strange LED?

tim_ber

Well-Known Member
Joined
12 May 2009
Messages
1,039
Visit site
I bought this 12volt LED to investigate.

Powered from 0V upwards, the top two leds light firstlly and then the side panels light up as 12V is reached.

I took the resistor out so I could run it from a lower voltage source (or so I thought).

The resistor was on the negative leg (that surprised me - should it?)

BUT the biggest surprise was that once I managed to solder it all back together, the side panels would not light at all.

No circuit board can be seen.

Those little 301 things, are they inductors or something? Is that why this will only work with a 12V supply?

Hope photos look ok.
Thanks
 
What voltage do you want to run it off?

I was just investigating.

With the Chinese (yes note there is no diode in there) Led I was seeing what ticks.

Without the resistor, I did run it up to (by memory) about 6V before I chickened out and stopped.

The top LED was obviously buring away furiously, but the side ones - not a jot.

Beats me.
 
V=IR
12=0.04A X 300 ohms.

So I guess the resistor I removed was just protecting the top LEDs, and the side ones are still protected for a 12V supply.

Interesting arrangement.

I think I've got that sorted now haven't I?

Cheers
 
I was just investigating.

With the Chinese (yes note there is no diode in there) Led I was seeing what ticks.

Without the resistor, I did run it up to (by memory) about 6V before I chickened out and stopped.

The top LED was obviously buring away furiously, but the side ones - not a jot.

Beats me.
You need to know several things. Here are just a few of them:

  • An LED is a diode.
  • A semiconductor diode doesn't obey Ohm Law.
  • A diode is not a resistor.
  • A resistor obeys Ohms Law.
  • You can tell very little about a semiconductor by looking at it.
  • Electronics is a science, unlike tree felling.
  • Three year BSc (hons) undergraduate courses in electronics are available, in which you can learn the basics.
I hope this helps.
 
You need to know several things. Here are just a few of them:

  • An LED is a diode.
  • A semiconductor diode doesn't obey Ohm Law.
  • A diode is not a resistor.
  • A resistor obeys Ohms Law.
  • You can tell very little about a semiconductor by looking at it.
  • Electronics is a science, unlike tree felling.
  • Three year BSc (hons) undergraduate courses in electronics are available, in which you can learn the basics.
I hope this helps.

Thank you.
My 4 years with a double degree in chemistry and physics didn't touch much on electronics alas.

I would have thought an ordinary diode would have stopped people blowing the Light emitting diode if they reversed the polarity accidentally was all.
 
I'd have thought the physics degree would have taught you enough about semiconductors to be getting along with LEDs.

Was a long time ago and there was little electronics involved. This was when the BBC acorn computers were in the labs.

Now if you want to know about quarks...

I've probably forgotten that too !

I do remember PNP or NPN stuff regarding semiconductors (I mean I remember studying it and how the charges moved) but I have obviously forgotten anything of use !

HOWEVER, the nice King Fisher avatar chap (I can't see it was Halcyon or something wasn't it) has now given me the knowledge that has just had me bypassing the 301's to see if I can get these LEDs going at a lower Voltage - and all it took was his answer - not another year in uni req'd . Yippee.
 
What you thought was a resistor could have been a current source IC.
Or it could just be a bigger resistor than 300ohms.
I would guess that if the top two LEDS light at about 6V, they are just two in series, while all the others are 3 in series.

Is there a better book than Horowitz and Hill?
 
What you thought was a resistor could have been a current source IC.
Or it could just be a bigger resistor than 300ohms.
I would guess that if the top two LEDS light at about 6V, they are just two in series, while all the others are 3 in series.

Is there a better book than Horowitz and Hill?

Thanks for that.

'The best self-teaching book and reference book in electronics … The beauty and fun of electronics shows through.' Radio Communication

I'm going to have a word with Santa.

And also see if I can find what I did with the bit I chopped out and double check it was a resistor.
Cheers
 
A couple of points:

In order to find out what's going on, your first step should be to trace the circuit, not randomly fiddle with components.

As has been pointed out, an LED doesn't behave like a resistor, it doesn't obey Ohms Law - there is not a linear relationship between the current drawn and the voltage across it. As an aproximation, white LEDs behave a bit like a constant voltage of about 3.6V with a very low value resistor in series. Below abut 3.6V they won't pass much current and won't produce much light, above 3.6V the current increases rapidly until destructon probably arounf 4V.

I imagine your device probably has the LEDs in pairs, in series, with the 300R resistor to set the current at about 20mA. The resistor value doesn't make much sense in any other configuration.

