ProtoDAC TDA1387 X8 project

[hifinet]

What happens when you let the browser with the wifi IP address open on the page for a while = Few minutes?

Times out with 'Site can't be reached' message. I tried on both mobile and PC browser, letting the browser run for a while.

Refresh a few times. My Zero W can need a few refreshes, or if you let it sit there, it should connect.

I've let is be for more than 30 mins ? no luck. i personally think there is some sort of network configuration that is getting wonky. Perhaps @Tim Curtis can chime in on where to look for the root cause.

Can you post the log ? m...Configure...System...Logs...download the Startup Log. Post it into the "code" window.

[varunach]

@hifinet I'm not able to get the Startup log because the RPi is unresponsive (can't access from browser nor SSH). But if I connect to ethernet it boots so I'm guessing the start up will look just fine. Is there another way to get the start up log from the SD card directly?

More perplexing stuff. My home has the wifi router in the living room. I'm able to connect to both ethernet and Wifi just fine when I connect to the Rpi4 ProtoDAC stack from my living room. But when I move the device stack to my office which is a bit farther and where this device is usually plugged in (about 25 feet from router) I can't connect to Wifi anymore. I also enabled 'Access Point configuration' to see if I can log in via http://172.24.1.1 if RPi4 can't connect to Wifi, but no luck. Moode interface is not accessible. This is the same location where the simple RPi4 + ProtoDac stack sat for the past month and I've been using Moode without any issues. Now with PurePi and Q7 it is acting up bizzarly. This is the same location where I have 2 Macs and 2 phones connected to Wifi and working just fine. I think I'm just gonna give up on this and chalk it up to weird internet blackhole.

I'm also wondering if there some sort of power line / ground interference in this room wall socket that is interfering with PurePi / Q7 clocks.

[OnkelPH]

@hifinet I'm not able to get the Startup log because the RPi is unresponsive (can't access from browser nor SSH). But if I connect to ethernet it boots so I'm guessing the start up will look just fine. Is there another way to get the start up log from the SD card directly?

More perplexing stuff. My home has the wifi router in the living room. I'm able to connect to both ethernet and Wifi just fine when I connect to the Rpi4 ProtoDAC stack from my living room. But when I move the device stack to my office which is a bit farther and where this device is usually plugged in (about 25 feet from router) I can't connect to Wifi anymore. I also enabled 'Access Point configuration' to see if I can log in via http://172.24.1.1 if RPi4 can't connect to Wifi, but no luck. Moode interface is not accessible. This is the same location where the simple RPi4 + ProtoDac stack sat for the past month and I've been using Moode without any issues. Now with PurePi and Q7 it is acting up bizzarly. This is the same location where I have 2 Macs and 2 phones connected to Wifi and working just fine. I think I'm just gonna give up on this and chalk it up to weird internet blackhole.

I'm also wondering if there some sort of power line / ground interference in this room wall socket that is interfering with PurePi / Q7 clocks.

I have experienced the same issue as you with Q7, so I’m just playing with the ProtoDAC at the time. I don’t have the PurePi though. I have a SD card with Volumio and that works fine with WiFi even with 3.3v on the ProtoDAC from Q7. One thing to mention, I logged on my router to see if the rpi was online, and it was.
I’ve just got som parts for a new 5v PS so my plan was to try again with the Q7 this week.

[varunach]

@OnkelPH Interesting. Yeah I simply couldn't figure out where things are falling apart because I can't even access the logs via the interface or SSH. In my case I can't even see the RPi when I login to the router. I have a ReceiverPiDDC, my plan is to build a RPi free DDC/DAC solution. I can then feed any digital signal to this including from a standalone RPi with Moode.

[OnkelPH]

@OnkelPH Interesting. Yeah I simply couldn't figure out where things are falling apart because I can't even access the logs via the interface or SSH. In my case I can't even see the RPi when I login to the router. I have a ReceiverPiDDC, my plan is to build a RPi free DDC/DAC solution. I can then feed any digital signal to this including from a standalone RPi with Moode.

In the Q7 ll setup it refer to DAC module Audiophonics I-sabre ES9028Q2M. In Volumio that works without any issues. 
I think receiverPiDDC is a good idea….

[hifinet]

@OnkelPH Interesting. Yeah I simply couldn't figure out where things are falling apart because I can't even access the logs via the interface or SSH. In my case I can't even see the RPi when I login to the router. I have a ReceiverPiDDC, my plan is to build a RPi free DDC/DAC solution. I can then feed any digital signal to this including from a standalone RPi with Moode.

But, you said you are able to access the RPi with Ethernet. If you can access with Ethernet, then you can get the WiFi IP from m...Configure...Network. Then power down the RPi. Disconnect power. Disconnect the Ethernet cable. Then reapply power, and you should have access through the WiFi IP.

