STC8A8K64 MCU ADC input is disturbed

elec32156

STC8A8K64 MCU ADC input is disturbed [Copy link]

 

The ADC port of the STC8A8K64S4A12 chip is used to collect the voltage of the linear Hall. The default output voltage of the Hall is about 1.65V. The actual test found that there was spike pulse voltage interference in the ADC input, with an amplitude of about 400mv and an interference level duration of about 3us (as shown in Figure 1). The ADC sampling trigger threshold is set to 12mv, which is easy to be triggered by mistake, causing the device to not work properly. It is not clear what caused this interference;

The following verification tests have been performed, but the cause of the problem has not been found:
1) The ADC pin is left floating, and the measured pin has high and low level pulse jumps (as shown in Figure 2). I don’t know if this level jump caused the spike pulse after the 1.65V was connected; the ADC input was disconnected, and the pin was connected to the ground with a 10K resistor alone. The measured voltage also fluctuated, and occasionally there were spikes, as shown in Figure 3

2) This chip has a total of 15 ADCs available, and only some of the outputs have interference. As long as the pin has interference, there will be such high and low level jump pulses when measuring its floating state, but the waveforms output by different pins will be slightly different;
3) A 1K resistor is connected in series between the Hall output and the ADC, and 100NF is connected in parallel to the ground. It is found that the interference spikes are gone, but the waveform becomes Figure 4, and occasionally there will be a drop of about 100mv.

4) Changing the VCC input filter capacitor of STC8 from 0.1UF to 10UF+0.1UF, no improvement; changing the magnetic bead between VCC and AVCC to a 22R resistor, no improvement; the schematic diagram of the power supply is shown in the figure

5) Considering whether there is any other circuit interference, all other devices on the board were removed, leaving only the microcontroller and the peripheral power supply resistors and capacitors, and the power supply was connected from the VCC pin filter capacitor to the 3.3V stabilized power supply with a flying wire. The phenomenon still exists.

I would also like to ask teachers and seniors to give me some advice, thank you very much!

宋元浩

You can try changing the board or the MCU.

whuer

The hardware is not well designed

The analog input channel can add an RC low-pass filter

freebsder

Try adding filtering

huo_hu

Does your interference signal have any periodicity? Try to program the microcontroller to do nothing to see if it is a software problem.

Num1

The signal doesn't seem to have any periodicity. I wonder if there is something wrong with your program.

elec32156

Song Yuanhao posted on 2020-8-7 21:27 You can try changing the board or the MCU

I have replaced multiple boards and multiple MCUs, and the phenomenon is the same.

elec32156

whuer posted on 2020-8-7 22:38 If the hardware is not well designed for the analog input channel, you can add an RC low-pass filter

My third experiment added an RC low-pass filter, but it was still not a stable input voltage. Occasionally, there would be a voltage drop of about 2ms. I don’t know where this came from.

elec32156

huo_hu posted on 2020-8-8 01:51 Does your interference signal have any periodicity? You can write a program that does nothing on the microcontroller to see if it is a software problem

This has not been verified yet, thanks for the reminder, I will verify it.

chunyang

Use two channels to monitor the ADC input and power supply simultaneously. Interference may come from power supply conduction.

damiaa

chunyang published on 2020-8-10 19:17 Use two channels to monitor the ADC input and power supply at the same time. The interference may come from the power supply conduction.

Most likely from the power supply

foxpro2005

Try to power the board with batteries. Also:

1. MCU analog AVCC power supply, plus π-type filtering

2.ADC pin input plus RC filtering

3.ADC plus software filtering

These are the most basic methods

吴下阿蒙

Keep going, I look forward to the next analysis.

hfe

Did you solve it in the end? I also encountered the same problem. When the external circuit is at a specific impedance and a specific voltage, the sampling value = 0 appears irregularly.

hfe

I called STC today and was told to connect a 103 capacitor to the ground. It was tested and it worked. The conversion result no longer showed 0. I will continue debugging. USB conversion will also be affected. Changing a USB conversion tool may have unexpected results.