Forums » General Electronics Discussion

Hello,

Thanks for the great help I already received on this forum. I think I'm reaching the final stage of my simulations.

Setting: I have a low cost RGB LED dimmer, PWM controlled at 100Hz, which has a lot of functionalities such as remote control, color control, setting of six custom colors, blinking, fading... I woud like to keep these functionalities as I would like to apply them in my application.

However, because the LED's are rotating at a speed of 100 rpm, the low PWM frequency make the different components (red, gree, blue) clearly visible during (night) photography.

Target: I don't want to touch the existing circuitry, but I'm willing to add an aditional circuit. I'm convinced that PWM is a very good solution for LED control, and after reading I'm convinced I would best "upscale" the frequency with respect of the duty cycle. Upsclaing would be from 100Hz to 1kHz or 10kHz (I really have to experiment this in real life).

My idea: I thought of adding two low pass (RC) filters to get a nice dc signal and then feeding this DC signal to a 555 to copy the original duty cycle. The frequency of the 555 would be set by the right settings of the concerned R's and C's. I've found a lot of examples in which the duty cycle is regulated by a potmeter, but I seem to miss a configuration where the dutycycle be regulated by a dc current.

Has anyone a link or example of such an implementation of the 555?

Kind regards,

Luc

Hello,

As an addtition to my request, I did find this circuit, but I don't know how to implement the mathematical trigger with real life components.

https://www.circuitlab.com/circuit/z242cn/555-timer-as-pulse-width-modulation-pwm-generator/

Kind regards, Luc

You could realise the glitch generating trigger using a quad EXOR 74HC86 or CD14070B device:

A trick you could try borrowed from the cinema and TV world is simply to gate the PWM signal on and off using a higher frequency clock.

That fools the eye by masking the LF PWM with an HF chopping signal.

Trouble is then, you can't get higher than 50% duty cycle ....

... unless you can replicate the original PWM circuit function but with a higher clock frequency and then chop the original PWM o/p with that. Same effect but you then get the full range of duty cycles.

A lot of effort that might be better put into building the PWM controller you actually need in the first place.

:)

Thank's alot, Signality,

You are absolutely right, I would love to build my own remote controlled PWM controller. Espcially, because I want to use add at least a second LED (yellow) to the RGB LED. As you might have noticed, those low cost LED strips and dito dimmers produce a terrible yellow (much to greenish). Designing a 4 or 5 or even 7 LED PWM dimmer would be my dream. I've read that some professional LED color makers use up to seven LED's (Selador white paper by ETC).

Unfortunately, I'm only an over-enthusiastic hobbyist. I last touched electronic components 30 years ago, and so I have to live with my limited abilities and knowledge in this field. Although I'm learning every day! I have no expertise to build a pwm with all the functionalities that I would like to keep (remote control; on/of; color control; custom colors; flikker; fade) and that I would like to add (more LEDs to control).

I'm convinced that programming a microcontroller cannot be avoided and, once again, it is out of reach because lack of expertise and material. Moreover, the remote control thing for controlling seven LED's seems extremely complicated to me.

So, I will have a deeper look into your proposed solution.

But as to restate my basic need (leaving the deaming aside), I would like to speed up my pulse train from 100Hz to 10kHz (e.g.) while preserving the duty cycle).

Once more, thank you for responding to my request.

Kind regards,

Luc