Sure, I agree they have to mill out the image, so as long as they can read your instructions/fab drawing, (Chinglishyou are good to go. Milling data is not part of the excellon drill file data.
I will take a quick look at your design, see if anything stands out.
The slots are in the board outline file. I've had them produced before on another project with no special instructions.
Okay, i see a lot of ecap symbols with wrong polarities. Having a polarizing ecap symbol does not necessarily mean you are using a polarized cap. It is a drawing convention issues, cap symbols with a curved pin, represents the aluminum can or negative polarity. I guess they are all 100n MLCC so i guess you are okay, I'll look for polarized ecaps in backwards.
U12,13 GP1FAV... do not need to be AC coupled.
Do they not go straight into a MCU port input?
I guess you could replace C28,29 with a series damping R of low ohms, 22-33ohm. They are std TTL signal levels I believe.
I used 100 ohm/1210/0.5W R's for R34,37
PCF8574's could be replaced with PCA9555, i have code for them as well. It is a big 24 pin dip pkg however.
I guess you have to use i2c thru a PCF8574 to make a SPI interface for the pga2311. Not sure why you need to do this however,it a PITA to program it, but if someone else has written the code, well i guess who cares. It is just a few pins from a mcu spi port, MOSI,SCK,SS plus the other control pins, ZEN, Mute.
That is it for now, lunch time.
U12,13 GP1FAV... do not need to be AC coupled.
Do they not go straight into a MCU port input?
I guess you could replace C28,29 with a series damping R of low ohms, 22-33ohm. They are std TTL signal levels I believe.
I used 100 ohm/1210/0.5W R's for R34,37
PCF8574's could be replaced with PCA9555, i have code for them as well. It is a big 24 pin dip pkg however.
I guess you have to use i2c thru a PCF8574 to make a SPI interface for the pga2311. Not sure why you need to do this however,it a PITA to program it, but if someone else has written the code, well i guess who cares. It is just a few pins from a mcu spi port, MOSI,SCK,SS plus the other control pins, ZEN, Mute.
That is it for now, lunch time.
The design is expandable to do pretty much whatever you would like it to do.
The Digital OI board will do two Coaxial SPDIF inputs and two optical digital ins. The control I'm using(LCDuino) won't communicate directly with the WM8805 so mine will only have a single optical input populated. The DAC board has an onboard port to connect to a WaveIO USB to I2S interface board Luckit | Audiophile grade DIY products to play HD audio files from a pc. This input is selected via an octal bus buffer and a relay by the LCDuino controller. The digital IO board has an alternate control port that can connect 4 more digital input. I think I have all the control busses in place to have this port connect to an advanced controller for possible addition of Rsavas' radio module and USB interface as well. Hopefully he can double check my work and let me know if iit looks like it will work.
The Analogue board will select DAC in or one of two Analogue coaxial inputs with an optional input expander board with the option for up to four more Analogue inputs. It has an onboard PGA2311 digital pot for attenuation. Output can be coaxial line out, headphones, or both.
I've designed my own LCDuino control board using Linuxworks' software but he asked me not to release it because he's selling his own boards for this. His LCDuino keeps growing as time goes on and getting more options all the time. He's done a good job on it I think.
I haven't made up any BOM's yet to get an accurate cost but I'm thinking this will end up between $300 and $400 with all the options populated plus supplies and chassis. I've designed it with higher quality op amps and there's quite a few of them so they add up quick. Running it bare bones like I want to will knock around $100 off the cost. Once the supplies are designed and the prototypes are built I will come up with a more accurate cost.
The preloaded Diptrace symbols for caps are all polarized drawings. The have a + on them if they are polarized caps,.
I'm trying to find the documentation I was looking at for U12 and U13 but they were showing AC coupling for those devices. It's a 1206 SMT package so a resistor could easily be populated there instead.
Linuxworks code is already written to control the PGA2311 via I2S and the Arduino doesn't have too many pins left so I ran it that way.
Thanks JW,
I understand about half of what you are saying, so I was just curious. I assume the LCDuino is some kind of Arduino clone. It seems strange that someone would use Linux as the source code for something like this and try and keep it closed source, that seems to go against the grain of Linux. I imagine it will be in the wild very quickly.
