Timeline for ESP8266/ESP-12F gets damaged when powered by LDO (LD1117v33) every time
Current License: CC BY-SA 4.0
28 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 24, 2021 at 0:45 | comment | added | Russell McMahon♦ | ... These may include touching with a finger, connection of test equipment, adding input serial resistance and more. Starting off doing what the manufacturer says is almost alwaays a frirst requirement. In some situations then changing to other arranegements and seeing what happens MAY be valid. | |
| Aug 24, 2021 at 0:45 | comment | added | Russell McMahon♦ | @moefear I spent significant time watching this question and trying to ensure that it got fair treatment. You dismissive comment abpout reading before I post is not appreciated :-) . The data sheetv that you cite also says "Only a very common 10 µF minimum capacitor is needed for stability." Your attempts to justify not doing what the manufacturers say is necessary, or what various experienced people here have been telling you make it very hard for people to help you. || A marginally stable device may "come right" when various informal changes are made. ... --> | |
| Aug 23, 2021 at 23:40 | answer | added | gmlux | timeline score: 3 | |
| Aug 23, 2021 at 17:34 | answer | added | Spehro 'speff' Pefhany | timeline score: 7 | |
| Aug 23, 2021 at 17:17 | comment | added | winny | Schematic or it didn’t happen. | |
| Aug 23, 2021 at 16:12 | history | edited | JRE | CC BY-SA 4.0 |
deleted 31 characters in body
|
| Aug 23, 2021 at 13:51 | comment | added | StarCat | @moe fear, no comment answers your question because we can't really determine the reason for your problem without a schematic, except for the obvious fact that you say you have no capacitors in your power supply. The discussion in the comments is about having a need for capacitors. You do need capacitors for a stable power supply, even if it seems you can solve your problem with a simple resistor. Please post a schematic, even if it's for just enabling others, who read the answer you posted, to determine if their situation is the same as yours. | |
| Aug 23, 2021 at 13:41 | comment | added | Kartman | Once you have a feedback loop, you need to ensure loop stability otherwise it oscillates. Adding/subtracting resistance changes the loop characteristics. Have a read of this: ti.com/lit/an/slyt194/… | |
| Aug 23, 2021 at 13:34 | answer | added | mo FEAR | timeline score: -7 | |
| Aug 23, 2021 at 13:33 | comment | added | mo FEAR | LDO output to GND (to which the LDO is connected to), is defacto a feedback loop. | |
| Aug 23, 2021 at 13:33 | comment | added | mo FEAR | A zener plus resistor would also require capacitors for stability, or how an LDO differs from a zener plus resistor, such that an LDO would need caps while a zener plus resistor would not? "Feedback loops" is a broad term that can be seen present in any circuit. Its more specific meaning usually relates to OP-Amps, it's not clear how that would carry over to a zener plus diode. Better said, that feedback loop exists anyways with or without a capcitor, it doesn't instantly come into existance once a capacitor is added. The fact that there is a closed path from | |
| Aug 23, 2021 at 13:28 | comment | added | mo FEAR | Still, no comment answered the question. The voltage output on a DMM & Oscilloscope appears stable enough, and the solution avoided the need/use of capacitors. The question is what exact process is going on to damage the esp, where is the damage happening, and why does a simple AMMETER or Resistor resolve the issue? Can anyone point me to a single reference that claims | |
| Aug 23, 2021 at 13:26 | comment | added | mo FEAR | @McMahon: Please read before you post. I'm not rejecting, I'm asking about the why. | |
| Aug 23, 2021 at 13:25 | comment | added | mo FEAR | I'm using ST st.com/resource/en/datasheet/ld1117.pdf | |
| Aug 23, 2021 at 13:04 | comment | added | Justme | @moefear Which exact LD1117 variant you have, manufacturer name and part number please? They might not be all alike. Preferably add a link to datasheet. | |
