l'd be happy with I2S output because that makes it easy to connect up a USB interface. I don't know enough to know what to suggest beyond that. How would you connect an ADC to a laptop? What's the best way? I have a USB 3.0 card, and USB 2.0 native on this laptop. I have almost zero experience with these kinds of converters. I have a MiniDSP USB interface, for example. I've used the Twisted Pear USB interface as well.
Oh cool. So, which ADC are you wanting to use? Did you want a separate thread for this? The PCM4222 looks great, but I'm not married to it...
http://www.ti.com/lit/ds/symlink/pcm4222.pdf
OPA1632 inputs looks good also.
I can help with a group buy or what ever.
AK5294A is what I' looking at.
I have a lot of research to do and I have to learn the Atmel procrssor.
This will take a while.
Just be sure it has good drivers for all OS (or class 2+ windows drivers). And support at least 192khz.
If you go back a few posts and look at what Demian linked to.
Use Arduino sure but the SDR-Widget doesn't use Arduino it uses the Atmel.
The project is open source and drivers are available for three OS and several OS versions.
I don't have time to create from scratch. It's easier to take code that's already developed and modify it for my purposes.
I don't mind learning new things but one at a time. Otherwise it gets too confusing.
Why is Arduino better? Why not Microchip?
Cheers,
Use Arduino sure but the SDR-Widget doesn't use Arduino it uses the Atmel.
The project is open source and drivers are available for three OS and several OS versions.
I don't have time to create from scratch. It's easier to take code that's already developed and modify it for my purposes.
I don't mind learning new things but one at a time. Otherwise it gets too confusing.
Why is Arduino better? Why not Microchip?
Cheers,
Last edited:
I can't promise there won't be headaches involved. But test applications can be tried before
committing to hardware. Hardware emulators.
The Audio Widget has most of what is needed. What remains to do is as follows:
1) use the USB module from Q N K T C USB-I2S Module and Analog Board
2) add ADC to the DAC (upgrade the DAC if needed)
3) add I2S isolation
4) add power supplies, not enough available from USB for this
I have worked out a number of the pieces and some is already in Borge's schematics. The rest is not really hard. Win interface would come from the ASIO effort, which would need an extension to work.
I must resist taking on this layout project for now. I'm too overbooked already. But I would be happy to provide the schematic pieces to assemble it.
A variation would be a really good recording interface as well.
1) use the USB module from Q N K T C USB-I2S Module and Analog Board
2) add ADC to the DAC (upgrade the DAC if needed)
3) add I2S isolation
4) add power supplies, not enough available from USB for this
I have worked out a number of the pieces and some is already in Borge's schematics. The rest is not really hard. Win interface would come from the ASIO effort, which would need an extension to work.
I must resist taking on this layout project for now. I'm too overbooked already. But I would be happy to provide the schematic pieces to assemble it.
A variation would be a really good recording interface as well.
Well i know xmos, via and amanero have no serious driver issues, the thing is to make it work with an ADC, rather than a DAC.
About the open source USB widget, I have no idea if it has windows drivers. Most people use a windows notebook I believe.
I'm considering both this (an used 0404 to mod) and the QA400 (also to mod), hoping for a lower noisefloor at HF, so it wont cover the harmonics. Both units dont have a state of the art ADC, but these modifications make them useful for audio testing.
About the open source USB widget, I have no idea if it has windows drivers. Most people use a windows notebook I believe.
I'm considering both this (an used 0404 to mod) and the QA400 (also to mod), hoping for a lower noisefloor at HF, so it wont cover the harmonics. Both units dont have a state of the art ADC, but these modifications make them useful for audio testing.
I think that it's sometimes easier (and more fun!) to start from scratch than to try to modify something that was bought in a store. I wish I knew more about digital audio to offer some help, but I find the effort so far very inspiring. Thank you. I can say that the PCM4222EVM doesn't appear to have a microcontroller on it. It uses microswitches instead I think. I'd be happy with microswitches, or better yet, a facility for front panel switches. I'm wanting to use this exclusively for FFT's, not recording or other purposes.
Attachments
Hi Demian,
I ordered boards for the SDR widget just to get me going.
