Timeline for LM2596 buck converter overheats converting 36DC -> 5DC at 600mA
Current License: CC BY-SA 3.0
20 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 26, 2017 at 15:27 | answer | added | Spehro 'speff' Pefhany | timeline score: 3 | |
| Sep 22, 2016 at 18:02 | history | edited | George Y. | CC BY-SA 3.0 |
added 137 characters in body
|
| Sep 22, 2016 at 18:01 | vote | accept | George Y. | ||
| Aug 13, 2016 at 16:08 | comment | added | Andrew Morton | If it still becomes unacceptably hot, it might be cheaper (and possibly better for peace-of-mind) to get a better-quality convertor than stick a heatsink on the regulator with thermal adhesive. Those capacitors may not be of the highest quality, so they will have a more-than-expected reduced lifetime in the vicinity of hot components. Is there any chance of reducing the input voltage, e.g. from a tap on the transformer windings? | |
| Aug 13, 2016 at 3:18 | comment | added | George Y. | I powered it from a 36V lab power supply; it still overheats, but a bit less; most likely would survive it. Will do more tests overnight. | |
| Aug 13, 2016 at 1:26 | comment | added | George Y. | The rectifier is a full wave one, not one diode. Will add the capacitor and try. | |
| Aug 12, 2016 at 16:35 | history | tweeted | twitter.com/StackElectronix/status/764138282716463104 | ||
| Aug 12, 2016 at 16:10 | comment | added | jonk | And that's exactly what I did, in my case. I added a 2200uF capacitor BEFORE the circuit. I guess it never crossed my mind to use the tiny cap on the board as the filter. | |
| Aug 12, 2016 at 12:04 | comment | added | Agent_L | I second the large input cap suggestion. Those converters were never meant to work on pulsating input. The input cap is to merely protect the upstream from noise, not to act as sole source of power when sine hits zero. Also, is your rectifier full wave? Just one diode won't cut it. | |
| Aug 12, 2016 at 10:02 | comment | added | Andrew Morton | Can you add a capacitor in the range 680uF to 2200uF and give it a quick test? I suspect that 100uF capacitor on the board is a low-ESR type so that the capacitor you add does not have to have a low-ESR. | |
| Aug 12, 2016 at 8:38 | comment | added | George Y. | The converter has a 100mF filter capacitor for input voltage, so I do not use anything else. | |
| Aug 12, 2016 at 8:35 | comment | added | Andrew Morton | What size filter capacitor are you using after the bridge rectifier? | |
| Aug 12, 2016 at 8:27 | answer | added | Chupacabras | timeline score: 3 | |
| Aug 12, 2016 at 6:29 | comment | added | Flanker | There are a lot of gossip about cheap fake LM2596 or LM2576s relabeled as LM2596. Switching frequency of 50KHz kinda confirms it (LM2596 is 150KHz). If it is indeed LM2576, then inductor must be 100uH, not 33uH. Not sure if it could lead to overheating... | |
| Aug 12, 2016 at 6:12 | answer | added | Bruce Abbott | timeline score: 12 | |
| Aug 12, 2016 at 4:48 | comment | added | jonk | It doesn't seem normal to me. I've used those converters, as well. I think I still have a few laying around somewhere. I supplied them from a transformer, bridge, and capacitor (2200uF) filter and loaded them with about 1A @ various DC voltages between 5V and 12V output, as I recall. The input was from a transformer rated for 5A @ 24VAC and was a higher quality (low regulation %) transformer. No problems that I recall, other than one that was due to a mistake I made and figured out. | |
| Aug 12, 2016 at 4:18 | history | edited | George Y. | CC BY-SA 3.0 |
added 98 characters in body
|
| Aug 12, 2016 at 3:44 | answer | added | user76844 | timeline score: 1 | |
| Aug 12, 2016 at 3:43 | history | edited | George Y. | CC BY-SA 3.0 |
added 49 characters in body
|
| Aug 12, 2016 at 3:33 | history | asked | George Y. | CC BY-SA 3.0 |