Wireless Audio - Component Selection
There are a number of RF transciever IC's used for conveying data. They come in
all sorts of flavours. The feature rich IC's are usually based on the transmission
of data using spread spectrum and phase or frequency shift modulation techniques.
Bluetooth, Zigbee and Wibree IC's fall into this category.
For this project
Bluetooth based IC's are a possible contender. Bluetooth 2.0 is able to work at
datarates of up to 2.1 Mbit/s. Drawbacks, of bluetooth chipsets, for this application
is power consumption and cost.
Zigbee is intended to be used in applications that require low data rates and long
battery life. Additionally the Zigbee protocol supports mesh networking, so its suitable for systems
that require a level of reliability. The low datarate makes it unsuitable for this project. At a later date I have a project that will utilise
the ZigBee protocol...
Wibree is a competitor to Zigbee. It has been developed in part by Nordic Semiconductor.
Again this technoloy is not suitable fo this project.
Nordic Semiconductor specialise in low power wireless applications. After a brief
search of their website I found the nRF24Z1. This chip has been designed specifically
for transmitting high quality audio!
Some of the key features of the nRF24Z1 are as follows:
- I2S Support. (upto 24 bit @48kbps). The IC will interface directly into
standard audio DAC/ADC.
- Low latency (as low as 6mS)
- Low power consumption (3V Supply)
- Programmable (EEPROM or micro)
On the downside this IC works in the 2.4 GHz spectrum, hopefully its frequency hopping
algorithms are robust enough to work along side Wifi or Bluetooth.
For the transmitter module a suitable audio ADC is required. I chose the PCM1803A from Ti.
It was not my first choice due to the requirement dual supply rails (3V3 and 5V0), however
the part is readily available, low cost and low power.
The audio source is passed through a buffer to ensure the input impedance is equal
or higher than the ouput impedance of the guitar.
Although this project has been designed to interface into a guitar, which only requires
a single analogue input, I have put the electronics on to support a stereo soruce.
This will let me experiment with other audio sources.
I have assumed the power will be coming from a single 9V battery. To generate the
3.3 V and 5.0 V supplies I have just used linear regulators. I did consider using
switching regulators, but I was concerned about noise and the extra expense.
I may experiment with switching regulators at a later date.
The electronics around the nRF24Z1 are based around the data sheet recommendations.
A serial eeprom is connected to the nRF24Z1, which shall contain the data necessary
for the nRF254Z1 to operate as planned. The serial eeprom shall be programmable
in circuit using an external programmer. Details with regards to the content is discussed
later.
All important signals are routed to a 2.54mm pitch header. This includes the SPI
lines to the eeprom for programmability the I2C lines to the nRF24Z1,
to provide expansion for some type of daughter board and allow me to experiment
with a microcontroller and the digital lines DD0 to DD2, again to allow some experimentaion
The plan is to eventually put the PCB's into plastic mechanical enclosures, but
to begin with I will just make PCB's to test the hardware. The schematic for
the tranmsitter is shown to the left(click to enlarge):
The receiver is very similar to the Tx circuit, the main difference is that a DAC
is required instead of an ADC. The DAC I went for, is again from Ti, the part number
is PCM1773. It is configured using hardware only, so no need for a micro. For this
part there is only a requirement for a 3V3 rail, so only a single regulator is required.
The electronics around the nRF24Z1 is very similar to that of the transmitter.
The (provisonal) schematic for the receiver is shown above(click to enlarge):
Page 3