I don't want to pay $200, when i can pay $250 and learn something. Wireless audio by someone who doesn't know anything about it.
Socket showed up, looks awesome. Long pins on the bottom, I could probably get away with a spring hook connection to the required pins, but I think it should be simple enough to modify one of the many qfn breakout board designs to accept a socket instead of a chip directly, and then I can stick it in a breadboard. Open socket pics below:
The pics make it look big.
I know it has been a while, but I believe I may be able to move forward with this project again soon. I found a socket for the QFN40 package that wasn't $80+. When it arrives, I will see how I plan on breaking out a chip to test/program before putting it on a board.
As per my last post, I bodged some thin copper wire as pins onto the ccdebugger header pads, in an attempt to see if the pogo pins were my issue. I have discovered that they are at least not my only issue. Still no love from the purepath wireless configurator software, it doesn't see the chip still. I don't have an oscilloscope to dig deeper, but I have metered the voltage at specific points to determine whether my connector was backwards or something. Next step is a step backwards, removing all the non-comms components from the board, to make sure there are no conflicts there. The only other thing I can think of that may be the issue, is if my reflow didn't take everywhere, but I can't think of a way to test that. Ideas are welcome. If the component removal doesn't do the trick, I will have to make a breakout board for just the wireless chip, and attempt to get comms going with it.
Hey all - Sorry I haven't moved forward with the project. Got another stuck in my craw, with the ESP8266.
I have a question for you fine folks - I was wondering about the possibility that my problem with communicating isn't with my board layout, but rather with the connection. I believe i mentioned in a previous post that I have stuck some pogo pins in the end of the female connector on the ribbon cable from the CCDebugger, and it has recently occurred to me that my failure point may be the point - of the pogo pins. They are a spear tip design (this product) with a tiny point. The picture at the link doesn't do them justice, you have to visualize the dimensions shown. That single point of contact is tiny. Easily smaller than any of the trace widths on the board, possibly even smaller than the pads on the IC's.
Could I have a power choke at that connection? I may kludge on some sort of header and try to fire it up again before de-soldering everything.
Thanks for any input!
I have populated the board, reflowed using a toaster oven. It worked well, as far as the reflow process went. unfortunately, there are no blinking lights and no communications with the programmer. I am now working on troubleshooting the circut. If I don't get anywhere doing that, I will probably mount the CC8520 chip on a breakout board and test it that way, before attempting another full blown transmitter. I will probably also attempt to remove the chip with the hot air process as well.
Boards showed up. What immediately struck me is the bare metal at the antenna connection points. I didn't think to remove it from the solder mask, as they aren't really pads, just connections on the signal plane.
Boards are slated to arrive on Saturday, 1.17, and with a holiday weekend, I imagine I will be able to find out if my design at least communicates with the programmer. I have the Master/Sink board designed (it was a very small change from the Slave/Source board), but I believe it prudent to put my hands on the first board and get a feel for how it came out before I submit the next set to OshPark.
I woke up last night, concerned that the programming pads were incorrectly oriented on the board, but my concerns were invalid. A check this morning confirmed that all was well.
I did notice that the current configuration is for a Slave/Source. Slave units are better suited to be the battery operated end, so it looks like I will be creating the wireless guitar pack first instead of the headphones. Not a huge issue, but not what I had intended on creating first. I should have created a Master/Source unit, or a Slave/Sink unit. Next will be the Master/Sink unit, to complete what is now the guitar pack.
I have adjusted components to what I plan on using, and have re-done the PCB. It is pretty small, 26x42mm. My first unit is intended to replace a cable from my monitor station to my in-ear monitor on stage, so plug in location is not a huge issue, but if the unit operates well enough, I will probably re-design the PCB layout again, with the intention of running a 1/4" jack out the center of the back, for the purpose of plugging into my guitar, to keep it tight to the guitar body. This particular piece is getting a 1/8" jack for headphones, and it may be just a whip out of it, I haven't decided yet. PCB screencap below, as well as the download links for the schematic and pcb for diptrace. As before, the DRC error circles are all related to the antenna. Next step is to have it printed out and attempt a toner transfer method with it. All those vias worry me though, I might not have a choice but to have it done from a board house. Scratch that, I submitted it to Oshpark just to see what the price was, and I can't argue with $8.50 for 3 boards, including delivery, so I now have boards en route.
First draft of the schematic and board (just to get an idea of how small I can go, and to attempt to work out the details regarding the PCB antenna.
Schematic first run:
PCB first run:
All done in Diptrace. The DRC error circles are all related to the antenna, it is a part created of shapes on the signal plane, using the dimensions for the TI inverted F antenna. I couldn't figure out how to make it see all the shapes as one piece, either in the part creator or the PCB layout, so if anyone has any tips regarding that, I would happily utilize them. I know that it is missing some passive components yet and a couple of the parts on there are just placeholders for the moment, so the next step is fine tuning the schematic, adding all the passives and replacing parts with the ones I intend on using. I will probably completely re-do the PCB layout as well, I think it can be smaller. Currently it comes in at about 43.5mm x 29.5mm. I will likely make it the width of the antenna, and drop everything under that. They aren't labeled on this layout, but U1 is the CC8502 wireless chip, and U2 is the codec.
I have also gone through all options for the programming of the unit, the PurePath Wireless Programmer software from TI is really easy to use, it limits all options based on which chip you select, and what role you want it to play. Many of the options in the software affect what pins you use, so it was a good thing I was thorough.