You need to go through this procedure with Moode. If you have ever connected with Ethernet, and if you want to connect with WiFi at the next startup, you need to do it that way. See line 193 "1. CONFIGURE FOR WIFI CONNECTION" of the setup guide.

[Tim Curtis]

Couple things:

1. Linux manages all network connections way before moOde even starts. All moOde does during its startup sequence is check for an IP address assignment.

2. WiFi suddenly dying or becoming unreliable can be a symptom of interference. The infamous HifiBerry issue from back in the day where one of its clocks generated RFI that interfered in the 2.4G WiFi band is an example.

To troubleshoot you need to start from a clean slate.
Use your full board stack.

1. Start with a fresh moOde 8.3.8 image
2. Connect using Ethernet
3. Set "Output device" to "I2S audio device"
4. Set "Named I2S device" to "ProtoDAC TDA1387 X8"
5. Reboot
6. Verify that everything works

Troubleshoot WiFi

1. Leave Ethernet connected
2. Open Network Config and define your SSID and password (use SSID for 5GHz band first)
3. Reboot

Post the contents of the startup log and output from the commands below

1. Startup log
The log can be viewed via the SSH command moodeutl -l or downloaded via the Logs section in System Config, or via Menu, System info from the any of the Config screens. The startup log will be at the end of System info output.

2. WiFi info

Code:

ifconfig

iwconfig

[pbjacob]

Hi, friends. I'm a new guy here, and very happy to obtain an account.

I have two questions, please help me.

1. Which set of capacitors on the protoDAC has greater impact on sound, the output polypropylene capacitors or the Vcc electrolytic capacitors? 
2. If the decoupling capacitors have a greater impact on sound, then is it okay to not purchase too expensive output polypropylene capacitors (after all, these polypropylene capacitors are generally more expensive than electrolytic capacitors)?

Looking forward to friends who have listened to it can share your experiences. Thank you very much.

[hifinet]

Thanks Tim for your help.

The power supply decoupling caps affect primarily dynamics and soundstage. I have listened to quite a few including audio grade from Nichicon, Elna and Audio Note. I prefer the Panasonic FR-A with ESR of 41 mOhm. Fortunately, these are very inexpensive caps.

The DC blocking film caps affect overall clarity and transparency. I chose the WIMA MKP 4 (HHHiFi score 7-) to keep the cost of the kit reasonable. They are NOS and leads are not steel. I bought them for a good price. They are good enough to give the listener a good idea of the capabilities of the DAC. They don't have any annoying properties to the sound, and they are pleasant, but certainly not the best in terms of transparency. ProtoDAC with the WIMA has compared favorably to much more expensive DACs.

Audio film caps can get expensive very quickly. I personally cannot afford boutique caps in my DIY stuff, and I haven't listened to any. As posted earlier, polycarbonate (MKC) are generally better than polypropylene (MKP), and more affordable than modern MKP audiophile caps. The kits will come with Chinese Audiophiler MKP (HHHiFi score 7) when I run out of the WIMA.

I should mention that if you can verify that the next component in your system (for example, preamp) has input DC blocking capacitors (rated > 2VDC), you don't need the output coupling caps. You can substitute a wire.

[OnkelPH]

Couple things:

1. Linux manages all network connections way before moOde even starts. All moOde does during its startup sequence is check for an IP address assignment.

2. WiFi suddenly dying or becoming unreliable can be a symptom of interference. The infamous HifiBerry issue from back in the day where one of its clocks generated RFI that interfered in the 2.4G WiFi band is an example.

To troubleshoot you need to start from a clean slate.
Use your full board stack.

1. Start with a fresh moOde 8.3.8 image
2. Connect using Ethernet
3. Set "Output device" to "I2S audio device"
4. Set "Named I2S device" to "ProtoDAC TDA1387 X8"
5. Reboot
6. Verify that everything works

Troubleshoot WiFi

1. Leave Ethernet connected
2. Open Network Config and define your SSID and password (use SSID for 5GHz band first)
3. Reboot

Post the contents of the startup log and output from the commands below

1. Startup log
The log can be viewed via the SSH command moodeutl -l or downloaded via the Logs section in System Config, or via Menu, System info from the any of the Config screens. The startup log will be at the end of System info output.

2. WiFi info

Code:

ifconfig

iwconfig

I don’t know how I did it, but it’s alive. I haven’t done anything special in the settings at all. I was just going to test a new 5V PS, that didn’t work out well, but I’m up and running with the Q7 and moOde on WiFi. So now it’s running on 3.3v from the Q7 while I figure out the 5v PS…