Sound Board Dead?

[Whitespace]

I have a sound board that I think died from ESD, as it used to work and then stopped. How could I test if it’s fried?

[wizard]

Hi @Whitespace ,
The main thing to check is that it shows up in the UI as Exp: SB1.0, in the top right status info.

If you have access to some tools, you could give it 3.3v power and check the TX signal for 115200 baud serial (like on a scope, or with a serial adapter or something). It should stream out data as long as it has power.

[Don]

Mine has done the same. It used to work fine. After the burn last year it’s been sitting for months unused. The other day I powered it up and I see the board listed on the screen so it does see it but no lighting changes when I select a sound pattern. I’ve reseated the board but no joy.

[jeff]

Hey Don - just to check, does the “SB1.0” in the menu bar go away if you disconnect it?

Also wanted to be sure you don’t have anything plugged into the line-in jack. It disables the onboard mic when a line-in is plugged in.

Last but not least, try enabling the Vars Watcher on a sound reactive pattern (in the edit page). Do you see values for energyAverage, and are they changing when noise is made?

[Don]

Hi Jeff,

Thank you for the reply. As luck would have it it now works fine. I haven’t done anything different. I powered it up to do the testing you asked of. For what ever reason it’s now working. There isn’t anything plugged into the audio in. I’ve never used that port.

Onto another topic, you had replied to me in the past when I asked if you can connect strings up in parallel. I believe you mentioned don’t do that. They should always be in series.

I have run 4 or 5 strings running in parallel without issue. However, I only run the sample .epe files that are posted on the site.

That said, how does the expansion board configure the output? If I attach 5 individual apa102 strings what does the PB see it as? Since each string is addressable wouldn’t that mimic a parallel configuration?

For the 102 or 9822 pixels on the expansion board would I share the clock signals?

For example, if I were to attach 7 strings that leaves one port (3 solder pads) that could be configured for the clock signals. There are three pads to source the clock signal but 7 strings of pixels. Would I share the clock signal to all 7 strings? After some experimenting it looks like only the one pad that is typically used for data is used for clock. It appears that one solder pad is shared for the other 6 strings.

I hope that makes sense.

Thanks again for your help.

Don.

[jeff]

@Don

For what ever reason it’s now working.

Great!

@Don

I asked if you can connect strings up in parallel. I believe you mentioned don’t do that.

If I said, that, sorry - I actually have a project I’m doing lots in parallel with right now. I know that it can glitch out in certain circumstances that can be hard to predict, especially with longer wire runs. If it becomes a problem, you can always use a SPI repeater or the expansion board.

@Don

That said, how does the expansion board configure the output? If I attach 5 individual apa102 strings what does the PB see it as? Since each string is addressable wouldn’t that mimic a parallel configuration?

You assign how each of the output channels is numbered, so that Pixelblaze still sees them as one long serial string (from an index perspective). You can technically skip some addresses (and I don’t remember if lets you repeat / parallel addresses). When I use it to replicate the same pattern on multiple strips or strands, I commonly do this in software.

Imagine I had 3 strips of 5 lights, and I wanted them to act as if they were parallel wired. Their actual index addresses would be configured in the expander setup page to go from 0 to 14 (0…4, 5…9, 10…14). I might write some code that makes sure the same color is rendered on pixels 0, 5, and 10 (and 1, 6, 11, etc).

@Don

For the 102 or 9822 pixels on the expansion board would I share the clock signals?

For example, if I were to attach 7 strings that leaves one port (3 solder pads) that could be configured for the clock signals. There are three pads to source the clock signal but 7 strings of pixels. Would I share the clock signal to all 7 strings?

Yes - since it has eight channels, you configure one of the channels to be the APA102 clock that’s shared / parallel across all APA102 outputs. That leaves up to seven APA102 data lines.

[Don]

Hi Jeff,

Awesome, Thanks so much for the input. Looking through past messages it wasn’t you that had replied . It was Scruffynerf. I ended up having to order two new Pico boards. One of the strings was reversed from the other so the power was backwards. It smoked the Pico board. That’s what I get for assuming all of the pixel strips are wired alike. They aren’t!
For anyone else reading this make sure you check your power polarities on the strips that you may be mixing. I think I was mixing apa102 with Sk9822 from different sources.