0
\$\begingroup\$

I am trying to figure out an appropriate resistor value for the following circuit.

schematic

simulate this circuit – Schematic created using CircuitLab

Input voltage - 3.3v Blue LED, forward voltage 3.3v LED : APT3216VBC My maths tells me that driving that LED at say 12ma, means that the resistor should be:

3.3-3.3/0.012 == 0

Checking with some LED resistor calculators, they suggest a 1 ohm resistor : http://led.linear1.org/1led.wiz

Given my voltage drop, and input voltage, what does the addition of this 1 ohm resistor actually make to the circuit?

Thanks

\$\endgroup\$
  • 6
    \$\begingroup\$ If you want to predict the current through an LED with that little headroom, you need a better model of the LED than "forward voltage is 3.3 V". \$\endgroup\$ – The Photon Jun 4 '18 at 15:57
  • 5
    \$\begingroup\$ I would say the LED/power supply combination was poorly designed. \$\endgroup\$ – Eugene Sh. Jun 4 '18 at 15:57
  • \$\begingroup\$ Is this for a blazing flashlight or apartment lighting application, or as some state LED on an electronic device? \$\endgroup\$ – Ale..chenski Jun 4 '18 at 16:19
  • \$\begingroup\$ Well I have the information available in the datasheet which is linked to in the question...so what is a reasonable resistor value and how did you reach that result? \$\endgroup\$ – RenegadeAndy Jun 4 '18 at 16:49
  • 1
    \$\begingroup\$ As state indicators, I re-formulated your question here electronics.stackexchange.com/q/378129/117785, and gave an answer. \$\endgroup\$ – Ale..chenski Jun 4 '18 at 18:52
5
\$\begingroup\$

If you want to predict the current through an LED with that little headroom, you need a better model of the LED than "forward voltage is 3.3 V".

You can see that this device has a typical differential resistance (at the operating point) on the order of 30 ohms, so the 1 ohm series resistor is probably not doing much:

enter image description here

On the other hand, what the graph doesn't show is that this curve likely moves around quite a bit if the device temperature changes.

I'd recommend either using a higher supply voltage so that a reasonable series resistor can be chosen, or treat the 1-ohm resistor as a placeholder to be adjusted once you've built the circuit and figured out what resistance is needed to avoid over-stressing the device at the worst-case temperature (and be prepared for pretty big changes in brightness with temperature).

(Also notice that the specified maximum forward voltage at 20 mA is 4 V, so you might have some trouble with device-to-device variation)

\$\endgroup\$
  • \$\begingroup\$ To elaborate for better teen-level understanding, if OP wants 12 mA and his supply is stabilized at 3.3V, then the R1 must be about 25 Ohms, according to the I-V chart. The R1=25 will provide about 300 mV voltage drop, moving the operating point to ~12 mA area. This should produce plenty of light, given the ~150 mcd LED rating. If this LED is used as some functional indicator for a gadget, OP might want to consider something like 1k resistor instead. \$\endgroup\$ – Ale..chenski Jun 4 '18 at 16:16
  • \$\begingroup\$ This would be an even better answer if tolerance of the 3.3V DC supply were included in the "need for better models". \$\endgroup\$ – glen_geek Jun 4 '18 at 16:25
  • \$\begingroup\$ ali -- how did you reach the 25 ohm value, or even 1k.....it's not clear how that is calculated. \$\endgroup\$ – RenegadeAndy Jun 4 '18 at 16:48
  • \$\begingroup\$ @RenegadeAndy, Old traditional LEDs for board indicators used to emit about 2-3 mcd of light. This is sufficient. This LED emits 150 mcd at 20 mA. The dependence of light versus current is about linear, so to get 1.5 mcd of light you can decrease the current 100-fold, to 0.2 mA . The forward threshold is 2.4 V, so you need a drop of about 0.9V. If R1 is 1k, it will feed about 1 mA into the LED, making 150/20 = 7.5 mcd of light. This is plenty, so you might consider even something like 3k ohms. \$\endgroup\$ – Ale..chenski Jun 4 '18 at 17:08
0
\$\begingroup\$

LEDs are not Ohmic elements. In other words the voltage across the LED is not a linear relationship. Nor are all LEDs created the same.

Instead once the forward voltage drop is exceeded the current through the LED will rise rapidly, a change of a few tenths of volts will create a large difference in current. The resistor forces the current to reduce by taking up more of the voltage drop across the battery compared to the LED until stable state.

\$\endgroup\$
  • \$\begingroup\$ so is 1 ohm here a safe choice? \$\endgroup\$ – RenegadeAndy Jun 4 '18 at 16:47
0
\$\begingroup\$

Is there a point in adding a 1 Ohm resistor to this LED circuit?

Not really, A 1Ω resistor will work as well/poorly as no resistor at all.

IMHO I think no resistor would work best for this LED. But I would not use this LED.

No resistor may be better than 1Ω. It is not likely that Vf will not exceed 3.3v somewhere in the visible range. When Vf reaches 3.3v current will stop increasing. This is very likely to happen before the 30 ma max.

According to the I-V curve typically the current will be 20mA @ 3.3v with no resistor. So adding a resistor would only make the LED dimmer.

The LED you specified has a max 4v Vf. While it can be run with a 3.3v supply when Vf when the LED work in the typical range. The 4v Vf max falls outside operating parameters.

so is 1 ohm here a safe choice?

Maybe, maybe not, very likely it would be safe. But not really any better or worse than no resistor.

The problem there is no minimum Vv specified.

The risk would be if the Vv were lower than typical, the current could increase to more than the max 30 mA rating of the LED. Not too likely but possible.

It would be reasonable to guesstimate Vv min could be 3v (10% below typical).
At 3.0v a 1Ω resistor would allow 300 mA. Not that it would be reasonable that this would happen, Vf will always increase as current increases. The question is: How reasonable is it that Vf will be under 3.3v @ 30 mA? Assuming the supply voltage will never rise above 3.3v.
As long as Vf is at or below 3.3V at 30 mA, no resistor is fine. As the current increases the forward voltage rises above 3.3 and it is at that point the current stops increasing. It is very likely with this particular LED Vf will always be greater than 3.3v at 30 mA.

The allowable LED current is 300 mA with a 1Ω where Vf = 3.0v
enter image description here

It would be very safe to use a 10Ω to limit the resistance to 30 mA at 3.0v.

enter image description here

But there you have the flip side problem. If the Vf is higher then the LED may not be bright enough. With a max Vf of 4.0v this is well within the realm of possibilities.

If this circuit were in a product to be sold to others, I would not use this LED.
If it to be used for myself, I would be okay with no resistor. If there is any possibility that a component will generate a field service call, I am not going to use that component no matter how remote the possibility. Even with no resistor.

I would use a higher power LED that has a Vf below 3.3v. A mid power LED will typically have a Vf of 2.8V at 100 mA.

I would want to see an I-V curve closer to this:

enter image description here

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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