[SOLVED] Pulsed LED brightness

[d123]

Hu,

I'm posting separate threads about one circuit to prevent too many questions making it complicated and messy to understand and answer. I'm trying to break a problem into manageable parts.

If I want three paralled LEDs to carry e.g. maximum 25mA each with a stable DC voltage and current, then make the DC signal/source 100Hz with a 10% ON duty cycle, the LEDs are very dim compared to the constant current.

The maximum LED current is controlled by an op amp constant current and NMOS block, the op amp voltage reference is variable but clamped at 1V to prevent blowing the LEDs with too much current.

Do I need to pass/pulse e.g. 10 times more current through the LEDs for the equivalent brightness achieved with constant DC as the duty cycle is only 10%? That makes no sense to me... Is the duty cycle too low: 1ms on, 9 ms off?

Thanks.

 

[FvM]

There is a small effect of duty cycle on brightness, but it's mostly determined by average current, at least within the specified peak current range.

If you seem to see a different behaviour, I'd bet that your LED driver does not achieve the intended average current.

100 Hz is fine, although you'll probably want a higher frequency to avoid stroboscopic effects wenn sweeping your eyes fast over the display.

 

[barry]

I have to disagree. Duty cycle most certainly DOES have an effect on brightness. Duty cycle is directly related to average current. If you've got a 10% duty cycle, your average current is 10% of maximum current.

 

[d123]

Hi,

Thanks.

If I'm understanding your answers and a couple of web pages I read about pulsing LEDs:

10 mA at steady DC has to be 100 mA at a 10% duty cycle to obtain equivalent brightness.

If the LED pulsing objective is to lengthen battery life, it's a pointless circuit add-on because: 10 mA at steady DC draws the same power and will drain the battery at the same rate as the 10% ON duty cycle at 100 mA. Is that correct?

 

[c_mitra]

10 mA at steady DC has to be 100 mA at a 10% duty cycle to obtain equivalent brightness.

When you say 10% duty cycle, you should specify that the duty cycle refers to the voltage applied. everything else remaining same, 10% duty cycle shall result in 10% average brightness. The average should be calculated over a complete cycle under steady state.

effectively, LED output is proportional to instantaneous power input (approx). This linear relation is often used in opto couplers.

I do not see how LED pulsing can lengthen battery life. It can, if (say) you turn on the LED 10% of the time (but not at high frequency). In that case, the average brightness is low (because it is off most of the time) but the status is visible (because it turns on often). In such cases, the LED is being used as a status indicator and not a source of illumination.

 

[d123]

Hi,

When you say 10% duty cycle, you should specify that the duty cycle refers to the voltage applied. everything else remaining same, 10% duty cycle shall result in 10% average brightness. The average should be calculated over a complete cycle under steady state.

effectively, LED output is proportional to instantaneous power input (approx). This linear relation is often used in opto couplers.

I do not see how LED pulsing can lengthen battery life. It can, if (say) you turn on the LED 10% of the time (but not at high frequency). In that case, the average brightness is low (because it is off most of the time) but the status is visible (because it turns on often). In such cases, the LED is being used as a status indicator and not a source of illumination.

Thanks for the clarification.

I can be very thick sometimes in my circuit design. To make a favourite joke of it: ... So all those 'free energy' circuit videos aren't truthful, after all.

Everyone's replies have solved my misunderstanding of LED average pulsed vs constant current.

Many thanks.