Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: Paul R.
high temp supply components don’t help if the LED itself gets too hot internally

The base of the bulb was hot enough you did not want to hold it. The LED's were still operating when I removed it and found it very hot. Heat means power consumption. LED's themselves don't get that warm. I even cut the plastic bulb housing and examined a failed unit. I am sure the LED's are still operational but the power supply failed. I've built many types of power supplies. I have held an Extra Amateur Radio license for many years and hold a GROL commercial License. I spent many years building scratch electronic devices. Have etched and build many PC boards from scratch.

172 posted on 08/21/2019 11:30:45 AM PDT by Texas Fossil ((Texas is not where you were born, but a Free State of Heart, Mind & Attitude!))
[ Post Reply | Private Reply | To 170 | View Replies ]


To: Texas Fossil

My point was, the power supply DESIGN might or might not be ok. Unless you know the requirements of the LED used, know whether or not inadequate parts were substituted, maybe ran a “P-SPICE” analysis or other simulation, know whether or not the heat sinking was adequate, and ruled out all those other factors I described, the power supply design might have been fine, but subverted by (for example) parts substitution. There’s no way to know though, on our end, what the design actually specified. Considerable (industrial) experience tells me that too often with Chinese products, the design is ok, but on, oh, say the 3rd run, somebody decided to substitute cheaper parts. Or the parts vendor themselves screwed up. Or... well, the list for even something as simple as a LED bulb can go on and on. Once the process is “past” the engineer (who often is not even Chinese), the Chinese are really bad about cutting corners or not observing details.

Also, uh, hate to disagree slightly, but high output LED’s most certainly can get quite hot (and fail because of that) if run hard and not properly heat sunk. It’s current times voltage drop of the diode = power to dissipate, similar to any other diode, really. Also, as the LED chip itself heats up, it’s light output drops. You can actually run one hard off a rock solid power supply and watch the output dim as the LED heats up. I’ve also seen this happen in LED head worn lamps where the LED runs hot, and once the housing heats up, the LED has nowhere to send its heat, gets even hotter, and dims. (The spec sheets from reputable LED mfgr’s don’t lie, and show this as well.)

That said, based on your further investigation details, one of those power supply problems is most likely the issue. The power supply provides more opportunities for something to go or be wrong, than the LED. A near-parallel is that over the years I’ve had quite a few electronic ballasts in 4’ fluorescent fixtures fail. There, one expects the bulbs to last a limited time, but to have the ballasts fail even 1/2 as often is really aggravating. Especially when the fixture is in a “difficult” spot to uninstall... But, fail often those ballasts do.

Back to the LED’s, the fact yours have a plastic housing is a clue that the heat sinking may have been inadequate. Glass / metal housings will typically do better.

Last, a good / easy test for anyone might be to purchase a few different brands of LED bulbs in the lumen output rating you need. Try them out in an easily accessible fixture like a table lamp, and see which runs coolest. Take the others back to the store(s).


174 posted on 08/22/2019 12:53:38 AM PDT by Paul R. (The Lib / Socialist goal: Total control of nothing left worth controlling.)
[ Post Reply | Private Reply | To 172 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson