Skip to main content

Timeline for Analogue input spikes

Current License: CC BY-SA 4.0

8 events
when toggle format what by license comment
Jun 7, 2023 at 16:23 comment added Nick S. @stelPet also, commercial dataloggers often use digital filters for filtering out 50/60 Hz and stuff like that. What is the application for your datalogger? Unless it's a high-speed acquisition FPGA-based datalogger, I doubt you'd have too much impact on your processing if you do minor digital filtering - it might be just easier to use Edgar's advice rather than fiddle around with soldering stuff to the module (at that point, you should be making your own module, I'm sure there are good charge regulator ICs out there).
Jun 7, 2023 at 16:21 comment added Nick S. @stelPet to me, my initial thought process points to an inductive spike. I'm wondering if putting a snubber diode across the inductor will help with them; if not, you would have to resort to implementing a hardware filter as described below. Two more things to consider - how are you checking the voltage? If you are using a scope with an isolated ground, then you'd have to make sure that the ground are bonded together with the module and the scope - there was a thread on AllAboutCircuits - when I'll find it I'll link it. You could ask in EE SE.
Jun 7, 2023 at 0:47 answer added Vile timeline score: 3
Jun 6, 2023 at 21:42 comment added stelPet No, i'm using a fully charged 18650 battery, not plugged in into an external charger.
Jun 6, 2023 at 20:59 comment added Nick S. Are you charging a battery while you are running the program?
Jun 6, 2023 at 20:37 comment added Edgar Bonet Re “i don't really want the code to do the filtering, as it's gonna cost me some speed”: A 3-point median is a very cheap filter speed-wise, and should be quite effective against this kind of spikes.
S Jun 6, 2023 at 20:00 review First questions
Jun 6, 2023 at 20:28
S Jun 6, 2023 at 20:00 history asked stelPet CC BY-SA 4.0