I'm using a breadboard power supply to power up two STM32F103, two Hall flow sensors, one IC555, and one 16x2 LCD with backlight. While everything is connected I have alot of noise in the system which is interfering with my measurements. Input to the breadboard module is 12V DC /2A. Here are the osciloscope images of 5v output and 3.3 V output. enter image description here enter image description here

I was thinking of soldering a 1000 microFarad capacitor before the voltage regulator , and adding in parallel to each output a ceramic 100pF capacitor. Will it work?, And how to actually calculate the exact capacitance required to filter this kind of noise? (Sorry for the badly photographed images)

Schematic of the whole system is located below. enter image description here

5V and 3.3 V is connected to the breadboard power supply mentioned above, while 12V is a seperate module. Hall sensors.

Water tanks PCB prototype Front panel

  • 1
    \$\begingroup\$ What's your schematic? Have you placed suitable decoupling capacitors for all the devices? Have you placed them in the right place? Have you put in bulk decoupling? If you don't know what decoupling capacitors are, do an internet search "decoupling capacitors". There are many many books written on the subject, so it would be a challenge to put all you need to know in an answer here. if you show your schematic and layout, we can probably advise. \$\endgroup\$ – Puffafish Jul 10 at 10:21
  • 1
    \$\begingroup\$ I agree that we need a schematic. While everything is connected I have alot of noise in the system which is interfering with my measurements If I said: "The noise / ripple on the supply is not an issue." How would you prove me wrong? What hard evidence do you have. You would not be the first to blame supply ripple, conclude that it is "too much", fix that ripple but still not get a better measurement result! Since you're using a breadboard module, you might have long wires and many contacts involved, that's a no-no for low noise measurements! \$\endgroup\$ – Bimpelrekkie Jul 10 at 10:26
  • 1
    \$\begingroup\$ @Bimpelrekkie The issue is i'm using Input capture to detect the length of the PWM signal (time spent from one rising edge to the next falling edge) to determine the water level in a tank via capacity. Also i'm using IC to count Hall sensor pulses in one second to determine flow. What i've noticed was that when i connect the IC555 timer to a stable laboratory power supply it generates nice clean signal, while using the module i get a noisy distorted waveform, especialy when the water level changes. So since the only thing i've changed was the power supply it led me to that conclusion. \$\endgroup\$ – Robert Sabljo Jul 10 at 10:36
  • 1
    \$\begingroup\$ OK, I see, that is good evidence. You're using the 555 timer for capacitance measurement. Like that it is likely to operate on a high frequency (for a NE555, I would consider to use the TLC555 instead) and can be quite sensitive to supply ripple. You can always just try extra supply decoupling and see what you get. There's not much to go wrong unless you connect polarized caps with the wrong polarity. If adding decoupling doesn't work, consider giving the 555 timer its own 5 V regulator. \$\endgroup\$ – Bimpelrekkie Jul 10 at 10:43
  • 1
    \$\begingroup\$ 1khz to 10khz OK, then the "test capacitor" has more capacitance then I expected. Then indeed no reason to use TLC555. Yes, the 555 on it's own voltage regulator form the shared 12 V. That way the two separate 5 V supplies (a clean one for the 555, dirty one for the rest) will start up simultaneously and there should be no issues. \$\endgroup\$ – Bimpelrekkie Jul 10 at 10:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.