Timeline for 10 µF and 0.1 µF capacitors not making any difference in filtering out noise
Current License: CC BY-SA 4.0
30 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 20 at 11:26 | comment | added | wyc | @hcheung Thanks a lot for all the great advice! | |
| Jul 20 at 10:20 | comment | added | hcheung | 80MHz is the minimum required for WiFi to work, it more than sufficient to handle the HX711, and this will reduce the noise slightly as well as power consumption... This is what I would do... Just change the frequency from the IDE pulldown menu. | |
| Jul 20 at 10:04 | comment | added | wyc | @hcheung Sorry, you meant I should configure the ESP32-C3 so that it uses 80MHz instead of 160MHz? Or you're suggesting I get a MCU of 80MHz that also has Wi-Fi? | |
| Jul 20 at 9:17 | comment | added | hcheung | You still be able to use WiFi if you run 80MHz instead of 160MHz, also turn off WiFi when you are doing measurement and only turn on WiFi when you need to send data. It will be less noisy. | |
| Jul 20 at 4:17 | comment | added | wyc | @hcheung Thansk for the suggestion. I do need Wi-Fi (I want to send the readings to the cloud). And I think the ESP32-C3 is the cheapest alternative. | |
| Jul 20 at 1:37 | comment | added | hcheung | I run my HX711 with a MCU running at 5MHz, not only be able to run on a coin cell battery, and with very low noise generated by the MCU. You are using an ESP32-C3 which has clock of 160MHz, plus WiFI/Bluetooth that all generates a lot of noise, this is why it is important to use the right MCU for the right application, not to pick the fastest and newsiest available in the market... If you don't need wifi, turn it off, run at 40Mhz. You can also do some simply band-pass filtering with software, read 10 readings, sort them out from low to high, and sum-up the middle 4 readings and divide by 4. | |
| Jul 20 at 1:34 | comment | added | hcheung | If you look closely there are already decoupling capacitors on the HX711 board, adding to the protoboard probably doesn't help much if you don't know what kind of noise you are dealing with. BTW, the wires on EE, BB and AA better to be twrist-pair than some random wire dangling around that would pickup the noise on the input of HX711. | |
| Jul 18 at 9:37 | vote | accept | wyc | ||
| Jul 17 at 20:04 | history | became hot network question | |||
| Jul 17 at 16:40 | comment | added | Chintalagiri Shashank | Having recently tried out the HX711, I can say with some confidence these errors are well within the typical noise floor for load Cells designed for > 10 kg. If you're using a much smaller load cell, then I would have to go and check, but I still suspect your problem has nothing to do with decoupling whatsoever. | |
| S Jul 17 at 16:23 | history | edited | Greenonline | CC BY-SA 4.0 |
Added image descriptions and hover tags.
|
| S Jul 17 at 16:23 | history | suggested | MarianD | CC BY-SA 4.0 |
Some fixes.
|
| Jul 17 at 15:14 | comment | added | Attie | I didn't mention earlier, but are you confident that these variants are actually above the typical / intrinsic accuracy you'd expect from your sensors, wiring, and HX711? Perhaps power has nothing to do with this... | |
| Jul 17 at 15:13 | comment | added | Attie | "the capacitors are as close to the 3V3 and GND of the HX711 as they can be" - no, they're electrically quite far away, and on an entirely different PCB! | |
| Jul 17 at 14:38 | review | Suggested edits | |||
| S Jul 17 at 16:23 | |||||
| Jul 17 at 14:07 | comment | added | Gos | The HX711, that I suppose that could be more affected by power disturbances produced by the ESP32 when there is high WIFI activity or whatever. Could be that the onboard capacitors are enough, but to add another one just in case will not hurt. | |
| Jul 17 at 13:45 | comment | added | wyc | @Gos Sorry, by board, you mean the ESP32 or HX711? | |
| Jul 17 at 13:42 | comment | added | Gos | Both boards have already 10uF capacitors that probably are sufficient for the few tests that you did. Maybe an additional one is not needed 99% of the time, but could be that time to time the ESP32 needs more power than normal, the supply is not responding fast enough and the wind is blowing from the north. Then to have an additional cap close to the board could help. Or maybe is never needed. | |
| Jul 17 at 13:38 | comment | added | user1850479 | Doesn't the HX711 already include it's own very high PSRR regulator onboard? If so I'd expect additional filtering to make little difference. | |
| Jul 17 at 13:38 | history | edited | wyc | CC BY-SA 4.0 |
added 72 characters in body
|
| Jul 17 at 13:23 | history | edited | JRE | CC BY-SA 4.0 |
deleted 6 characters in body
|
| Jul 17 at 13:10 | comment | added | wyc | @Attie Sorry, but the capacitors are as close to the 3V3 and GND of the HX711 as they can be (the HX711 goes on top of those round female pin headers). Or you mean the electrolytic capacitor shouldn't be behind the ceramic one? Or you meant too far away from the ESP32-C3? | |
| Jul 17 at 13:09 | answer | added | kruemi | timeline score: 3 | |
| Jul 17 at 13:03 | comment | added | wyc | @Attie The oscilloscope readings look almost the same. 3V3 and GND. Circuit: i.sstatic.net/oRsxJjA4.png. Circuit with caps: i.sstatic.net/JpvTHec2.png. But since the readings are almost the same, the noise should also be pretty much the same? | |
| Jul 17 at 13:00 | comment | added | Attie | If the capacitors you're referring to are on the perfboard, then they're likely the wrong choice (high ESR for the electrolytic), and too far away (long legs, pin interconnects) to be useful. | |
| Jul 17 at 12:58 | comment | added | Attie | It's impossible to draw any useful conclusion from 10 samples with unknown period between them - think of it as "extreme aliasing", the signal could be doing anything between the samples. You'd be much better off trying to use (borrow?) an oscilloscope to actually inspect the signals. | |
| Jul 17 at 12:56 | history | edited | wyc | CC BY-SA 4.0 |
added 58 characters in body
|
| Jul 17 at 12:16 | history | edited | wyc | CC BY-SA 4.0 |
deleted 2 characters in body
|
| Jul 17 at 12:07 | history | edited | wyc | CC BY-SA 4.0 |
added 69 characters in body
|
| Jul 17 at 12:02 | history | asked | wyc | CC BY-SA 4.0 |
