LED failures?

[jol50]

I have a couple MTX amps, newish Asian ones, and the power LED have blown among other issues. I put a new one in and get 2.74v, on a typical blue 3mm LED. I used this digikey number 365-1173-ND for the new LED.

That voltage is ok right; with a Vf of 3.4v and 4v max? Should I assume the factory LED were poor quality? I cycled it a few times and did not see a spike on the meter, though I suppose it might not pick up I'd have to try the scope. Brightness stays the same so I didn't bother, and in limited testing the new LED seems fine. These are jackhammer sub amps, is there some abuse the owners might have applied? I just don't want them to do it again later and aside from overvolting or significant age I don't know what else makes them puke. The particular amp I tested worked fine other than the bad LED.

[dangus]

Current is more meaningful than voltage... the LED should be fed through a resistor that limits current to some reasonable amount (like 10 to 30 mA). LEDs behave a lot like zener diodes: they drop a nearly constant voltage over a wide range of current.

Unless someone screwed up and used the wrong value series resistor, they could just be bad LEDs.

[TO-3]

Well, I would not say that they act like a zener, as the voltage does NOT stay fixed as current is increased, they do act like a diode, as the output intensity is proportional to the current that is applied to it. If you can find the series resistors, for the LEDs, see what the values are, and you will want to replace them to provide the correct current, as stated before, between 10-30mA. Most likely, they are coming off off a the 12V bus, so: (12V-Vf)/20mA=R. So, (12V-3.4)/20mA = 430 Ohms.
Possibly the resistors were stuffed with too low of a value.

[pheonix358]

Hi, if you measure while the car is off you get 2.74V. What about when the car is running? Your feed voltage can vary between 12.6 and 14.4. What is it on your car? Measure the volyage across the led while the car is running and put a small say 20Ohm resistor in series with the led to limit current.

Terry

[Kabigon]

Originally Posted by TO-3

Well, I would not say that they act like a zener, as the voltage does NOT stay fixed as current is increased...

Zener voltage increases with zener current.
This is especially true with lower voltage zener diodes.

[djQUAN]

cheap LED's fail for no reason even when operated well within their ratings. they fail as a short circuit. that's my experience with them. :P

I usually get around 3V across blue LED's. 2.7V means they are running at low current or the LED has low Vf. 5mA is already too bright as a power indicator. 10-30mA would be very bright. IMO

[jol50]

It is pretty bright head on but not much different than other amps with bare LED showing (and they are clear LEDs), not much different than than one original I had that worked but was a while ago not exactly sure how bright it was. I had 4.x volts and 1.x volts to it from 12v ground, it is in a protection circuit type of thing though odd all three are to same 'ground' that showed 1+ volts. Not sure where 'ground' side went, will have to setup to run it out of case upside down.

Should I try to measure MA on it?

I found people asking about LED problems with this same model line of amp on the net, so it must be common to a point.

Also found this amp is not fixed, now it works fine but when I turned it off the overload LED stayed lit dimly. Then today I tried it again and it did not, but looking really close I could see the overload LED was still lit just a little. It appears to go to a SMD transistor. But...I fixed another identical amp that works fine, so I'm going to check the voltage on those LED and find a difference. Not very familiar with this issue so not sure if a transistor might be leaking, out of spec resistor, guess I will look for those where the current is different.

This was all done on a 13.8v power supply. All the lights light up for a moment when remote is applied, power LED stays while powered. Thanks for the help I'll post what I find. This amp is a JH300, it actually has no sink but a PC cpu type cooler for the transistors with a variable fan on them. China one side board, but some cool ideas you can stack them three high as well.

[toobnoob]

you have a schematic?

[I Am An Idiot]

Find one of the resistors feeding the LED and double it's value. I doubt that you will see any dimming of the LED and it will be safe from overcurrent, if that is what is taking them out.

[jol50]

I don't have schematic.

Nice... I tested the one with another LED strip I put new LED in. So I take the bad case that belongs to it and put one in that and test it, no more LED staying on. I put it all together and hook up speaker and input and test, it works great. Turn it off and again it stays lit. Take the input and speaker off and try again, it stays lit (only when amp off). I let it sit and the other protection light comes on too but not the power LED. Soon as I turn it on just the power runs like it should. I wonder why it is OK one time and not the next. So yeah, have to take it back apart and see where power is on it from, if it decides to do it again. Maybe the cold spray can help me.