Modular multi channel DIY AD/DA Box

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Well, that's why I decided to make it modular. The pinlayout of the modules will be published (for the DAC you see it in the schematic). I'm thinking of making something like a standard connection and so that everybody can design modules for it, a kind of open hardware  ::)

Raphael
 
Hi,

to keep this thread alive: The PCB company informed me that the prototype pcb of the DAC is shipped today.
So if everything goes well I can stuff the board over the weekend and test it.

Raphael
 
Today the pcb and the parts arrived. :)
As you can see on the picture I placed 6 of the modules on one pcb because I need 24 output channels. That was cheaper then ordering 6 modules as prototypes. So I can use the weekend to stuff the board and to test it.

Raphael

 

Attachments

  • DSCN3370_klein.jpg
    DSCN3370_klein.jpg
    255.2 KB · Views: 700
The inputs on every module are I2S. For the first run I'll connect it to a SHARC evaluation board because their I've direct access to I2S outputs. If successful I'll then proceed with making an ADAT interface (3x lightpipe = 24 channels) that I can connect it to an RME Hammerfall.
 
Just to let you know, I've successfully build the first module with 4 channels running. A signal was sent via an ADC->i2s into my DAC board and I was able to see the signal on the outputs with an oscilloscope.
I've to leave now, later this evening I'll post some pictures of the prototype, I think you want to see it. :D

Raphael
 
Here is a picture how the board looks after stuffing it (except the pinheaders).

 

Attachments

  • rkn0902_proto.jpg
    rkn0902_proto.jpg
    249.3 KB · Views: 567
And here under test conditions. The SMD soldering was done by hand with a standard solder iron. No special tools for SMDs were used. So, don't say it is not possible to do it yourself. It is and it is much easier then it looks and faster then wired parts. :)
 

Attachments

  • rkn0902_test.jpg
    rkn0902_test.jpg
    209.6 KB · Views: 475
Nice!!!

The SMD resistors I can understand, but the big chip? Did you solder that by hand as well... that's cool....

I'm planing on building more digital stuff so I'll be interested to know any tips and tricks you have on soldering those chips.

Nice work Raphael.

J
 
rkn80 said:
The inputs on every module are I2S. For the first run I'll connect it to a SHARC evaluation board because their I've direct access to I2S outputs. If successful I'll then proceed with making an ADAT interface (3x lightpipe = 24 channels) that I can connect it to an RME Hammerfall.

What are you going to use for the ADAT interface? The obvious choice is the Wavefront part but I don't know if they sell small quantities.

I have an ADAT transmitter design in VHDL. One day I'll actually test it.

-a
 
Matthew Jacobs said:
The SMD resistors I can understand, but the big chip? Did you solder that by hand as well... that's cool....

I'm planing on building more digital stuff so I'll be interested to know any tips and tricks you have on soldering those chips.

Indeed there is a trick for those chips. I learned it while I was working for a company making audio electronics: You have to clean the pads from tin-solder with desoldering wick if the pcb company tin-coated the pads so that they are really planar. Then you put the chip on it and make to solder points on opposited sides. The chip is now fixed and the pins should be on the pads. Then you start on one non-fixed side and put a lot  of tin-solder on the pads/pins like you want to make a big shortening over all pins. You do that on all sides. In the last step you heat up the tin-solder again but now you move carefully the solder iron parallel to the case side, the tin-solder will follow the iron and free the space between the pins. At the end you can remove the big tin-solder lump with your desoldering pump. Repeat that on all sides. You have to be careful that you you do not deform the pins but it works fine and at the end you have nicely soldered the chip on your pcb. Finally you have to inspect all sides (with a loupe if your eyes ar enot so good) if there are still some shortenings. If so you can easily remove them with the solder iron because the tin-solder will flow to the pins and pads and free the space between the pins. (I forgot the name of the physical phenomen that causes this behaviour).
That's it. With this method I had already success when I tried it the first time: A TSOP case and after doing that two or three times you'll find it easy to do.
Well it sounds a little bit complicated perhaps I should make some pictures when I do it for the other DAC chips. :)

Raphael
 
Andy Peters said:
What are you going to use for the ADAT interface? The obvious choice is the Wavefront part but I don't know if they sell small quantities.

I have an ADAT transmitter design in VHDL. One day I'll actually test it.

-a

Yes, I want to use the Wavefront parts. The main reason is that I want to avoid the need for programming devices. I think most people (including me) do not have the stuff to program a fpga. I know that there are some free fpga designs for ADAT (and other protcols). But if you are willing to do that, go head. Put a pin header on your board that has the same pin layout like my dac board and you can connect it easily to my DAC pcb. :)

Raphael
 
Thanks Raphael for the soldering tips, makes a lot of sense. Sounds pretty easy to do as well. Some close up pictures would help if you get the time of course.

Are your boards complete? have you tested them yet?

J
 
Well, I tested the 4 channels you can see populated on the picture. They are working so far. Today evening I tuned some filter componentes for a good frequency response. Btw. the DACs give now a maximum output voltage swing of ~21dBu (adjustable via resistor selection).

I see that there are different thoughts about the digital inputs. All of the mentioned the formats are possible as soon as a board is designed for it. Again I'll first go for ADAT (48kHz and 96kHz SMUX) so if someone is willing to design an AES/SPDIF board you're welcome.

Raphael
 
Back
Top