# A bit of a silly question, LED current limiting resistor

Discussion in 'General Electronics Chat' started by scubasteve_911, Jan 3, 2008.

1. ### scubasteve_911 Thread Starter Senior Member

Dec 27, 2007
1,202
1
Hello all,

I feel silly asking this, but I really want to make sure before I do it and realize I was wrong.

If I have a 3.3V supply and it's within 2% capable of 6A, can I directly connect a 3.3V LED?

More specifically, this LED
http://www.stanley.co.jp/device/pdf/led_chip02/s1113f.pdf

I know how to calculate for a simple current limiting resistor, but there is no difference between my supply and LED drop, so it doesn't seem needed.

Steve

2. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
You're correct.

If (Vcc) - (Vled) = 0, then the resistance required to limit current is also 0.
If the result (Eresult) were greater than 0, you'd have to calculate for R = Eresult / Iled.

Funny, I was guessing blue or blue/green before I looked at the sheet

3. ### Ron H AAC Fanatic!

Apr 14, 2005
7,050
657
You may get more or less than 10mA. See the highlights in the chart below.

File size:
38 KB
Views:
34
4. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Ron H brings up a good point; they may vary a bit, aka "your mileage may vary"

Blue LED's tend to require higher voltage to obtain their rated current.

As you get towards green, the voltage tends to be lower for a given current.

It sure wouldn't hurt to verify at what voltage the particular LED you're planning on using reaches it's rated current.

It's OK to run it at less than it's rated current. Exceed the current rating, and life is bright but short.

5. ### scubasteve_911 Thread Starter Senior Member

Dec 27, 2007
1,202
1
Thank you both very much! I really wanted to make sure, nothing more embarassing than a 500\$ prototype PCB with a burned out LED because you didn't do it right

I understand that I will get some variance in current due to temperature, supply tolerance, but as long as it gives me about 2000 hours of service, I'll be more than happy

You guys are awesome, thanks again!

Steve

6. ### Ron H AAC Fanatic!

Apr 14, 2005
7,050
657
Prototype? Is this for a commercial project? I know that you can buy products that have diodes without current limiting resistors (e.g., laser pointers, I think), but I would never do that. If this is a one-off, it might be OK, but who pays \$500 for a one-off PC board?

7. ### Audioguru New Member

Dec 20, 2007
9,411
896
The LED might work at its typical voltage of 3.3V. Or it might burn out at 3.3V and need only 3.1V. Or it might not light at 3.3V and need 3.8V.

It needs a voltage higher than its max voltage so that a current-limiting resistor can average-out the differences in its voltage.

I have had many customers pay a fortune for my one-off circuits since nobody else did what I do:
1) The Canadian government was in charge of all airports. In their wisdom (?) they spec'd an FM radio station broadcast quality PA system.
I doo'd it. It sounded perfect. Then the government sold all the airports to a company that didn't care about how the PA sounds and the new ones sound as bad as an old telephone.

2) Telephone boardroom conferences sounded very muffled until I made and sold many transmit equalizers that made the sound crisp and clear. I sold every one I demo'd.

3) The government wanted a huge intercom system with many custom features for the health plan department. I doo'd it. It did exactly what they wanted and no other company could. Then after I installed it they decided to use the telephone system instead.

4) A bank head office wanted their wireless boardroom tele-conference system scrambled. I doo'd it with suppressed-carrier single sideband. It descrambled perfectly but was completely unintelligeble when scrambled. Then they went bankrupt for cheating their customers.

Custom stuff is expensive. I made lots of custom circuits. One at a time.

8. ### scubasteve_911 Thread Starter Senior Member

Dec 27, 2007
1,202
1
Yes, this is a prototype for my degree project. I require two 676-pin Spartan 3 FPGAs to implement my design, not to mention the dozens of ICs to support my application. If you are aware of PCB prices for prototypes, then you won't be shocked by a 500\$ price tag for a 6-layer PCB.

I'm not even sure what you are suggesting, except that you would never buy something without a current limiting resistor or pay 500\$ for a custom PCB.

Steve

9. ### hgmjr Moderator

Jan 28, 2005
9,030
214
Is this LED only serving as a power on/off indicator or is its purpose more involved? Explain please?

hgmjr

10. ### scubasteve_911 Thread Starter Senior Member

Dec 27, 2007
1,202
1
Agreed, it might be nice to be on the safe side and choose a lower voltage LED, but something attracted me to the elegance of a matched LED. My supply voltage will be within 2% of 3.3V, so we're talking about +/-66mV. The LED forward voltage characteristic is actually quite repeatable and I doubt it very highly that it will be far off of what they specified. If my supply deviates from about 2.5V to 3.5V, I will still be in the SOA for the device.

I would have to second your statement about custom things, they just cost you money. A lot of standard products will not suffice. In my particular instance, I probably could purchase evaluation boards, but this is for a competition.

Steve

11. ### scubasteve_911 Thread Starter Senior Member

Dec 27, 2007
1,202
1
It's to indicate which FPGA I am programming via JTAG, so it isn't very critical at all. Maybe I am just really into that wonderful blue-green LED to let me know what FPGA I have selected?

Steve

12. ### hgmjr Moderator

Jan 28, 2005
9,030
214
If it is not too late, you may want to select an LED that has a lower forward voltage rating than the one you have chosen. I share audioguru's concern that whatever LED you use should require a current limiting resistor and still provide a clear indication even in the brightest of ambient lighting conditions it is likely to encounter.

hgmjr