DC110V RGB LED optocoupler question

[jmbinette]

Hi to all,

I have an application where I am using a RF RGB LED controller to drive a High Voltage 110Vdc RGB LED Strip.

During my tests, I used a small lenght of the LED strip connected directly to the RF controller and all functions work perfectly using the remote : Colors change, DIM, etc.
The LED Strip controller DIM is a square wave PWM at 2kHz.

For my specific application, the LED strip will be 20m which consumption will surpass the capacity of the controller. I then found an amplifier (Manufactured by XUNATA) to boost the available power to 750W.
My problem is that the DIM function does not pass through the amplifier. I do not have an oscilloscope but I measured 2kHz at the input of the amplifier and 120Hz at the output.

I have reproduced, to the best of my hobbyist knowledge, one input circuit (RED) on EasyEDA. It is attached at the end of this post.
The source of the problem seems to happen at the 817C optocoupler. For some reason, the input (or output) saturates and cannot replicated the 2Khz.

I tried removing competely C5 thinking it was maybe creating a low-pass filter. No effect.

After reading a little about optocouplers, I figured that maybe I needed to play with the CTR ratio.

I tried increasing the opto input current by adding another 51k resistor in parallel with R4. No effect.
I tried decreasing the opto input current by adding another 51k resistor in series with R4. No effect.
I tried increasing the opto output current by adding another 51k resistor in parallel with R5. No effect.
I tried decreasing the opto output current by adding another 51k resistor in series with R5. No effect.

The only thing I found that made an effect, was to apply a 470ohm resistor across C5 in order the reduce the voltage across optocoupler pins 1 & 2 at 0.5V (Instead ov 1V). This has the effect of having a little DIM capability but not on the full range.

One weird thing I cannot explain yet : When DIM is 100% and I change colors, change of color is instant. When I lower the DIM and then change color, even though it does not reflect of the brightness of the LED, it does passes by a "dim" state before changing to the other color for a fraction of a second.

Any suggestions or something I am missing ?

ps. The amplifier states AC110V input/output but the output is really 110Vdc

[BillyO]

According to your schematic you have a .5V Zener across an LED that probably requires at least 1.4V just to get up to speed.  Why?  The way it's drawn the LED will never come on.  Did you make an error in the drawing?

[jmbinette]

I will certainly not rule out the possibility of drawing mistake ! Included a close up pic

I tried removing the zener also with no effect. Also removed the cap and only kept the 470ohm resistor and have very limited dim range.

Right now, even if it's not "full dim range" the best scenario is capacitor in parallel with the 470 resistor.

[tooki]

Not that it has anything to do with the problem, but am
I seeing this correctly, that the LED strip gets rectified DC that has no smoothing whatsoever? So it’s going to flicker at 120Hz (a horrible, flickery mess of a frequency) at every brightness level, even 100%?! If so, that’s horrible.

[BillyO]

You will never get 2kHz out of that amplifier.

Once the triac fires, it will stay on until the current drops to close to zero.  That only happens at 120Hz.  That amplifier was never intended to operate at 2kHz and just cannot.  Do you have a link to or a scan of the spec sheet for it?

[jmbinette]

Hi Bill,

Thanks for your input !

I don't really have an official spec sheet but I am attaching screenshots.

This is really the only product I found so for to do the job. Do you think I could replace the optocouplers or modify rhe circuit to attain a 2khz dim with amplification ?

I am also open to add some DC smoothing after the rectifier.

[BillyO]

No, changing anything to do with the optocoupler is not going to help.  In fact, that's probably one part you would not have to change.  I don't think there is any reasonable way to get this "amplifier" to work with a 2KHz controller.  Once triggered the triac is not going to stop conducting until the current through it approaches zero.  You'd need a 120Hz or 60HZ PWM controller that is simonized to the mains frequency.  In other words, one that turns on at varying times during the AC cycle .. working in effect like the old fashioned triac light dimmers.

The kind of "amplifier" that would work at 2KHz would be one that used a high voltage MOSFET, a filtered supply an a regulated low voltage supply for the control circuitry.  It would be far easier to find a PWM controller that worked at mains frequency.

[inse]

Is it possible, that the circuit looks like this?
I couldn’t make sense of the original drawing.

[BillyO]

Unlikely, but I don't have the device here to poke around on.  However, having said that, the 120Hz operation is a function of the triac (the big output device) and no amount of rewiring is going to change that.

[jmbinette]

Thanks, do you have a suitable example of such a device ? And/or just a pcb ?

Thanks !

[BillyO]

Unfortunately no.   How about the manufacturer of the controller, do they have a high current driver module?  Or do they mention one in their documentation?

Also, there has to be someone out their making a controller that works at 120Hz.

The thing with 120Hz is that it will flicker.  It will act like strobe light and could cause some people serious problems.  That's why decent controllers run at higher frequencies.

Something like this maybe? https://www.amazon.ca/XUNATA-110-130V-Bluetooth-Control-Controller/dp/B07NV8M55P/ref=pd_lpo_1

Although they do not say what frequency it is.  It is also Xunata.

[jmbinette]

In fact, it would be the other way around. The Skydance controller is perfect for my application and is standard across my other installations (Zigbee, RF remotes, etc.)

It's the amplifier that I would be willing to find a suitable replacement.
Even design/modify a 1500W controller to accept the 2Khz signal by the Skydance controller.

[Zero999]

Your "amplifier" is not an amplifier. It's a thyristor based switch and is completely unsuitable for PWM.

To dim 120V AC use phase control, which would involve a microcontroller and many other components.

Dimming 120VDC is easier, but it's still not simple to design an isolator which is fast enough to work at 2kHz.

[jmbinette]

OK, what about a setup like the sketch attached ?

I would use 2 separate controllers :
1 - RGB controller for solid color
2- DIM controller where I would control a 2khz 120Vac

The amplifier would use the RGB input signal as color selector and be supplied by a 120Vac input that is dimmed @ 2Khz

Is it true to say that it would work only for RED, GREEN and BLUE solid colors ? No in-between as the 2Khz is filtered out ?