HQ Wireless Audio

Not much to add, just commenting that I did end up purchasing a CCDebugger, as it was much less likely to fry my chips than my hand soldered fine wire idea. I will add header pads to the pcb, I also ordered some pogo pins to facilitate it.

Found out that I don't have the means to produce a PCB footprint that is sufficient for this chip at home. I am currently using the toner transfer method. It might be possible to have it printed at a kinko's or something to produce fine enough traces, but then I am also unsure about whether it will etch properly. I contemplated printing it any way (the pads run together a bit) and then cutting between the pads with an exacto, but I have opted to attempt to solder individual wires to the pads. See new picture for progress. I may scrap that whole method and just have some boards printed up after going over my design with a fine tooth comb. We'll see.

So the CC8520 wireless audio chip showed up today. This thing is dang small. Looks like I get to learn another thing - reflow soldering. I have a nice heat gun that I may be able to do it with. Picture added to page, q-tip for scale. Because it was closer than a ruler.

Further reading has made me realize that I need something specific to do the programming of the wireless radio chip. The Pure Path Wireless Configurator calls for the CC Debugger, a TI product for SPI programming. Retails for $50, and I found a few clones for around $30-35. The configurator help file lists a couple of other programming methods, one of which still requires the CC Debugger. The other is direct access to the chip's bootloader through the SPI interface, using the command set defined in the user guide. This is the path I will attempt first, as I believe I can utilize stuff I have here to get it done, namely an Arduino Duemilanove. I have not delved into SPI yet, so this should be interesting.

I have learned that if I want to learn how to do something, there are at least 3 other things I will have to learn how to do, just to get to what I originally set out to learn.

Update to goals-

I believe I know the amount of latency that will be introduced by the wireless communication, which is somewhat configurable, depending on a number of programmable factors. On the low end, it looks like 16ms is achievable, and expandable up to 64ms. I am having trouble finding any reports online about how much latency is allowed before the human ear picks it up, I saw some reports that stated as low as 0.1ms, and others that ranged far above that. One of the most compelling pieces of info on this subject that I came across was a post by a TI customer in their forum, in which the customer was unhappy with a noticeable delay using a setting of around 42ms. The TI rep stated that for lip syncing purposes, latency around 20ms was desirable. The CC8520 chip trades wireless range and connection robustness for latency. Previous attempts at a project of this nature that I have done involved a bluetooth radio, which introduced a delay of about 150ms, and was unusable in a live environment, so at least the maximum is well below my previous attempt. This does not take into account the latency introduced by the CODEC, but I believe that is in the realm of 0.5-1.5ms.

Project started on Hackaday.io, to track my progress. I have ordered some parts from TI, and will post pictures of them when they show up. I need to learn how to solder an exposed die. I also need to figure out how to prototype with the chip, without making it unusable in a final custom PCB. I have begun reading through the datasheets for the CC85xx wireless audio chip and the TLV320AIC3204 audio codec chip, as well as downloaded and begun poking through the PurePath wireless configurator software. I will post items of note and intended paths here, until I actually get to work with the chips themselves.