Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53  Location: romania Thanks: 2 times
|
When powered on the input impedance of b3 mck is more than 2000 megaohms(the maximum of my multimeter) so it`s fine. In all the photos yellow trace is for cronus output that goes to amanero and blue for the output that goes to the dac. In the bottom left side of some pictures you can see also the Vpp of the wave. In the first photos you can see cronus when powered alone. In each photo counter is set for a different channel.  3.JPG (111kb) downloaded 2 time(s). 3.JPG (111kb) downloaded 2 time(s).Here I powered up b3 also and attached it`s mck to cronus output. 2.JPG (118kb) downloaded 2 time(s).In this one I pulled up cronus clock selector so the clock is 49mhz this time. 5.JPG (120kb) downloaded 5 time(s).A curios photo is this one when I connect all b3 inputs to the cronus.In the other photos I was connecting only mck ufl. 6.JPG (113kb) downloaded 15 time(s).
|
|
|
|
|
|
Rank: Administration
Groups ready for retrieval: 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)
|
If it possible a supply is sagging? (Cronus supply in particular) The mck clock amplitude lowers significantly when connected to the DAC and looks very much like it is attenuated. That means it is either being loaded or has low supply voltage to begin with... What is the impedance of the clock input on the B3 with no power applied to the B3.
|
|
|
|
|
|
Rank: Administration
Groups ready for retrieval: 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)
|
Repeat the test you did above - but measure the clock signal at the Rhea as well as at the MCK output. This will tell us if the problem lies between the Clock and the DAC (say possibly the MUX chip).
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
b3 input impedance when powered off is ~120k Each rhea module outputs the right frequency. Yellow is mck/2 that goes to amanero and blue each rhea module 7.JPG (114kb) downloaded 14 time(s). 8.JPG (122kb) downloaded 14 time(s).Vdd at each clock pin is 3.3vdc. When I measure the dc at the output pin I get 1.6vdc unloded by the dac and 1.35vdc with dac mck connected.
|
|
|
|
|
|
Rank: Administration
Groups ready for retrieval: 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)
|
Does the clock at the Rhea continue to look the same when the MCK is connected on the DAC?
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
Here yellow is connected on cronus mck output that goes to dac and blue on each of the rhea modules accordingly to how I pulled the clock selector up or down. 9.JPG (126kb) downloaded 19 time(s). image10.JPG (129kb) downloaded 20 time(s).
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
In the last photos the dac was always connected to cronus.
|
|
|
|
|
|
Rank: Administration
Groups ready for retrieval: 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)
|
Ok - great info. Try operating with just the 45Mhz Rhea - take the other one off the Cronus.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
If I pull out the 49mhz clock( clock selector set low) at the outputs of cronus I will have 45mhz for b3 and 22.5mhz for amanero with dac mck disconected and mck on amanero output with dac mck connected. Instead if I switch high the clock selector I will get nothing at any of the outputs. I tested this with 45mhz clock pulled out too and I does exactly as above. Edited by user Thursday, March 3, 2016 5:50:40 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Administration
Groups ready for retrieval: 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)
|
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
I did test it with b3 connected and disconnected as written above.
Nothing changed from the past but at least now I know that rhea modules and clock selector are working fine in sync and async
|
|
|
|
|
|
Rank: Administration
Groups ready for retrieval: 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)
|
Ok thanks. So here is the next thing to try. Leave just the 45Mhz clock (this is important) and leave CS low. Connect a jumper from the rhea output to the MCK output of the Cronus while connected to B3 (this will effectively bypass the MUX) - What I am trying to do is find out if the MUX is somehow dead/dying Edited by user Thursday, March 3, 2016 11:31:32 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
I have connected b3 directly at rhea ufl without soldering any jumper to bypass the mux. Yellow is mck output to amanero and blue is mck output(at cronus output) First I have pulled out the 49mhz clock and connected dac ufl to the 45mhz clock ufl. Now what goes to amanero is right. image11.JPG (117kb) downloaded 5 time(s).The same thing I did to test the 49mhz clock. Pulled out the 45mhz clock and set high the mux. image12.JPG (116kb) downloaded 5 time(s).Since everything looks normal at the outputs I mounted the combo in the system and gave it a try to stream a 384khz file. I I had to leave both clocks on cronus because with only the 49mhz clock amanero won`t initialize properly and won`t start streaming.B3 mck input was connected to the 49mhz clock ufl directly. Still nothing. At this point do you still think it`s worth trying to solder the jumper you was saying? I read quickly your post and lost the part with the jumper 
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
For my peace of mind I have unmounted quickly the combo and soldered in the jumper image15.JPG (97kb) downloaded 12 time(s).Here you can see cronus outputs without b3 mck connected. image13.JPG (110kb) downloaded 7 time(s).And here is when I connect b3 in. image14.JPG (117kb) downloaded 8 time(s).
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
Is normal that b3 attenuates every mck signal that I connect it to? Edited by user Friday, March 4, 2016 7:08:54 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Administration
Groups ready for retrieval: 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)
|
No - it is absolutely not normal... and I can't say I have ever seen it before... :( It is as if the XO on B3SE is not correctly tri-stating. But I don't see how that could happen unless it is damaged. You already measured the resistance to the clock signal on the B3SE and it was high - so it seems to be something that happens only when the DAC is powered up. Almost as if there is some DC bias on the B3 mck net.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
I measured now dc bias on cronus output with a dmm and is ~45mv dc. If this matters.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
Hello again!
Few weeks ago after exchanging some messages with Brian he sent me another hermes/cronus combo.
Last friday the package arrived.
I had controlled and soldered everything with care and very much attention.
In the end I replaced the old hermes/cronus combo with the new one but I got exactly the same results as first. The dac will play at a double pitch when connected in sync with cronus.
I really need some help please to find out what is really wrong with my setup.
Now I have a doubt about the dac working properly.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 6/3/2010(UTC) Posts: 53 Location: romania Thanks: 2 times
|
Having little spare time these days I had disoldered the 100mhz vcxo that is on b3. Because I didn`t want to heat everything up with an air gun I used a soldering station with two irons each one with a double soldering tip and we were in two  The vcxo came down very quick. Now when I connect the dac mck to cronus mck and run sync everything works fine until 192khz. To stream 384khz files I have to turn off oversampling filter else the dac won't lock . The dac should stream 384khz files with 45 & 49mhz clocks in sync mode and oversampling filter on, right?
|
|
|
|
|
|
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.
