So you run the dac async(clock comes from 100mhz oscillator onboard) and take from the cronus only d1, d2 and dck ?
Did you setup amanero in slave mode?
When you say low resolution files at what sample rate do you exacty refer? Do you manage to play both 44.1k and 48k files without any issues??
Can you check if you mounted on cronus the lower freq oscillator in x1 place?
Spdif is working as advertised at all resolutions using the onboard clock.
Cronus is connected with u.fl cables: d1, d2 and dck.
Amanero comes from TPA with the correct firmware.
Lower oscillatot is in position x1
44.1 plays as it should, anything else gives problems.
Everthing used to function as it should untill I pulled out the usb cable as I described in a previous posting. My guess would be the problem is somewhere in Cronus or Hermes...
Cronus is connected with u.fl cables: d1, d2 and dck.
Amanero comes from TPA with the correct firmware.
Lower oscillatot is in position x1
44.1 plays as it should, anything else gives problems.
Everthing used to function as it should untill I pulled out the usb cable as I described in a previous posting. My guess would be the problem is somewhere in Cronus or Hermes...
I had a similar problem although much more severe. See post 23 in the Hermes/Amanero thread.
I'm not sure what you mean by the isolator. The Potato is IC4 on the Cronus. It's a flip flop and all the data is processed through it as near as I can tell. The pins on that IC on my Cronus looked "dry". Touching them with an iron to re-flow the solder took care of my problem. My Cronus/Amanero combo has worked flawlessly since.
Hmm strange...
Tried to play today some 192k pcm files in sync mode and the lock led starts to flicker.
Doing the same thing in async it works.
In sync I can play dsd files and pcm files until 96k.
Measured all voltages under load(playing 192k and 96k pcm files) on the reclocker and buffalo which are all ok except the 1.2V which is 1.16V.
Could this be causing the problem?
Tried to play today some 192k pcm files in sync mode and the lock led starts to flicker.
Doing the same thing in async it works.
In sync I can play dsd files and pcm files until 96k.
Measured all voltages under load(playing 192k and 96k pcm files) on the reclocker and buffalo which are all ok except the 1.2V which is 1.16V.
Could this be causing the problem?
Hermes-Pi
Hello guys!
After experimenting with Amanero/cronus bundle on my raspberry Pi 2 (great sound, but usual troubles with usb on the pi when the network is in use...), I finally took time to try to adapt the wonderful hermes-cronus-rhea solution to raspberry pi i2s output.
First step was to write a linux sound driver accepting an external bitclock (as masterclock is not an option with raspberry pi)
Then build an openelec system to test it.
This is just a first attempt so only one sample frequency is possible. I set openelec to resample everything to 192 kHz, then use the Cronus on-board divider to get the right bitclock (49.152MHz :4)
And I use the Hermes-Amanero for isolation.
It works wonderfully !
I'll try to go a bit further with a little board dividing the master clock depending on the original sample frequency.
I'll use two idt 542 clock dividers, a 4:1 multiplexer like the 74AC151 and an I2C GPIO expander (mcp23008), selecting the correct bitclock on the fly with my custom alsa driver sending I2C commands.
I'll keep you posted when I'll have results.
(But no hurry, I'm slow...)
Cheers!
And again a big thank to Russ and Brian for this great modular solution!
Hello guys!
After experimenting with Amanero/cronus bundle on my raspberry Pi 2 (great sound, but usual troubles with usb on the pi when the network is in use...), I finally took time to try to adapt the wonderful hermes-cronus-rhea solution to raspberry pi i2s output.
First step was to write a linux sound driver accepting an external bitclock (as masterclock is not an option with raspberry pi)
Then build an openelec system to test it.
This is just a first attempt so only one sample frequency is possible. I set openelec to resample everything to 192 kHz, then use the Cronus on-board divider to get the right bitclock (49.152MHz :4)
And I use the Hermes-Amanero for isolation.
It works wonderfully !
I'll try to go a bit further with a little board dividing the master clock depending on the original sample frequency.
I'll use two idt 542 clock dividers, a 4:1 multiplexer like the 74AC151 and an I2C GPIO expander (mcp23008), selecting the correct bitclock on the fly with my custom alsa driver sending I2C commands.
I'll keep you posted when I'll have results.
(But no hurry, I'm slow...)
Cheers!
And again a big thank to Russ and Brian for this great modular solution!
Last edited:
Now what happens if you restore the setup to its "last good" (temporary) configuration? Perhaps a photo would inspire some ideas?
It is actually in the "last good" situation. Pictures are here (Hope you like aircraft)
https://www.flickr.com/photos/24454328@N03/
Scroll down a bit to see the pictures. Pictures are from some time ago but the measurements are still the same...
Situation is the same: playing about two 24-192 albums, then distortion, flickering lock-led then no lock...
Russ said it looks like something is running out of power but I have no idea where to look next (and I do not have a scope).
Only two elementary thoughts and probably not the 'root' issue... If simply running the usb source directly into the amanero still works, why not turn it 180 degrees and access it directly through the back panel, eliminating the internal cable? And second, I'm starting to get the sense that more lock problems are reported with rigs using Hirose U.Fl connectors than other routes for i2s. Of course that could be bad epidemiology and incorrect but I don't use them and I don't have problems. What if you experiment with mating the amanero and Buffalo edge to edge with 2cm jumpers (or other extremely short direct-soldered hookup)? Best of luck, sir!
I have the top of and a usb cable straight from the Auralic Aries to Amanero.
I've heard about the Hirose cable giving problems too (not in audio). What I didn't hear is if the problems were with the cables or with the connectors on the pcb. Perhaps you know?
I could mate Cronus and Buffalo close together but that would require a lot of effort... I'm not sure I am ready for that
I've heard about the Hirose cable giving problems too (not in audio). What I didn't hear is if the problems were with the cables or with the connectors on the pcb. Perhaps you know?
I could mate Cronus and Buffalo close together but that would require a lot of effort... I'm not sure I am ready for that
Trying to find a solution to my problems I had changed the controller of the dac as Russ suggested.
Now when I turn off the osf it works properly in both sync and async. Looks like it was a firmware issue with the original b3se controller.
Now I still have one problem to solve...
I use only i2s input.
When running in async everything works ok on pcm signals until 192k and dsd 256x and the controller shows up the right sample rate frequency
When running in sync I can stream files up to 96k(the controller display shows up 192k).For 192k the dac will lock only if I turn off the oversampling filter.It won`t lock to 384k signals.
Now when I turn off the osf it works properly in both sync and async. Looks like it was a firmware issue with the original b3se controller.
Now I still have one problem to solve...
I use only i2s input.
When running in async everything works ok on pcm signals until 192k and dsd 256x and the controller shows up the right sample rate frequency
When running in sync I can stream files up to 96k(the controller display shows up 192k).For 192k the dac will lock only if I turn off the oversampling filter.It won`t lock to 384k signals.