You will notice the pins on the assembly are labelled "+" and "-", so the complete light is polarised and won't work if wired backwards. Most LEDs will only tolerate a relatively low reverse voltage, but I guess the series connection means they should survive reversal at 12V, so ne need for any diodes.

Hope this helps.
 
Looks to me like there are parallel groups of 3 LEDs with a 300R in series with each group, and a missing diode/resistor/fuse in series with the whole lot. Taking it out either blew up the most sensitive LEDs, or there is a bad connection.

No matter what you do, you can't run this on less than about 12V, as the 3 LEDs in series drop almost 11V (assuming they are white).
 
Last edited:
Here's a very simplistic tutorial on LEDs.

First something on diodes. Simply, they're like a one-way valve. If you apply a voltage in one direction, they conduct current. Apply a voltage the other way around they don't. But, to get them to conduct, you need to apply enough volts to switch them on. Rather like having enough water pressure in a pipe to open a one way valve against a spring. That's called the forward voltage. Once a diode switches on, it's resistance is very close to zero. So, apply volts without limiting the current & you'll blow it up. If you apply volts in the reverse direction, then you can get it to conduct if you apply enough. That's called the reverse voltage & in fact some diodes are supposed to work in the reverse direction. But, generally the reverse voltage is quite a bit higher than the reverse voltage. An LED is designed to emit light when conducting in the forward direction.

The forward voltage for LEDs varies quite a lot from type to type. Some are around 2.5V & are designed to conduct around 0.02A of current. Put too much current through them & they get too hot a fail.

OK - so what's ohm's law? Any conductor has a resistance. So put a voltage across it & the current that flows is the voltage divided by the resistance - simples. If we want to work out what resistor to use to pass a current from a voltage source then just flip the equation so R = V / I.

So the sum for LEDs is quite simple using ohm's law. If we want to put 0.02A from a 12V battery down an LED that has a forward voltage of 2.5V, then we have 9.5V to drop across a resistor. 9.5 / 0.02 = 475ohms. It's silly to do this since most of the power is dissipated in the resistor. So, we can connect two diodes in series & get more light from the same power. Now the diode drop is 5V leaving 7V left for the resistor which would now be 350ohm. With 4 LEDs, we only have 2V left & a 100ohm resistor etc.

I'm not sure how your light is actually connected but since you removed a resistor, you would have isolated some or all of the LEDs. If you want to reduce the operating voltage, you need to reduce the resistance. But you will need enough volts to switch on the LEDs in the first place.
 
Last edited:
LEDs

From the photos I am pretty sure the LEDs are known as 5050 LEds. These are square surface mounted LEDs with 3 LED in each package. If you reduce the voltage on a variable supply you can confirm when you see 3 tiny sources of light. They are actually a blue or green LED which excites the phosphor to produce white light. There are 6 connections on the back of each LED and they are then wired 3 in series. So the lowest voltage the whole thing will give any light on is something above 9v.
I have actually removed the 5050 and wired all 3 diodes in parallel so that it will run on about 4 volts. It still needs a series resistor though. As said each series package runs at about 20 ma full brightness.
I think you will find the 310 ohm resistors feed a portion of the lamp. Not as you might imagine all LEDs with one resistor simply to keep the power dissipation in range of the tiny SMD resistors.
Of course my guess may be wrong and they may be SMD LED one diode per pack.
You may have cooked some of the LEDs by running with no current limiting resistor although somehow I doubt it and think you have probably broken a connection. good luck olewill
 
Is there a better book than Horowitz and Hill?

Rather a broad question...

Certainly sets the standard for undergrad electronics courses, but the current, 2nd edition, dates from 1989, so it's a bit outdated, to the extent that some of the devices it talks about are obsolete and unobtainable - a 3rd edition has been rumoured for the last couple of years, may surface some time in 2013/2014.

Rather overkill for understanding LEDs, though - Ohms law, plus knowing the current draw and forward voltage of the LED in question covers 99% of what you need.

If you're starting electronics from scratch, perhaps this http://www.amazon.co.uk/Electronics-Owen-Bishop-Bristol-Oxon/dp/0750655453 or this http://www.amazon.co.uk/Getting-Started-Electronics-Forrest-Mims/dp/0945053282 might be more suitable choices.
 
Last edited:
Used similar to these to light shop fittings, they come in long lengths, up to three metres and can be cut down every 30mm, if they are reversed, they survive quite well, but twisting them is an easy way to break the printed tracks.
 
Top