$10 O-Scope revisited

What are you developing?
mausi_mick
Posts: 147
Joined: Fri Aug 12, 2016 1:40 pm

Re: $10 O-Scope revisited

Post by mausi_mick » Thu Feb 23, 2017 8:24 am

Hi ThunderOwl, hi Rexnanet,

the different Pins could be a problem, I have changed it partly for shorter wires, or save pins (e.g. for restart PIN on 3.3V).

Another problem is that not all pins of the STM32F103 chip are connected to the board, neither on the blue pill nor on maple mini:
on maple mini for instance their is no PIN for PB9, which needs Rexnanet for the encoder-buttom.

I try to test his(Rexnanat) program for the MCP6S92 ... and I hope, that it would soon go on with the second channel ...

User avatar
ThunderOwl
Posts: 135
Joined: Fri Oct 28, 2016 6:04 am
Location: Riga, Latvia
Contact:

Re: $10 O-Scope revisited

Post by ThunderOwl » Thu Feb 23, 2017 9:00 am

I see, PB9 on Mapple-type boards is used to "USB enable" switching. So, I guess, resistor on board can be found where to solder. Anyway, but, how about you 3 "main developers" for this moment :), come together and agree on unified connections? :) Then we will have "hardware platform" with easily exchangable software. I think Rexnanet has a good point of his version of TFT wires, to have that as a basis, probably is possible to agree on encoder inputs, generator (optional) control pins, amplification (for MPC) control pins?
In every man's childhood, first 40 years are the hardest

User avatar
rexnanet
Posts: 205
Joined: Wed Mar 16, 2016 10:34 am

Re: $10 O-Scope revisited

Post by rexnanet » Thu Feb 23, 2017 9:37 am

ThunderOwl wrote:I see, PB9 on Mapple-type boards is used to "USB enable" switching. So, I guess, resistor on board can be found where to solder. Anyway, but, how about you 3 "main developers" for this moment :), come together and agree on unified connections? :) Then we will have "hardware platform" with easily exchangable software. I think Rexnanet has a good point of his version of TFT wires, to have that as a basis, probably is possible to agree on encoder inputs, generator (optional) control pins, amplification (for MPC) control pins?
I think its a feasible solution :)
The pins I choose are mainly to make it easier to connect the wires (dupont, sequencial if possible) as I still don't have it on a board.
Encoder pins are very easy to change. No special functions here, only interrupts.
MCP are also easy, general IO.
And also we could separate a bit the code functions into files. I've made that with the MCP6S92 library. That makes it easier to apply changes.
Maybe the graphical functions could be separated also... have to check if its better or not.

I've only made some preliminar tests with 2 MCP's in series (to get up to 1024x gain) but wasn't working yet and didn't had time to debug then. Will work on that ASAP.
2 channel is not forgotten :) First: get 2MCP's in series then resize the quadricule to 300x200 to make it easier on the display math :)

P.S. I only use the Blue Pill :)

P.S.2. I just saw that AD9851 is not in a separate lib, maybe that could be done... mausi_mick?

mausi_mick
Posts: 147
Joined: Fri Aug 12, 2016 1:40 pm

Re: $10 O-Scope revisited

Post by mausi_mick » Thu Feb 23, 2017 9:56 am

Rexnanet, ThunderOwl,

it's possible to change the Landscape -format to Portrait,
than we could display the Signal with full 8 bit resolution (256) in the high and had place for scope-settings (delay, trigger, ...) in a own in the 84 x 240 field on the top or on the ground ?

User avatar
rexnanet
Posts: 205
Joined: Wed Mar 16, 2016 10:34 am

Re: $10 O-Scope revisited

Post by rexnanet » Thu Feb 23, 2017 1:34 pm

mausi_mick wrote:Rexnanet, ThunderOwl,

it's possible to change the Landscape -format to Portrait,
than we could display the Signal with full 8 bit resolution (256) in the high and had place for scope-settings (delay, trigger, ...) in a own in the 84 x 240 field on the top or on the ground ?
You could do that but it would be a bit narrow... the scopes usually have 10x8 quadricules or 12x8 in a wide screen. That approach would be a bit visually odd imo.
Also the signal has 12bits but can be trimmed down.

With the 300x200 for the quadricule we get an aditional space on the bottom that might be used for the AD9851 settings.

mausi_mick
Posts: 147
Joined: Fri Aug 12, 2016 1:40 pm

Re: $10 O-Scope revisited

Post by mausi_mick » Thu Feb 23, 2017 3:01 pm

Rexnanet,

I think with the 8 bit it's difficult too, because the MCP6.. has only 1% accuracy and you have to select other components to better than 0,5 %.

And I think better than 7bit it's for hobbyist enough . :lol:

User avatar
rexnanet
Posts: 205
Joined: Wed Mar 16, 2016 10:34 am

Re: $10 O-Scope revisited

Post by rexnanet » Fri Feb 24, 2017 9:50 am

This pin issue can be solved by creating an #ifdef BLUEPILL_BOARD and #ifdef MAPLE_BOARD and place all pin assignments inside each of those defines.

This way we can simply set the variable we want and that's it :)

What do you think?

User avatar
rexnanet
Posts: 205
Joined: Wed Mar 16, 2016 10:34 am

Re: $10 O-Scope revisited

Post by rexnanet » Fri Feb 24, 2017 10:05 am

mausi_mick wrote: MCP6.. has only 1% accuracy
What are you referring to?

The resolution is affected by noise, not by the accuracy of the components.

User avatar
ThunderOwl
Posts: 135
Joined: Fri Oct 28, 2016 6:04 am
Location: Riga, Latvia
Contact:

Re: $10 O-Scope revisited

Post by ThunderOwl » Fri Feb 24, 2017 10:10 am

rexnanet wrote:This pin issue can be solved by creating an #ifdef BLUEPILL_BOARD and #ifdef MAPLE_BOARD and place all pin assignments inside each of those defines
That sounds smart to me. What others say? :) (who actually understand coding, unlike me :D okay, i understand just a little bit... :) )
In every man's childhood, first 40 years are the hardest

User avatar
zoomx
Posts: 523
Joined: Mon Apr 27, 2015 2:28 pm
Location: Mt.Etna, Italy

Re: $10 O-Scope revisited

Post by zoomx » Fri Feb 24, 2017 11:38 am

ThunderOwl wrote: That sounds smart to me. What others say? :) (who actually understand coding, unlike me :D okay, i understand just a little bit... :) )
This is an automatic way to change code before it is compiled.

#ifdef BLUEPILL_BOARD mean that if in the code there is a #define BLUEPILL_BOARD, all the subsequent lines are added to the code to be compiled until #endif. If there is not the #define BLUEPILL_BOARD, the lines will be removed from the code.

#define BLUEPILL_BOARD and #define MAPLE_BOARD must not be present at the same time.

Post Reply