USB mass storage for Ultimate Source project - brainstorming end
I found two devices that could act as an USB Host, stream data through SPI and meet the speed criteria of minimum 200KB/sec needed for 44.1/16
1. Vinculum's FTDI-VDRIVE2
Pros: is always SPI Slave, has relative simple command set for FAT32 accessing, dev time seems low. It's cheap, ~25euro
Cons: support says it can read continuous 200KB/sec, only 10% more from the required transfer rate might not be enough. It could be possible with low fragmentation drives and very optimized code. This transfer rate would be enough for 44.1/16 only
2. FEZ Domino
Pros: supports says the SPI can give ideally 3MB/sec, so we could expect to handle ~700KB/sec enough for 96/24. It runs .NetMF, so development is easy
Cons: it acts as SPI master always thus the data clocking is difficult; some kind of protocol should be implemented between the slave dsPIC33 and FEZ and extensive buffering on the dsPIC side... development time needed unknown
These are just information. I'm still researching FEZ Domino as it would be nice to have 96/24. If I decide on one I will open a new thread for it.
Anyway, Koon's board will be needed for any of the above, however, in case of 48/96Khz a second XO is needed (easily solvable with a piggy back board and a relay)
Cheers, Zsolt
I found two devices that could act as an USB Host, stream data through SPI and meet the speed criteria of minimum 200KB/sec needed for 44.1/16
1. Vinculum's FTDI-VDRIVE2
Pros: is always SPI Slave, has relative simple command set for FAT32 accessing, dev time seems low. It's cheap, ~25euro
Cons: support says it can read continuous 200KB/sec, only 10% more from the required transfer rate might not be enough. It could be possible with low fragmentation drives and very optimized code. This transfer rate would be enough for 44.1/16 only
2. FEZ Domino
Pros: supports says the SPI can give ideally 3MB/sec, so we could expect to handle ~700KB/sec enough for 96/24. It runs .NetMF, so development is easy
Cons: it acts as SPI master always thus the data clocking is difficult; some kind of protocol should be implemented between the slave dsPIC33 and FEZ and extensive buffering on the dsPIC side... development time needed unknown
These are just information. I'm still researching FEZ Domino as it would be nice to have 96/24. If I decide on one I will open a new thread for it.
Anyway, Koon's board will be needed for any of the above, however, in case of 48/96Khz a second XO is needed (easily solvable with a piggy back board and a relay)
Cheers, Zsolt
Last edited:
Looks like very good solution for "all in one" player :
ADI ADSP BF592 Blackfin Ultra Low Cost 800 MMACs DSP : Electronics News from ElectronicSpecifier
ADSP-BF592 | Low Cost Blackfin with Optimized Peripheral Set for Industrial and General Purpose Applications | Blackfin Processors | Embedded Processing and DSP | Analog Devices
Two I2S and SPI , 400Mhz . Nice one for a new project , what do you think ?
ADI ADSP BF592 Blackfin Ultra Low Cost 800 MMACs DSP : Electronics News from ElectronicSpecifier
ADSP-BF592 | Low Cost Blackfin with Optimized Peripheral Set for Industrial and General Purpose Applications | Blackfin Processors | Embedded Processing and DSP | Analog Devices
Two I2S and SPI , 400Mhz . Nice one for a new project , what do you think ?
Came across this sd card player board - has ipod docking compat. and i2s out
Music Shield | Australian Robotics
Music Shield | Australian Robotics
Looks like an interesting project. Might try "porting" it to dsPIC33FJ64GP802 that I'm working with at the moment, see if I can get it to drive the onboard DACs direct. We also have UDA1355 S/PDIF codecs at work, can do I2S in for analog and/or S/PDIF out. Sounds like an interesting combo to play with.
you cannot buy afaik, you can only make it, i presume you followed my link. why have you made 3 posts above saying the same thing? we understood the first time
it is a DIY project with the emphasis on the Yourself part, there is not, nor has there ever been a kit or commercial PCB, formal instruction manual or forfmal support network, if you are looking for this, i'm afraid you will be disappointed
it is a DIY project with the emphasis on the Yourself part, there is not, nor has there ever been a kit or commercial PCB, formal instruction manual or forfmal support network, if you are looking for this, i'm afraid you will be disappointed
Last edited:
Is there a BOM and board files available for this project? It looks the cheapest and the only actual DIY one you can build yourself.
All the info you can get is here :
dsPIC33 and SD Card WAV to I2S, SPDIF Player "Ultimate Source" V2
No PCB, Gerber, BoM, ..., etc. available. So you have to generate your own.
