Jump to content

Lamp issue


Recommended Posts

Hi all.i got a set of 100 led lamps that run on 3 x aa battery so 4.5 volt.as they are on when juke is on I've run them on a regulated supply at 3.80 volt.they worked fine for 2 days but now gone out,supply is still correct.no sign of damage.so I dont understand it.

Link to post
Share on other sites
30 minutes ago, warlord said:

Hi all.i got a set of 100 led lamps that run on 3 x aa battery so 4.5 volt.as they are on when juke is on I've run them on a regulated supply at 3.80 volt.they worked fine for 2 days but now gone out,supply is still correct.no sign of damage.so I dont understand it.

Have you got 3.8 volts DC?

Link to post
Share on other sites
2 hours ago, warlord said:

Hi all.i got a set of 100 led lamps that run on 3 x aa battery so 4.5 volt.as they are on when juke is on I've run them on a regulated supply at 3.80 volt.they worked fine for 2 days but now gone out,supply is still correct.no sign of damage.so I dont understand it.

your just cursed when it comes to any form of lighting, the gods of illumination have it in for you...

Edited by jocky581
spelling
Link to post
Share on other sites
2 hours ago, Brigham said:

Have you got 3.8 volts DC?

Yep,3.8 volt.under what it should be.unless the supply is not good enough.it was a kit from maplin,1.5 amp out,0 to 35 ac in.and I had 28 ac in.3.8 out.

Link to post
Share on other sites
22 minutes ago, warlord said:

Yep,3.8 volt.under what it should be.unless the supply is not good enough.it was a kit from maplin,1.5 amp out,0 to 35 ac in.and I had 28 ac in.3.8 out.

I hope the power supply kit had a massive heatsink? 

If it's the one I think it is, the voltage regulator would have been dropping nearly 36 volts DC which will cook it in no time, even with a huge heatsink. 

A transformer with a 6 volt output would have been plenty.

Edited by CanonMan
Link to post
Share on other sites
19 minutes ago, warlord said:

Think you are right,fuck it,I give up.

you may just have got a duff set, Did you use the battery box to hook up to, or cut it off, remember that LED's are actually current devices, 0.6v DC with 10mA average, current will increase with brightness, and this might show as an increased supply voltage, are you sure the batteries were series connected, to give 4.5 volts, or were they parallel to give more current at 1.5 volts, it possible the battery box has a simple LED driver circuit. but at least this time you got a couple of day light.

Link to post
Share on other sites
2 minutes ago, jocky581 said:

you may just have got a duff set, Did you use the battery box to hook up to, or cut it off, remember that LED's are actually current devices, 0.6v DC with 10mA average, current will increase with brightness, and this might show as an increased supply voltage, are you sure the batteries were series connected, to give 4.5 volts, or were they parallel to give more current at 1.5 volts, it possible the battery box has a simple LED driver circuit. but at least this time you got a couple of day light.

Sounds like a power supply issue, see above.

Link to post
Share on other sites
28 minutes ago, warlord said:

Yep,3.8 volt.under what it should be.unless the supply is not good enough.it was a kit from maplin,1.5 amp out,0 to 35 ac in.and I had 28 ac in.3.8 out.

 

3 minutes ago, CanonMan said:

Sounds like a power supply issue, see above.

I can only go by what is said, yes it might be a PSU issue, hence why I asked about the battery box, over volting an LED will drive the current so high and blow it, I would get a set of fairy/xmas lights, fancy flashing sequences, and check the input voltage, switched mode might be 100 to 240 volts, and pick up a mains take off from the PSU, nice if you can find a set that will take 48 volts,   but best to keep what ever was designed to light the things up by the manufacturer, and control the supply to that, simple relay from the lighting circuit, ect, but thats just my opinion.

 

  • Like 1
Link to post
Share on other sites
2 hours ago, CanonMan said:

I hope the power supply kit had a massive heatsink? 

If it's the one I think it is, the voltage regulator would have been dropping nearly 36 volts DC which will cook it in no time, even with a huge heatsink. 

A transformer with a 6 volt output would have been plenty.

Its this one mate 

20201229_204449.jpg

Link to post
Share on other sites

Ordered 3 more sets.so I'm going to test one set with batts,till they go flat,and one set wired to a proper psu set at 3 volt and soak test for 4 days non stop.ive checked existing set and it's the controller in batt box that's failed again,not the LEDs.