| Aug 23, 2021 at 12:50 | comment | added | Blup1980 | @moefear I don't get why you push for answers other than the one asking you to add a proper output caps. As we are mostly experienced EE people here, I really suggest that you follow our advices. If it doesn't works, then OK, we can try to think about other causes. But before that, we should make sure that the obvious capacitor issue is not the root cause of your problem, | |
| Aug 23, 2021 at 12:08 | comment | added | Russell McMahon♦ | @moefear The comments about capacitors, whoch you are rejecting, are made not inly by "most people" but in the manufacturers datasheet here - Page 1: " ... Only a very common 10 µF minimum capacitor is needed for stability ..." | |
| Aug 23, 2021 at 11:44 | comment | added | Peter Smith | Every variant of the 1117 I have ever seen requires output capacitors (the suggested value in this datasheet is 10uF) but they also need a minimum ESR for loop stability. The typical requirement is around 0.5 ohm. | |
| Aug 23, 2021 at 11:26 | comment | added | StarCat | Not using capacitors creates exactly the kind of problems you are having (instability). The resistor that you've used as a "solution" to your problem, is not really a solution but a stopgap current limiter. The datasheet very clearly states that capacitors are needed for stability (which means they are necessary). Please look at common power supply designs that use linear regulators and especially the role of capacitors in such designs. The ESP8266 is a complex device that can generate quick changes in power supply load. Capacitors help keep the supply voltage constant. | |
| Aug 23, 2021 at 11:14 | comment | added | mo FEAR | BTW... the Datasheet specifies the capacitors are recommended, but not necessary for (to theory of) operation. I assume they're not as simple as a zener and a R, but that itself works without any caps, and is the whole point of a zener. That's why I assume it is OK to expect such a simple setup to work, even if not ideally under all circumstances. What is shocking is how it's presence depends on, and only on, the presence of the Ammeter, even though I don't think it purposefully internally has any "stabilizing" capacitors attached anywhere (atleast not any relevant to the path being measured). | |
| Aug 23, 2021 at 11:09 | comment | added | mo FEAR | @Mat: I connect the Ammeter directly between the LDO and the VDD pin of the esp-12F module. It's not very good quality, it has an internal resistance of around 1 Ohm, which gave me the idea of trying a separate Resistor. The ground of the esp goes back to the battery pack, after joining with the ground of the LDO. there is nothing else. Even with an oscilloscope, I can't discern anything unusual, even if I set it to trigger at any voltage over 5V. | |
| Aug 23, 2021 at 11:07 | comment | added | mo FEAR | @PeterSmith, there are no capacitors. But I just found out that a simple R works. It solves the problem for me. So anyone else with this issue can try that. It might not work in all situations, but is simpler in my opinion that several caps. | |
| Aug 23, 2021 at 11:02 | comment | added | Kartman | The use of capacitors is very relevant. Ignoring them suggests you need to learn a bit about how regulators work. I’ve observed regulators oscillating and they do it well and the outcome is not good for the load. That was with a 7805. LDOs are even touchier. Even the wrong sort of capacitor can induce oscillation. They’re a linear device and they operate on feedback. As for your tiny down converter, without knowing the specs of said supply we’d only be guessing. All I can suggest is that you read the manufacturer’s datasheet and comply. Or keep blowing things up. | |
| Aug 23, 2021 at 10:34 | comment | added | Peter Smith | What output capacitor(s) [type and value] are attached to the LDO? | |
| Aug 23, 2021 at 10:25 | comment | added | Mat | Please add a schematic of your setup (including where/how you connect your ammeter), the details matter. | |
| Aug 23, 2021 at 10:21 | history | edited | mo FEAR | CC BY-SA 4.0 |
added 705 characters in body
|
| Aug 23, 2021 at 10:20 | review | First posts | |||
| Aug 23, 2021 at 10:58 | |||||
| Aug 23, 2021 at 10:16 | history | asked | mo FEAR | CC BY-SA 4.0 |