Alternatively, this is as close to you can get :
QA-550--SD Card WAV Digital Audio Player
Disclaimer : I have no links with QLS.
Just a happy customer for a quick interim solution while building our own.
Patrick
dsPIC33 and SD Card WAV to I2S, SPDIF Player "Ultimate Source" V2
No PCB, Gerber, BoM, ..., etc. available. So you have to generate your own.
Alternatively, this is as close to you can get :
QA-550--SD Card WAV Digital Audio Player
Disclaimer : I have no links with QLS.
Just a happy customer for a quick interim solution while building our own.
Patrick
Thank you Patrick. I suppose I can build everything on a proto board with SMT adapters for the chips, and then I'd have to design an S/PDIF transmitter, and hope the firmware and building everything actually works! Ideally, an SD card player seems like the best transport nowadays. With parts becoming hard to find for broken CD transports, this solution only makes sense.
Looking at the schematic, it looks pretty simple, and not too difficult to accomplish. The issue is that as well as the dsPIC33, some other parts I won't be able to get through hole, and I'll need to get a programmer or build one myself. For the S/PDIF transmitter, I was thinking of using the DIT4096.
Looking at the schematic, it looks pretty simple, and not too difficult to accomplish. The issue is that as well as the dsPIC33, some other parts I won't be able to get through hole, and I'll need to get a programmer or build one myself. For the S/PDIF transmitter, I was thinking of using the DIT4096.
Last edited:
I received the breakout boards for all of the chips and soldered them on. Even after all these years I can still do it. None of them are aligned perfectly, but there are no bridges and no damage.
The chips are the dsPIC33 on the red board, a DIT4096 on the green board, and a 74VHC125 on the dark one.
An externally hosted image should be here but it was not working when we last tested it.
The chips are the dsPIC33 on the red board, a DIT4096 on the green board, and a 74VHC125 on the dark one.
I've started assembling the main board.
The pin header where the button is, is the programming header. I connected all power and ground to each other and PICkit software detected it. The two DIP 14 sockets are for the oscillator and the 74VHC125 and the DIP 16 socket is for the 74HC173. The DIP 28 socket is for the DIT4096, and I'm using a TOTX173 toslink transmitter. I can't get the software to build in MPLAB though.
An externally hosted image should be here but it was not working when we last tested it.
The pin header where the button is, is the programming header. I connected all power and ground to each other and PICkit software detected it. The two DIP 14 sockets are for the oscillator and the 74VHC125 and the DIP 16 socket is for the 74HC173. The DIP 28 socket is for the DIT4096, and I'm using a TOTX173 toslink transmitter. I can't get the software to build in MPLAB though.
Here is the perf board almost completed:
On the bottom left is the LCD header (I'm using instead a blue OLED), on the top right is the SD card header (I'm using a micro SD card instead). Only things that are not finished are the bypass capacitors of the 5V regulator, the wiring of the DIT4096, and the coax S/PDIF driving circuit. Also, the dsPIC33 has to be programmed. Has anyone been able to build the source?
An externally hosted image should be here but it was not working when we last tested it.
On the bottom left is the LCD header (I'm using instead a blue OLED), on the top right is the SD card header (I'm using a micro SD card instead). Only things that are not finished are the bypass capacitors of the 5V regulator, the wiring of the DIT4096, and the coax S/PDIF driving circuit. Also, the dsPIC33 has to be programmed. Has anyone been able to build the source?
I've made some progress. To build the firmware, use the version here. I programmed the dsPIC33, fixed an open between the reclocking section and main ground, and got this:
I'm not exactly sure the OLED display is jumpered properly for the right format, or something else is screwed up. I'm waiting for the micro SD card to arrive to see if I can get something to look like it's playing.
An externally hosted image should be here but it was not working when we last tested it.
I'm not exactly sure the OLED display is jumpered properly for the right format, or something else is screwed up. I'm waiting for the micro SD card to arrive to see if I can get something to look like it's playing.
CMIIW, VCC OLED too noisy, you need atleast RC filter for OLDE VCC. For firmware i compile using MPLAB Ide 7.60 and burnit with pickit-2 clone. This is my build
But no display not even black bar on LCD, even i turn the contrast trimpot. emuman100, can you send me your firmware?
Didiet
An externally hosted image should be here but it was not working when we last tested it.
But no display not even black bar on LCD, even i turn the contrast trimpot. emuman100, can you send me your firmware?
Didiet
- Status
- Not open for further replies.