OSC Arduino and questions on the LM3914 as "VU" Faders

Hey guys. Thinking of scrapping together a 48 “channel” control surface using some arduino megas and OSC. (No motorized Faders, all rotary encode. those mfs expensive lol, Sorry guys).

Aside from the goal, I’m using OSC to gather strip dB messages with an Arduino mega and turning them into PWM duty cycle value that is then low passed (47k ohm, 0.8uF) and converted into a DC voltage to drive the LM3914 on pin 5.

How sensitive are the LM3914s to ripple noise? Would ripple cause inaccurately displayed values?

I would not want any unnecessary flickering between LEDs if said issue would arise, but too high of a capacitance(I think) on the lowpass would obscure and not display the dB in a quick enough fashion as it reacts to the OSC messages being received.

Order of operations as follows:

OSC(strip channel dB) - > Arduino - > PWM signal - > Low pass filter(Conv. duty cycle to voltage) - > LM3914

Thanks guys. Been trying to plan this project for over a year but life’s been in the way.

LM3914 Datasheet

How sensitive are the LM3914s to ripple noise? Would ripple cause inaccurately displayed values?


Just a thought - is there a reason to convert to analogue and then back to digital (e.g. PWM to analogue and then back to digital using the window comparator chain in the LM3914)? If the idea is just to display the (channel) dB values received by the Arduino in visual form, you might consider chaining a set of shift registers together, use the parallel outputs to drive the LEDs, and then just clock a ‘control surface wide’ word out of the Arduino through the chain to update the LEDs directly every frame. This might put more processing load on the Arduino but it would likely be cheaper and more reliable - less possibility of noise, calibration drift etc.

Why not keep i digtal, OSC --> Arduini–>I2C --> LED river

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.