First, for fun and learning. Second, not amortizing my own development hours, lowering PCB BOM costs by not having to buy something commercial available. Third, maybe to do something for the community.
Forgot the fourth: also to integrate microcontroller code for the other peripherals.
Forgot the fourth: also to integrate microcontroller code for the other peripherals.
Last edited:
🙂
So excuse my naivety. But the ICs are like €10, then adding few € for SRAM, bunch of capacitors, resistors, USB connector and LDO does not arrive at €90-95 for an I2SoverUSB or Amanero?
Not including the PCB cost because it’s be integrated on the PCB for the DSP/DDC/DAC I would be making anyway.
Just tell me what I’m missing 🙂
So excuse my naivety. But the ICs are like €10, then adding few € for SRAM, bunch of capacitors, resistors, USB connector and LDO does not arrive at €90-95 for an I2SoverUSB or Amanero?
Not including the PCB cost because it’s be integrated on the PCB for the DSP/DDC/DAC I would be making anyway.
Just tell me what I’m missing 🙂
If you are talking about XMOS - you need EEPROM, not SRAM, which is also does not need if you will use XUFxxx chips.
And not including YOUR time cost, for developing firmware.
Thanks for the pointers on that XMOS variant.
And yes, donating my own time is what I meant with not amortizing my own hours.
And yes, donating my own time is what I meant with not amortizing my own hours.
In case of I2SoverUSB: isolators, CPLD, proper clocks. XMOS is >15€ with VAT. PCB manufacturing & shipping.
You are right, it’s good value. Though for my application I can do without the clocks and reclocking CPLD.
You are welcome 🙂
Don't forget that you need also to spend $27 on this:
https://www.digikey.co.il/en/products/detail/xmos/XA-XTAG/2183686
I2SoverUSB is EUR 75. That's really not bad for a quality solution. Amanero is good too, but I could wish for galvanic isolation directly on the board. Amanero is EUR 79, so about the same.
I've looked at cooking my own USB->I2C solution using something like an LPC43xx, but from what I gather UAC2 is half-baked at best from LPC and I would need to write drivers for Windoze, which I've never done. I'm also not a software guy, so even though I can write software I don't enjoy it. Paying someone EUR 75-79 doesn't sound so bad all of a sudden.
Tom
I've looked at cooking my own USB->I2C solution using something like an LPC43xx, but from what I gather UAC2 is half-baked at best from LPC and I would need to write drivers for Windoze, which I've never done. I'm also not a software guy, so even though I can write software I don't enjoy it. Paying someone EUR 75-79 doesn't sound so bad all of a sudden.
Of course not. Even at today's cost of steel and aluminum you don't get $70k worth of materials when you buy a BMW either. The idea that businesses should not charge more than the cost of materials is naive at best. Providing a complete solution adds value. That value costs money.
Tom
@roderickvd There is not available a DIY friendly USB-I2S device, based on a cheap MCU with simple opensource code, you can be first and make one 👍
From my perspective XMOS is DIY friendly:
The advantages of a DIY USB-I2S are obvious, but I think the biggest ones are:
- they provide free IDE
- they provide audio reference software with examples
- they provide free documentation
- they have a dedicated forum
- XU/F 208/216/316 chips available to order
The advantages of a DIY USB-I2S are obvious, but I think the biggest ones are:
- you can design it on the same 4 layers PCB with the DAC chip/s, so no fancy connectors (Amanero style 2x10 pin connector is not the best solution here)
- linear power supplies
- galvanic isolation (properly designed! on a 4 layer -at least- PCB)
Regarding USB solutions, it turns out Andrea Mori is working on a USB board that can run from something like Accusilicon clocks, it has connectors to hold a Simple DSD Converter board, and is galvanically isolated. It will also have Thesycon drivers for Windows. Includes u.fl connectors to drive a PCM or DSD dac. Don't when it would be available though. Just mention it since it could be a good low-ish cost option for people who don't want to design their own.
This is good USB-I2S DIY fully open project for USB Audio 2.0 : http://pavouk.org/hw/audiosystem20/en_at32uc3a3256usbi2s.html
It is atmel.
It is atmel.
Yeah that’s the SDR-Widget / Audio-Widget based one. Also this one by AMB: https://www.amb.org/audio/zeta1/
Should work very well for PCM, though no support for DSD and seemingly no longer under active development. So a candidate to breathe some new life into, when not taking an XMOS or similar route.
Thanks for the suggestions everyone, keep ‘em coming if you know more!
Should work very well for PCM, though no support for DSD and seemingly no longer under active development. So a candidate to breathe some new life into, when not taking an XMOS or similar route.
Thanks for the suggestions everyone, keep ‘em coming if you know more!
Pic of Andrea Mori prototype below. Don't know if the final version will look any different.
Believe outlined areas were where some ongoing work was still underway.
Believe outlined areas were where some ongoing work was still underway.
That's basically why I've started looking at USB->I2S solutions.
You could integrate the Amanero (or any other USB->I2S board) pretty easily by adding connectors and standoffs on your DAC board. Then just screw and solder the Amanero in place and you're good to go. Amanero also offers the option of licensing their circuit and code so you can integrate their solution on your DAC board. It looks like Schiit does that.
Tom