LIS25BA can record, but there is "interference"

littleshrimp

LIS25BA can record, but there is "interference" [Copy link]

 

The operating voltage of LIS25BA is 1.8V. With so many development boards available, it is really difficult to find a 1.8V debugging board.

Because we need to use TMD communication, the common microcontroller does not have this function

At the beginning, I chose a Cypress PSoC6 because the board can be directly set to 1.8V power supply. After tinkering with it for a day, I found that its TDM function only supports 32-bit slots.

Later, I changed the debugger to 1.8V on the NUCLEO board, but the USB could not communicate. After being reminded by a kind netizen, I read the manual and found out that the USB needs a voltage of 3.0V or above.

Later, I re-soldered several jumpers of IKS01A3 and finally got the 3.3V<->1.8V lines of SCL, SDA, MCLK, WS, SCLK, and SD together.

After debugging, it is now possible to record, but there are some irregular error data in the normal data

When it is still, it looks like the following effect

This is the effect when using audio software

Put the sensor on your throat and record

The sound can be heard clearly, but the problem is that I don't know what causes this glitch.

2.mp3 (30.26 KB, downloads: 6)

2019-11-19 21:56 上传

点击文件名下载附件

The VDD noise of the sensor measured with an oscilloscope is quite large. The problem does not seem to be caused by this. It seems more like it is caused by timing or misalignment.

When monitoring the MCU data, you will often see single data that is about twice the normal data, and you cannot find the reason

In order to facilitate the observation of STM32's SAI, only the Z-axis data is enabled, and the other configurations are as follows

cruelfox

You flying line...

littleshrimp

cruelfox posted on 2019-11-20 09:51 Your fly line...

It seems to be the result of the flying wire. I reorganized it and used tape to stick it on my throat to record. Sometimes the effect is very good.

4.mp3 (39.12 KB, downloads: 4)

2019-11-20 10:05 上传

点击文件名下载附件

led2015

Glitches are always caused by various interferences. In addition to flying wires, are there other interferences?

littleshrimp

led2015 published on 2019-11-20 21:28 Glitches are always caused by various interferences. In addition to flying wires, are there other interferences?

Flying wires should be the main reason. After reorganization, if you don't touch the DuPont wire bureau with your hands, there will be few burrs

led2015

littleshrimp posted on 2019-11-20 21:33 The flying wire should be the main reason. After reorganizing, if you don’t touch the DuPont wire bureau with your hands, there will be few burrs

Indeed, I see that your use of the line is so messy, which is also a kind of helplessness. So is the waveform perfect now?

littleshrimp

led2015 Published on 2019-11-20 21:35 Indeed, I think your use of the line is so messy, which is also a kind of helplessness. So is the waveform perfect now?

Almost perfect

https://bbs.eeworld.com.cn/thread-1099176-1-1.html

yujie2510

If I tie the wires like this and don't touch them, will it always be perfect? Or is it a little messy from time to time? Is it related to the signal wires not being isolated? It seems that the LIS25BA has high-frequency signals

littleshrimp

yujie2510 posted on 2019-12-16 13:39 If the wires are tied like this, and you don't touch them with your hands, will they always be perfect? Or is it a little messy from time to time? Is it related to the lack of isolation of the signal line? It seems that LIS25BA has high-frequency signal...

Basically, it is the problem of Dupont line. The interference caused by the long line