We found a Philips Hue "Lightstrip" that was not working, and it turned out to be a duff 240 V AC to 24 V DC PSU. When I measured the resistance across the mains input pins, it gave "open circuit" even at the 200 M ohm range of my multimeter (trying both polarities). So maybe duff transformer or other mains-voltage circuitry, I deduced.
But when I measured the same pins on a *working* PSU which drives the lightstrip OK, it also gave open circuit.
Has anyone come across a SMPSU which has "infinite resistance" (ie open circuit even on multi megohm range) and yet which works fine? Do they use capacitors as "dropping resistors" (dropping impedances) on the mains pins?
?Seeing? a very high resistance at the input of a SMPSU isn?t unusual.
From past experience, a common fault is ?dry joints? - often you can see then, especially on ones which have mosfets / transistors which are through hole - those are the pads which tend to fail. Clean away the old solder, resolver and it may work. I usually do all the switching devices ( the ones on the heat sink)
I'd do that, but this PSU seems to have no screws (not even concealed under labels) so the two parts of the case are evidently glued together or else held by springy tabs that need ten hands to release them all simultaneously.
I also discovered that Philips made two different versions: some have a barrel connector between the PSU and the control logic (and the LED strip hard-wired to the controller) whereas others with a subtly different part number have the PSU and controller hard-wired and a multi-pin connector between controller and LED strip. By chance we have some of each type ;-)
If its a modern "light weight" SMPSU, then the transformer is not connected across the input pins like it was on old linear regulated supplies. Quite often the first thing connected (excluding some EMI filtering) is the rectifier.
Some meters may show a very high resistance in one direction at least (>
7Mohms on the one I just tried) - but that "resistance" is equally likely to be the meter just charging the reservoir cap in the PSU.
(You might also see a different behaviour on a diode test mode if the meter has one)
Yup - especially if you are not applying enough voltage to forward bias the rectifier diodes.
Not usually unless they are supplies designed for very low loads. Not typical for your typical external wall wart.
Ah! I bet that's it. I wonder what voltage a multimeter typically applies to a load when it is measuring resistance.
I remember in Elec Eng at university we were given a project in teams to design a SMPSU with certain criteria as regards permitted variation of load voltage as more current was drawn, and permitted range of input voltages (provided with a Variac for the test).
For some reason that escapes me, a dropping resistor was needed on the input to the transformer - or maybe it was between bridge rectifier and switching transistor. (This was an "old fashioned SMPSU" where the transformer was fed at mains frequency rather than switching frequency as happens nowadays to keep transformer as small as possible).
One of my team-mates (and I wish it had been me!) had the brilliant idea of using a series capacitor instead of a series resistor to do this, with a size based on impedance at the relevant frequency. This was a lossless solution unlike a resistor, and meant we achieved significantly better efficiency than any other teams. We won the prize!
(I remember that the switching transistor that we were required to use did
*not* like to be turned on permanently when full load was being drawn. The feedback circuit limited the M:S ratio to a little less than 100% even with heavy load. But it only took a loose wire during development, before everything had been permanently soldered, and the transistor lost its switching signal and was on permanently. We all did it at one time or another. There was a noise like rustling autumn leaves for a couple of seconds, and then the transistor blew up. Ours left a dent in the ceiling! After that, we were all made to wear safety goggles just in case...)
Not on a LED one, certainly but of course you didn't mean SMPSU, because most LED power supplies are not that smart. see yer Big CXlive videos. a series capacitor with a surge resistor into a bridge rectifier and a smoothing cap - or maybe not - and then into a chain of series LEDS is just cheap and works fine.
OOI I just connected my Fluke 117 back to back with my old Maplin Precision Gold M125. If you put the fluke in auto range and measure resistance, then the other meter measures around 0.5V and the Fluke sees
10 MOhms. If you put the fluke in manual range, then you can get a maximum of about 2.5 volt out of it on some ranges (with the resistance being out of range on all but the highest). Diode test mode uses 2.5V all the time it seems.
The other way around gets similar results although it sees the input impedance of the fluke at around 11 MOhms, and outputs about 0.4V
This dry joint problem used to afflict old fashioned dimmers too. It must have something to do with the two materials and the solder. You could at least when I could see, see the kind of well around the pins after a few years. Must be chemical or something I don't know, but some high power audio amps had similar issues. Pioneer used to use wire wrap on the pins. Brian
It also leaves the load at mains voltage, as there is no transformer to isolate it. I used to have a handy Pifco torch which directly plugged in to charge it up. It was old school Nicads and traditional bulb. One day I came home to find the living room full of smoke of the magic kind that smelt of melted plastic and the like. The capacitor dropper had self destructed. It took days to rid the place of the pong. I don't know what they make them of, but back in those days it was something less than pleasant!
It's because there was too much vibration on the conveyor belt out of the solder bath before the solder had set. This was a problem with large tags that retain heat for two long. A crack forms all round the spill, and goes open circuit some time in the future.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.