0
\$\begingroup\$

I got a 10watt LED from Ebay but never thought that the driver and heat sinks are must-haves for this little master. When googled, I found several circuits of LED driver. But none of them are precisely a 'simple' circuit with takes 12v 0.5mA ~ 1.5A DC input and lit the LED to full brightness. So here are my questions folks:

  1. I got few 220V to 12v transformers (with 500mA, 1A, 2A current ratings). Can you please give me a simple circuit diagram (preferably with a linear regulator or NE555 or any other easy-to-find IC/Transistor) to lit the 10 watt LED?

  2. Just saw a video, where the maker claims, even though the forward voltage rating of the LED is 9-12v, it will burn itself if connected to 12v and furnished a meaningful explanation. Whereas, we has shown that the LED glows ok with a simple 9v NiMH battery. I knew these LEDs need 2-3AMP current at 12 volts. What's the real fact of voltage and current for these LEDs?

Many thanks in advance. Please help.

\$\endgroup\$
4
  • 5
    \$\begingroup\$ How saddly, this forum is becoming a place where 90% of people is asking about LED lamps over and over, mostly about the junk they bought from chinese sellers. \$\endgroup\$ Commented Feb 9, 2016 at 20:02
  • 7
    \$\begingroup\$ @MarkoBuršič Proposal: chineseparts.stackexchange.com To cover all the PCBs without schematics, parts without datasheets, broken, counterfeit, used, preprogrammed-OTP and other amusing subjects. \$\endgroup\$ Commented Feb 9, 2016 at 20:07
  • \$\begingroup\$ Sorry Sir, I am not a Pro. By profession I am a Java Architect working in one of the top IT firms in the world. Electronics was my hobby in school days. I know, I have miles to travel to line up with the latest trends of Electronics. I often buy these 'Chinese' things from Ebay, cause they are cheap and serves my hobby purpose. And I thought just like Stackoverflow, StackExchange is a good place for novices too. If any resolved threads are there on this forum, could you please share the link? Thanks again. \$\endgroup\$
    – sribasu
    Commented Feb 9, 2016 at 21:23
  • \$\begingroup\$ 2-3 amp at 12V is 24-36 watts, not 10. JFYI. Thus my speculation that 900 mA might be a reasonable value if the thing even is a "10W" part (eBay unspeced parts - who knows?) \$\endgroup\$
    – Ecnerwal
    Commented Feb 9, 2016 at 22:18

4 Answers 4

0
\$\begingroup\$

Yet another simple LED driver request and yet another application for the LM317 as constant current source to be a simple (if somewhat inefficient) LED driver, but, being in current source mode, at least a good one for LEDs.

schematic

simulate this circuit – Schematic created using CircuitLab

You can get brave and figure a value for the resistor, or you can sneak up on it with adding resistors in parallel as shown at an early stage, given you have sketchy parts.

1.25 volts is maintained between Adj and Out (the principle of the LM317's operation) - so each 10 ohm in parallel will add 125mA to the output in this configuration; or if you think the thing should take 900 ma, you could put together some combination that would get you close to 1.38 Ohms. Note that you'll burn over a watt in that resistor as well as the dissipation in the LM317 itself (which is best regarded as a "smart resistor" at least for power use.) Other driver toplogies can be more efficient - this one is mostly simple, and involves just a few components.

As commented above "2-3A at 12V" would be 24-36 watts. 10 watts at 12V would be 833 mA, 10 watts at 9V would be 1111 mA. It is important to note that LEDs are primarily current driven devices - that is, the current, and especially the maximum current, is what you need to control.

\$\endgroup\$
2
\$\begingroup\$

I actually built myself a small lamp from one of these 10W "12V" datasheetless LEDs like this:

  • buy half a dozen 1W 0.5ohm resistors
  • bolt LED to aluminium heatsink (this is essential)
  • put LED in series with resistors and power supply
  • remove resistors until measured power in LED is about 9W (not 10W: derating it improves life and I don't trust the manufacturer)

(If the LED is the type built on an aluminium PCB, it will be a tremendous hassle to solder to it...)

\$\endgroup\$
2
  • \$\begingroup\$ Aw! That dirty trick is great! Thanks @pjc50 \$\endgroup\$
    – sribasu
    Commented Feb 9, 2016 at 21:25
  • \$\begingroup\$ Just make sure that everything stays cool (it'll get warm, but should still be touchable) and electrically insulated and you avoid the use of flammable materials in construction, and it'll be OK. \$\endgroup\$
    – pjc50
    Commented Feb 9, 2016 at 21:57
1
\$\begingroup\$

You are going to have to measure, at the very least, the forward voltage of the LED when driven with perhaps 100mA.

When you know the approximate forward voltage, you can then calculate the required current. From those characteristics, you can either build or purchase a suitable power supply / driver circuit.

You know that the rated power is 10 Watts. The rest is easy, once you know the forward voltage of the LED.

\$\endgroup\$
0
\$\begingroup\$

I got few 220V to 12v transformers (with 500mA, 1A, 2A current ratings). Can you please give me a simple circuit diagram (preferably with a linear regulator or NE555 or any other easy-to-find IC/Transistor) to lit the 10 watt LED?

If you want to utilize the POWER LED to full potential then you should measure the thermal response of the LED in Current-Time domain. And it is related to your second question:

just saw a video, where the maker claims, even though the forward voltage rating of the LED is 9-12v, it will burn itself if connected to 12v and furnished a meaningful explanation. Whereas, we has shown that the LED glows ok with a simple 9v NiMH battery. I knew these LEDs need 2-3AMP current at 12 volts. What's the real fact of voltage and current for these LEDs?

Production of LEDs(particularly POWER LEDs) has some scattering within the process so the their Typical forward voltages are spreading between a voltage level(9 - 12 V in your case). When dealing with 10W power LED, you should aim a test method based on constant current rather than constant voltage. You can have the LED tested with this simple configuration:

schematic

simulate this circuit – Schematic created using CircuitLab

Start From drawing 100 mA and go up to 2-3 A. See the forward voltage for a wide current sink. Then draw 0.5 A and check the temperature of the LED after a few hours. And then, you will have a countable data about your LED. If for 0.5A your LED does not produce Heat much, draw more current and repeat the test. Then you have a reliable data on your hand.

\$\endgroup\$
3
  • \$\begingroup\$ How are you going to draw 2-3 amps through a 50 ohm resistor from a 12.5 volt supply? 12.5 volts divided by 50 ohms is 0.25 amps. \$\endgroup\$ Commented Feb 9, 2016 at 21:14
  • \$\begingroup\$ I added initial configurations in the simulation. R1 starts from 5 Ohm to 50 Ohm and V3 is 5V square PWM. So from 100 mA to 1 A LED currents generated. So yes should have mentioned that the user should change the R1 value to get different LED currents. \$\endgroup\$
    – Alper91
    Commented Feb 9, 2016 at 21:17
  • \$\begingroup\$ ...my 2N3904 datasheet lists 200mA as absolute maximum collector current, which might cause problems with real parts at 0.5A rather than poorly simulated ones that don't blow up when they should. That really SHOULD be the one thing simulated parts get right. \$\endgroup\$
    – Ecnerwal
    Commented Feb 9, 2016 at 22:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.