Link to post
Share on other sites

I've attached the datasheet for that particular power supply. If you read it, you'll see that it recommends a 9V transformer for 3V to 5V output, so if you've been running it on 28V AC you'll have cooked the LM317 voltage regulator chip:

For a 28V AC supply, the rectified and smoothed DC input to the voltage regulator is 39.6V (28 * 1.414) and the output is set to 3.8V so the voltage regulator has to dissipate the difference, which is 34.6 volts. It'll struggle to do this with a heatsink, let alone without one.

If you use a 9V transformer and set the output to 4.5V DC, that should be enough when wired to the terminals of the battery box to get it working. You might have to replace the voltage regulator though, it's probably fubar because of the 28V transformer.

k1823.pdf

Edited by CanonMan
Link to post
Share on other sites
15 minutes ago, warlord said:

Ordered 3 more sets.so I'm going to test one set with batts,till they go flat,and one set wired to a proper psu set at 3 volt and soak test for 4 days non stop.ive checked existing set and it's the controller in batt box that's failed again,not the LEDs.

Does the controller work with a 4.5V supply? If you're only giving it 3.8V that might not be enough to get it going.

Link to post
Share on other sites

They are the regulators which have no protection when they fail, they DON'T fail SAFE, the regulators go short and the input voltage passes straight through to the output, especially those bought of the ebay!

you need a switch mode type which WILL fail SAFE

lots of fake ones about that don't meet the specification they should due to fake Chinese regulators, buyer beware

Link to post
Share on other sites
2 minutes ago, Andrew96_ said:

They are the regulators which have no protection when they fail, they DON'T fail SAFE, the regulators go short and the input voltage passes straight through to the output, especially those bought of the ebay!

you need a switch mode type which WILL fail SAFE

lots of fake ones about that don't meet the specification they should due to fake Chinese regulators, buyer beware

Unlikely to be a fake LM317 in a Velleman kit though, in my experience.

Link to post
Share on other sites

When the lights went out the reading was 3.8 volt DC out.they worked fine at this for 2 days.the supply unit can take ac or DC in upto 35 volt.and I've tried the psu and wether you put 20 volt ac or 30 volt ac the output remains the same.anyway,many thanks for all replies,I do appreciate it,my tests will prove what's happening when my new sets arrive.i think that cheap psu board is maybe not good enough.

Link to post
Share on other sites
1 minute ago, warlord said:

the supply unit can take ac or DC in upto 35 volt.and I've tried the psu and wether you put 20 volt ac or 30 volt ac the output remains the same.

Yes, but the larger the difference between the input voltage and the output voltage, the more heat the voltage regulator is going to chuck out and the greater the chance of it failing altogether. That's the point I was making.

Stick with a 9 volt transformer and a decent heatsink on the voltage regulator and it should be fine.

 

 

  • Like 1
Link to post
Share on other sites
45 minutes ago, CanonMan said:

Unlikely to be a fake LM317 in a Velleman kit though, in my experience.

true, but thermal runaway exists! temp goes up, output voltage goes up, lights blow, then no current being drawn so temp returns back to normal, so does the voltage, nothing to show what happened as everything looks innocent! Those regulators do not fail SAFE! they fail UNSAFE and blow up what is connected to it

that's my 2p worth

  • Like 2
Link to post
Share on other sites
2 minutes ago, Andrew96_ said:

true, but thermal runaway exists! temp goes up, output voltage goes up, lights blow, then no current being drawn so temp returns back to normal, so does the voltage, nothing to show what happened as everything looks innocent! Those regulators do not fail SAFE! they fail UNSAFE and blow up what is connected to it

that's my 2p worth

I know. I've never tried to drop 30+ volts with one though.... 😁

 

Edited by CanonMan
Link to post
Share on other sites
2 minutes ago, Andrew96_ said:

also I can see no 0.1uf decoupling capacitors on the input and output of the regulator, this wont help the stability either in my opinion, Velleman kit or not!!!

I hadn't spotted that. Oops.

Link to post
Share on other sites

Yes,you guys have verified my suspicions,the psu board maybe ok for motors but not good enough for LEDs.it maybe a week or 2 but I will let you know what happens with my tests.thanks all again.love your input.

  • Like 1
Link to post
Share on other sites
47 minutes ago, warlord said:

Yes,you guys have verified my suspicions,the psu board maybe ok for motors but not good enough for LEDs.it maybe a week or 2 but I will let you know what happens with my tests.thanks all again.love your input.

get a good quality switched mode psu, check the specs for input voltage range, output voltage and minimum out put load, expect you will be fine, especially if you can find one that will meet your 28 volt input.  have fun, should last years and bring a sparkle back to that jukebox

 

  • Like 1
Link to post
Share on other sites
×
×
  • Create New...