anyone suggest replacing the single-stage preamp with a dual-stage, as in the Minipulse Plus. That might be a workaround for the "almost impossible to get" LM357, and bearing in mind that any LF357s that are available appear to be fakes.

[Searchy]

That's exactly what i was talking about.
The way the opamp is used (misused) is not the way it's properly done.
It's a totally unstable/unpredictable walk at the limits of the silicon/input rails offsets etc.
It's simply a bad design and this also reflects in the circumstance, that you have to follow all the values as precisely as possible to hopefully get it work (and best get a opamp produced in the same week in the same year ).
Really... the opamp stage is a nightmare.

[ivconic]

Does anyone have an alternative to the LF356 working?
And is there nobody except me who thinks this OPAMP "design" is an absolute nightmare!?!?

No, is not.
Use LF357. Preferable "H" version.
Obey everything on original schematic traced by Qiaozhi.
Make proper coil.
No nightmare at all.
Everything is well analyzed and elaborated in first dozen pages on this topic.
Just read and obey.
While chasing suggested debug numbers (by author); one must understand wider context of this design and how things usually operate in PI technology.
If one don't understand common PI working principles; is not recommendable to start "career" with this project.
Debug numbers are suggested. And is good to have close values to those suggested.
But keep in mind that literally EVERY component in RX frontend will dramatically affect those numbers, starting from coil, damper resistor and all other components in chain ending with opamp input.
It's not just about opamp. And in most of the cases is not about opamp at all.
In most of the cases is about coil. Than chain of components until opamp input.
For example; changing 100 ohms resistor value to say.. 180 ohms; will dramatically affect debug numbers. Or changing tantalum 2u2 value will affect debug numbers even more.
Changing damper and/or pass through 390 ohms resistors will also dramatically affect debug numbers.
Etc.
It is complete signal path and wider context, not just one single component to suspect it.
When lacking of proper value, quality and tolerance of one component in chain; one can try to compensate that by little bit of trimming other components in chain.
ALL THE TIME with coil far away from any metal, noise source and any EM field, of course.
Keep in mind that debug numbers are not fixed and not suppose to be fixed.
Author gave suggested values with coil attached but with no signal of detection at all.

[Searchy]

Does anyone have an alternative to the LF356 working?
And is there nobody except me who thinks this OPAMP "design" is an absolute nightmare!?!?

[chuubaba]

Someone suggested the use of 5k 2w potentiometer instead of damping resistor and posted schematic illustrating the remedy. Practically do we remove both damping resistors or R11 only?

[gustavodigitales]

Hi . does anyone know if the latest version has the counter active ?