[sdiy] Improving LED response when driven from half-wave rectifier

[Tim Ressel]

Exponential?  Sorry, but that didn't sound right. After all, power increases linearly with current, and light output is power, yes?

I checked datasheets from Kingbright and Avago and they both show luminous intensity increasing linearly with current.

As for Neil's complexity comment, it's five resistors instead of 2, and they cost $0.0018 each. Admittedly I like the current pump a lot since it saved my bacon on a test fixture design. Maybe I'm biased.

I do like the circuit with the LED in the feedback path of the opamp, and indeed that is how one drives the LED in a Vactrol. But the LEDs I'm driving are on the front panel on their own control board. Its easier to have them ground referenced.


--Tim Ressel




----- Original Message -----
From: KA4HJH <ka4hjh at gmail.com>
To: sdiy <synth-diy at dropmix.xs4all.nl>
Cc: 
Sent: Wednesday, September 5, 2012 11:40 AM
Subject: Re: [sdiy] Improving LED response when driven from half-wave rectifier


On Sep 5, 2012, at 2:36 PM, Neil Johnson wrote:

> Hi,
> 
> Tim Ressel wrote:
>> I use a modified Howland Current Pump to drive LEDs, especially bi-color.
>> 
>> Here is the schematic:
>> https://dl.dropbox.com/u/5536133/led_driver1.pdf
> 
> Well, I guess if you like complexity...
> 
> You can do it with an op-amp and single resistor if you're ok with the
> LED fully floating.

The brightness of LEDs varies exponentially with the current so if you really want to get fancy you need an expo convertor. This is easily done with an MCU and a look-up table. It looks quite nice. 8D


Terry Bowman, KA4HJH
"The Mac Doctor"

_______________________________________________
Synth-diy mailing list
Synth-diy at dropmix.xs4all.nl
http://dropmix.xs4all.nl/mailman/listinfo/synth-diy