Been working on my own controller board. Next up is my own DC SSR's I _hope_ to run all power through the controller fed by a typical PC power supply. I currently have 5V run out to the SSR's to feed the LEDs, but I may switch it to 12V.
Actually, after reviewing the pictures, I realized that I could not possibly solder the grounds to the molex power headers. I'll have to revise that. Not to mention the ethernet ports. So I'll revise this and post up Rev2. I will also probably switch from 5V out to the SSR's to 12V.
- Runs on olsen/grinch plugin
- For running LED strings on DC current
- (Theoretically) Can chain straight into grinch/olsen controllers
- Dim-able channels **
- Jumpers to run AC or DC SSR's
- Jumpers to disable dimming
**Channels can be dimmed in batches of 8. That means all 8 channels are dimmed simultaneously. Dimming is done with a PWM pulse. Work in progress.
Pics:



I need to develop a plugin that will output a PWM signal to the parallel ports. I know how to do it in C using usleep. PWM is just setting different on/off times. I will need also to add the ability to have the dimming of channels 1-8, 9-16, etc. joined together, either by interpolating different values for a set or forcing the same value for each set of channels.
Based on what you've written, I don't think that this is going to work, because the timing using usleep on whatever OS you're using isn't going to be sufficiently consistent. The OS is going to be preempting the usleep, causing your timed delays to vary all over the map, resulting in undesired flashing or dimming. Before spending any money on PCB prototypes, I'd suggest writing and thoroughly testing the results, using a logic analyzer to check the timing and consistency of the results.
Ultimately, I will not be using a C program to control dimming. It will be a Vixen plugin. I've recruited some assistance in designing the plugin, because it's far beyond my skills.
Rev2 is finished. I added LED channel indicators, changed the SSR voltage to 12V, and did my best to reduce the number of vias.



I've also been considering adding the option to use either through-hole or surface-mount LED drivers. Maybe in a future revision.
Attached are the PDF's for any fellow DIYers who'd like to make their own boards. 😀
My comment applies to any program running on a PC under any non-null operating system. Any program (including a Vixen plugin) is going to have trouble implementing a PWM algorithm in software, at least at any reasonable rate. Unless the PWM algorithm is implemented in those IC's on your board, you're going to have great difficulties.
I'll write up a quick C program for testing. Tomorrow I'll bring my laptop to class and hook it up to a scope and check on the consistency and accuracy of the results.
