0
\$\begingroup\$

I made a small prototyping board where accelerometer data is read into a microprocessor, showing the data from the accelerometer after it is sampled, gives a very noisy signal. (50 mv peaks)

I had a similar problem before on a breadboard and switched to another breadboard/layout to resolve this but sadly didn't think any more of it. I think the problem is do to interference from the 5v power regulator or bluetooth module, however when using a oscilloscope I can't find the noisy signal anywhere on the board!

I would be greatly interested in any information which could lead me to find to source of the noise or solve it.

Edit1: The noise increases when the bluetooth module is sending data. So its almost certainly crosstalk or interference through gnd/vcc, any way to resolve this without resoldering?

Edit2: board layout http://imgur.com/bNLDdlz

\$\endgroup\$
5
  • \$\begingroup\$ Show the interface between the accelerometer and the and the micro. What is the input impedance of the ADC input on the micro? \$\endgroup\$
    – Andy aka
    May 22, 2015 at 23:32
  • \$\begingroup\$ @Andyaka I'm not sure how I can measure the input impedance while everything is soldered. For the interface maybe my board layout is of any use? \$\endgroup\$
    – YetAnotherStudent
    May 23, 2015 at 11:55
  • \$\begingroup\$ Read the data sheet of the ADC/micro \$\endgroup\$
    – Andy aka
    May 23, 2015 at 11:56
  • \$\begingroup\$ I'm using a PIC18F, nothing is mentioned about the input impedance in the data sheet (that I can find). The analog input model does suggest that I_leakage is around 100nA. Are you sure it could be related to the input impedance? Shouldn't there also have been a problem on every setup I made with these components before this one? \$\endgroup\$
    – YetAnotherStudent
    May 23, 2015 at 12:19
  • \$\begingroup\$ @Andyaka , forget to reference you in my previous answer. \$\endgroup\$
    – YetAnotherStudent
    May 23, 2015 at 12:31

1 Answer 1

Reset to default
0
\$\begingroup\$

The problem was a combination of an almost empty battery and a bad board layout. The ground of the Bluetooth module went to the the battery via the microprocessor which uses 20 times the current. Rerouting this line (and vcc) away from the microprocessor straight to the power regulator removed a lot of noise, further replacing the battery made it work great wireless.

However when using the 5V USB as source (connected to the microprocessor) there was still a lot of noise, this was solved by temporarily attaching a capacitor (~1000 uF) over the Bluetooth's vcc and gnd.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.