Can't beat the price. I'll start with that.
I Got some good advise from George on how not to burn your nose when soldering smt.
The instructions on the web for FLIP are out of date. I can't find the device in the device list for the newest version of FLIP.
Best thing for me is to get fully engaged. You can save yourself a lot of time by pointing me to the best literature source. Then I don't bug you for the answers. When do you see your work load tapering off. The hardware is not difficult for me. My troubles are on the software side.
Where is the source code for all this? I can't seem to find it in the expected places.
Cheers,
Last edited:
The Emu 0404 uses the AK5394a, the best currently available DAC. The implementation needs help for measurements. And a well tuned clean system to work. It just barely works at 192KHz sample rate.
The XMOS driver is closed and expensive. The Via stuff doesn't work right yet. I'll look into the Amanero. The CMEDIA CM6631 works but is not really a DIY project either with NDA's around the firmware and driver.
The SDR Widget is already mostly done for what we are discussing.
Borge may be interested in making a board or device if we can get 40-50 people interested. Or you could buy a Lynx Hilo. . .
I don't know if the the driver issues have all been resolved with the Widget project but there are windows driver available using an open source ASIO driver. Apparently windows did not include the high speed USB as native in windows which seems kind of retarded so it's a make work project. It's probably the reason for audio driver problems with windows.
I suppose I could just mod the digital interface in one of the USB EMU's with a digital port and outboard the rest. This would not be difficult to do but for paralleling ADC's dsp is required and EMU has driver issues anyway.
There are some future projects, some ideas I want to try which require dsp and the Atmel looks like an inexpensive entry into dsp. DSP is probably one of the most expensive ventures to get into on the support side of things. TI and AD are not giving the software away. What I need is a multi port high speed I2S interface that goes in both directions.
I was on the beta program for the QA400 and dug into the front end hardware. I'm not sure there is much more to gain from the QA400. The QA400 has a target market with manufactures in China. It's intended to be simple to operate with out a lot of setup. It also had to fit a particular price point and for that reason things were not optimized as much as they could be on the input side. The power supply isn't there to do much with the input circuitry. It is what it is. The sound cards on the other hand.......
One thing that might work to lower input noise on the QA400 is to reduce the VCOM bias resistor from 100k to 10k or lower if it can be tolerated. If the QA400 is being driven from a low Z source and fundamental suppression is used the high input Z is not required.
It's not the R loading on an op amp output that's an issue its the output load current. Distortion will not rise significantly from loading if the signals are tiny.
If one looks closely at the data sheets for these ultra low noise op amps out there (1.1nV/rtHz or less) the noise rises significantly with source R. generally the noise is worse at 10k or 20k than it is for op amps spected at 5nV/rtHz. Jfet input op amp are usually better at high source Z. I'm not sure there is much to gain in modding the QA400.
A front end such as what Demian has done is what's needed.
Cheers,
Last edited:
Likewise. But I don't have the skills nor the tools to pursue such a project myself.
The akm is one of the best sigma-delta ADC, but is not really usable above 50khz.
A 18 bit r2r one would be more optimal if starting from scratch.
But I'm well fine as a start with improved emu or qa400.
You surely know more than me about driver status for usage in a custom project.
I have seen the via used in USB-i2s-spdif output interfaces and it is the best currently available jitter-wise. Do we care for jitter? Probably not. I'm curious to learn what problems there are with drivers.
Amanero could make some custom boards if the numbers are fine as he should be grateful to this forum for the many boards sold (i think close to 300).
The drivers are good, firmware is updated quick and the support is also great.
I2s is not isolated and we want to add that for an ADC.
The free asio driver for windows is utter crap, I wouldn't touch it with a 3ft pole.
E-mu drivers are gold compared to that.
Last edited:
This project will be exclusively the for purpose of measurement.
Just a high quality ADC stream. I will also look into forcing periodicity with the sampling. This would avoid the need for windowing the FFT. A first start approach to this would begin with synchronizing the beginning of the sampling with the zero crossing of the input sine.
I'll keep the boards modular so it's all plug and play.
Cheers,
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.