I understand about half of what you are saying, so I was just curious. I assume the LCDuino is some kind of Arduino clone. It seems strange that someone would use Linux as the source code for something like this and try and keep it closed source, that seems to go against the grain of Linux. I imagine it will be in the wild very quickly.
Linuxworks is a member here. The LCDuino is an Arduino LCD combination with a volume control pot and an IR reciever for remote control. The sofware is Arduino which is a cut down version of C.
Okay,
The Linux name threw me on the software control. C language then, or C++ or one of the variants. I see in other threads a big disagreement between using analog and digital volume controls, analog seems to have some real benefits but then you have to convert back and forth from digital.
The Linux name threw me on the software control. C language then, or C++ or one of the variants. I see in other threads a big disagreement between using analog and digital volume controls, analog seems to have some real benefits but then you have to convert back and forth from digital.
To elaborate , the PGA is just a good Jfet precision op-amp with it's feedback ratio
being switched by a digitally controlled resistor network.
To say it is digital , is like saying the slewmaster is digital because of the Arduino
in the protection circuit. "Analog with digital control " would be a better assessment.
Oh , the subjective assumptions never end ....
OS
Always happy to answer any question , Kind !
I was just disgusted by the lack of knowledge by some "analog only" members.
That PGA will even preform cancellation of the digital resistors noise. External
control is Embedded in the feedback loop and inside the chip.
A analog equivalent would have wire/traces , contacts .... many more opportunities
to be corrupted by the "dirty" (analog) world (hum - RFI)
.OS
Always happy to answer any question , Kind !
I was just disgusted by the lack of knowledge by some "analog only" members.
That PGA will even preform cancellation of the digital resistors noise. External
control is Embedded in the feedback loop and inside the chip.
A analog equivalent would have wire/traces , contacts .... many more opportunities
to be corrupted by the "dirty" (analog) world (hum - RFI)
OS
I know you aren't a fan of the NE5532 op amp. Should I be using an OPA device instead of the NE5534 in the buffer? I'm surprised TI didn't recommend something from the Burr Brown line for there and instead of the LT1028s.
Would it be worth adding a header to the analogue board for these connections?
Perhaps a little reading
. There might be a reason in some circuits to use a bandwidthlimited device.
You are talking about the DAC buffer/filters , right ?
OS
Yes I mean the DAC buffers. I remember reading TI's explanation how wonderful and versatile the NE5534 was and it kind of struck me like a sales pitch. That's why I'm wondering. I really don't know enough about op amps myself. I can figure out what pin configurations will work, but that's about it.Perhaps a little reading
limited device.
You are talking about the DAC buffer/filters , right ?
OS
I would think just making sure you have access, like a 40/50 mil test-point via pad would suffice.
Talking about buffers, the pga2311 requires a low Z source to obtain its thd+n spec. So for the external analog source(s), it is a good idea to use a unity gain buffer and add the i/p lpf there as well.
If you layout using ne5534an there are a few upgrades. Look at opa1611 as an upgrade. See fig 35 in the ds it is the exact same as the one in the pcm1794a spec.
One suggestion to make, is to do a dual footprint for the dac I/V converter ics, so-8 and dip8. This way if someone wants to use a opa1611 that is only in avail in so-8 you have it there and they do not need the adapter. You can try and put the so-8 on the backside, to see if it routes any easier. I think just to staggers them on one side would be fine, but if someone puts in a socket for the dip8, makes it hard to get at the so-8 unless it is on the back.
Even lt1468 looks to be a good sub for a ne5534a
Another note is that you are settling with a pga2311 rather than have the option of using a pga2320, having the larger dynamic range. This could also be an option, but you need provisions to have +/-5 or +/-15V supplies. yet another part to have dual footprint for.
Just ideas, things to ponder.
Sure the pga23xx has an internal gain stage(s) with max gain of 15.5 dBv IIRC. The point was distortion rises with increased source R, so you want to buffer, to guarantee a low R source. With external sources you do not have this low R guarantee.
If you do not want another buffer after the DAC I/V then buffer each ext source.
If you do not want another buffer after the DAC I/V then buffer each ext source.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.