Timeline for Are LEDs in modern streetlights usually pulsed? If so, roughly what frequency?
Current License: CC BY-SA 3.0
17 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 15, 2017 at 13:14 | comment | added | uhoh | There's multiple helpful answers here, but this answer zeroed in on what it was that was making me think they might be pulsed and then zeroed it out. Thanks for your help! | |
| Jul 15, 2017 at 13:13 | vote | accept | uhoh | ||
| Jun 14, 2017 at 17:54 | comment | added | uhoh | @Makyen sometimes working towards an answer takes a little time. In this case my question hypothesiszed that they would be pulsed based on some thoughts of optimum current. This answer is examining that premise and finding some issues with it. When an easy/immediate answer is not forthcoming, sometimes one has to work a little harder. If you know the answer, please post it! | |
| Jun 14, 2017 at 17:30 | comment | added | Makyen | While this is an interesting discussion about driving LEDs, and does address some of the assumptions stated in the Question, it does not actually address the primary questions stated in the Question: "Are LEDs in modern streetlights usually pulsed? If so, roughly what frequency?" | |
| Jun 14, 2017 at 16:01 | comment | added | uhoh | @Harper I used to work with people who were making QW LEDs, but that was a while ago. I do look at data sheets, but not for high power LEDs, at least not yet. Once I click your link, that will all change. Thanks! | |
| Jun 14, 2017 at 15:42 | comment | added | Harper - Reinstate Monica | @uhoh that efficiency graph he presented is in the data sheet of every LED emitter I've ever seen. Have you read any LED data sheets? They are "enlightening" (ha ha) to say the least. Here, pick any random one (you can filter on specifics) and click "Datasheet". | |
| Jun 14, 2017 at 11:45 | comment | added | Barleyman | @Andrés PWM on low duty cycles is a pretty nasty load for PWM thought. You would like to have a large capacitor to supply majority of the transient load. Not that 29mA is a great load if your max output would be 2.9A. | |
| Jun 14, 2017 at 11:17 | comment | added | uhoh | OK I'll hold off on accepting anything until I get something definitive about street lighting, but your answer and discussion are really helpful. Thanks! | |
| Jun 14, 2017 at 11:10 | comment | added | Andrés | But the efficiency that matters is the efficiency of the power supply + LED and power supplies ussually have increased efficiency with higher power. So is the "cascade" of both efficiencies that matters in the end, and that could be a design decision as well. | |
| Jun 14, 2017 at 11:09 | comment | added | Barleyman | @uhoh The only reasons to use PWM is that if you need very high instaneous light output (optical sensors for example) or you want to control brightness. For lighting there's no reason for PWM if you don't want a dimmer. | |
| Jun 14, 2017 at 11:06 | history | edited | Barleyman | CC BY-SA 3.0 |
(Junction) temperature diagram
|
| Jun 14, 2017 at 10:46 | comment | added | uhoh | That's pretty amazing, it seems GaN devices have improved a bit over the years, it's almost Nobel Prize-worthy! OK would you go one step farther and suggest that this might fairly well rules out any intentional and/or substantial pulsing of the LED current; that there would be no reason from the LED side to pulse it? Further, as long as the frequency from the power system is filtered somehow, that the LED current would really most likely be DC in a typical outdoor/streetlight application? I'd like to accept the answer as long as it addresses the question sufficiently. | |
| Jun 14, 2017 at 10:44 | comment | added | Barleyman | There's also a second order effect, light output goes down with junction temperature. So PWM with higher current will run with worse efficacy to start with. It will also run hotter for the same light output that makes things worse. So it's definitely better to use constant current from system efficiency point of view. | |
| Jun 14, 2017 at 10:35 | comment | added | Barleyman | @uhoh Yes, that is DC. With PWM the active pulse would behave much the same way. | |
| Jun 14, 2017 at 10:15 | comment | added | uhoh | Wow! To double check, is the plot really showing DC current, not PWM average current? And an LED that can handle 400 mA DC continuously is still at least locally linear all the way down to around 5 mA DC (give or take)? i.sstatic.net/dSQbw.png This is all really measured, and not just a plot/extrapolation of some parameterization? | |
| Jun 14, 2017 at 9:51 | history | edited | Barleyman | CC BY-SA 3.0 |
added 159 characters in body
|
| Jun 14, 2017 at 9:44 | history | answered | Barleyman | CC BY-SA 3.0 |
