Rank: Advanced Member
Groups: Member
Joined: 11/3/2010(UTC) Posts: 52 Location: Belgorod, Russia
Thanks: 6 times Was thanked: 1 time(s) in 1 post(s)
|
Hi guys,
I've connected master clock on new Cronus 1.3 to my two BIIISEs, via two u.fl cables (terminated ends go to the Buffalos, while the non-terminated are soldered to the MCK pads that is used for SMA connector I guess ). The firmware on the BIIISE is the latest from GIT.
I've tried both default switch settings, and with OSF off, DPLL set to high.
Upon turning on everything works fine, but after several minutes, music disappears. After power cycle music starts again, but for even less amount of time. After the next power cycle it won't come out anymore.
After letting it rest for some 20 minutes, it starts again fine, but losing lock again after some time.
Cronus and two BIIISEs are powered by separate Placid HDs. All have sufficient amount of shunt current. Buffalos are using new series Tridents. It all works fine in async mode of course.
What could be the problem?
Thanks, Fedor
|
|
|
|
Rank: Administration
Groups: Administration, Customer Joined: 10/24/2006(UTC) Posts: 3,979 Location: Nashville, TN
Thanks: 25 times Was thanked: 89 time(s) in 83 post(s)
|
Hmmm - you might need to observe what changes...
Dual mono is very difficult to wire.
Aren't the uFLs terminated on both ends?
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 11/3/2010(UTC) Posts: 52 Location: Belgorod, Russia
Thanks: 6 times Was thanked: 1 time(s) in 1 post(s)
|
Hi Russ, thanks for the reply. I've looked at the scope - the MCK pattern did not change (although it is not rectangular, like the other clocks. But that's probably my cheap scope won't allow such high frequency). Here's how it's wired: About the u.fl termination - I've been only able to find the u.fl to SMA terminated cables in my town, but anyway I wanted to make those connections as short as possible, so I trimmed and soldered them to the pad. Is there anything else, beside the MCK, that I could scope? Or measure with DMM? Edited by user Sunday, November 27, 2016 12:54:57 PM(UTC)
| Reason: Not specified
|
|
|
|
Rank: Administration
Groups: Administration, Customer Joined: 10/24/2006(UTC) Posts: 3,979 Location: Nashville, TN
Thanks: 25 times Was thanked: 89 time(s) in 83 post(s)
|
Thanks for the pic - that helped me understand a lot better.
Ok a couple of ideas.
1) The B3SEs are already terminating the incoming signals - so you can just jumper the output Rs on the cronus.
2) You should measure bit clock in particular - but I would take a look at them each if you can.
It never hurts to double check solder joints - even on the SMT parts - dry joints can be hard to detect.
Cheers! Russ
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 11/3/2010(UTC) Posts: 52 Location: Belgorod, Russia
Thanks: 6 times Was thanked: 1 time(s) in 1 post(s)
|
I've removed the output Rs on Cronus and reflowed all the SMT parts and Potato chips. Still no good. Another observation - when it stops working completely (after a couple of power cycles), the Mute LED on Buffalo is always on. But I guess it is how it should be, considering that there nothing to lock to :) Here's the clocks: Bit clock: D1: D2: Edited by user Sunday, November 27, 2016 7:45:02 PM(UTC)
| Reason: Not specified
|
|
|
|
Rank: Administration
Groups: Administration, Customer Joined: 10/24/2006(UTC) Posts: 3,979 Location: Nashville, TN
Thanks: 25 times Was thanked: 89 time(s) in 83 post(s)
|
Those signals look great... It makes me wonder if the loss might be at the source end.
|
|
|
|
Rank: Administration
Groups: Administration, Customer Joined: 10/24/2006(UTC) Posts: 3,979 Location: Nashville, TN
Thanks: 25 times Was thanked: 89 time(s) in 83 post(s)
|
Another thing make sure the DPLL is set very high. It should freewheel.
|
|
|
|
Rank: Administration
Groups: Administration, Customer Joined: 10/24/2006(UTC) Posts: 3,979 Location: Nashville, TN
Thanks: 25 times Was thanked: 89 time(s) in 83 post(s)
|
Do not turn off the OSF. :)
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 11/3/2010(UTC) Posts: 52 Location: Belgorod, Russia
Thanks: 6 times Was thanked: 1 time(s) in 1 post(s)
|
Russ, I've made some progress. I've replaced the new clocks with the ones from my old Cronus (they are the same 45 and 49 MHz Rheas) - and now the lock is fine - music is playing, but as soon as I do some action, like setting music on pause or manually changing track - the lock is lost again. (If the tracks are in queue, and changing by themselves - the lock stays put)
So probably you are right - something's happening on the BBB and Botic end. I'm now using Botic V7rc1, I'll try downgrading to v4 or 5.
Turned on the OSF, but that did not help :)
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 11/3/2010(UTC) Posts: 52 Location: Belgorod, Russia
Thanks: 6 times Was thanked: 1 time(s) in 1 post(s)
|
Well, unfortunately, downgrading the Botic did not help. I've reflowed the elements on Hermes just to be sure, but still no go. As long as I change the track, there is no music.
|
|
|
|
Rank: Administration
Groups: Administration, Customer Joined: 10/24/2006(UTC) Posts: 3,979 Location: Nashville, TN
Thanks: 25 times Was thanked: 89 time(s) in 83 post(s)
|
You might just retouch the Rhea pins a bit.
What player are you using? MPD?
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 11/3/2010(UTC) Posts: 52 Location: Belgorod, Russia
Thanks: 6 times Was thanked: 1 time(s) in 1 post(s)
|
I've spent all day doing anything to fix this - I've reflowed the Rhea pins, used aplay to play some wav instead of MPD, changing firmwares, etc.
But nothing helped. What helped is connecting MCK to the E_CK header(only clock pin, without ground) on Buffalos, not to u.fl. At first I thought that my cables are no good, but after some fiddling it became evident that reason for all this was connecting grounds of the two u.fl cables. It created ground loop, which I guess was interfering with the MCK signal. My ground scheme is star ground, so this connection was breaking it it seems.
So now I have just one question - is it ok to leave ground of the u.fl cable connected only on one end to the main system ground, wouldn't it work as an antenna? Or is it better to use just single core cable, like some twisted pair?
Anyway, I'm sorry I've doubted your and Miero's products :)
|
|
|
|
Rank: Administration
Groups: Administration, Customer Joined: 10/24/2006(UTC) Posts: 3,979 Location: Nashville, TN
Thanks: 25 times Was thanked: 89 time(s) in 83 post(s)
|
I actually just use a single stranded hookup wire not shielded at all. It works fine - it will be more of a transmitter than a receiver :) No worries. Cheers! Wiring is the hard part. :) Edited by user Monday, November 28, 2016 11:03:56 PM(UTC)
| Reason: Not specified
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 11/3/2010(UTC) Posts: 52 Location: Belgorod, Russia
Thanks: 6 times Was thanked: 1 time(s) in 1 post(s)
|
|
|
|
|
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.