STM32G0 oscilloscope pen

An upgraded version based on [STC Oscilloscope Pen]

The STM32G030F6P6 microcontroller is used, which supports dual-channel adjustable frequency and adjustable duty cycle PWM output, as well as ADC dual-pass input, and the microcontroller is overclocked to a main frequency of 100M

The board can also be integrated into other modules for collection

Display video [link] , open source warehouse [GitHub]

The current functions are:

AD input voltage 0-3.3V

12bits@5Mhz

1-2-5 time base step from 200ns to 500us

Vertical magnification 1-2-5-10 times

Three triggering methods, 10 trigger levels from 0-90%

Waveform measurement function

1-2-5 time base switching PWM output frequency from 10MHz to 200Hz

10% step switching PWM output duty cycle from 0% to 100%

3 independent buttons for multi-level menu switching

1 programmable LED light

USB serial port communication is reserved, but I can't get it out.

----------

BOM is not trustworthy, please go to X letter to buy similar packaged components (referring to connectors)

----------

The code is being organized and is not open source yet. Currently, only firmware downloads are available. It is open source. See above.

If a bug occurs during operation, please send a message via Lichuang Kaiyuan/b site private message/comment area or email. I will modify the program and update it in time.

If you need more cooperation, please only specify the email address  Defenver@foxmail.com

------------Explanation session----------

The principle is actually not difficult. Just use DMA+ADC+TIM to cyclically sample, and then transfer the data to the kernel for processing.

According to the actual operation, it is found that the ADC clock seems to be overclocked. It only needs to ensure that the AXI bus clock is aligned. Therefore, it is overclocked to the 100M ADC clock and the sampling delay is appropriately increased, so that the ADC sampling rate reaches the terrifying 5M (TSMC yyds)

The timer uses the TRGO of TIM1 and is connected to the ADC as the clock trigger source. Then the DMA transfers the data to the memory for subsequent processing. You can see the related tutorials of Avnet here to understand the principle.

There is also a PWM output. The principle of this thing is very simple and I won’t explain it. You can refer to Avnet’s tutorial to understand the principle. Generally speaking, compared with other series, G0 has a simpler adjustment method in this aspect.

Next are other peripherals. The previous project used a 0.96-inch screen. Although it was usable, it didn't display much, so I changed it to a 1.14-inch screen here.

The original button felt too bad, so I replaced it with a pulsator switch, and the feel is indeed much better.

The serial port is broken. It seems to be a chip problem. Let's wait for the successor to fix it (

Finally, there is the software part. The software part is not much different from the previous two oscilloscopes. Basically, everything is thrown into while(1) for processing.

The page display part uses the classic point drawing function, the trigger point part uses the ternary operator to judge in order to save code size, and finally the page switching part uses the switch statement with key values ​​to switch options, etc.

The code and ideas of this project are roughly the same

See the actual picture below

----------