The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Digiface + 2 digimax FS Audio Interfaces
Old 15th May 2015
  #1
Digiface + 2 digimax FS

Here's the setup:

RME Digiface latest drivers/firmware

Two Digimax FS's

Only recording at 48/24 so no SMUX (sp)??

How do you connect them?

I've got it all plugged in and I'm getting clicking/popping jitter noise... I'm certain it's user error so I won't bore anyone with how I've set it up... I just am hoping the GS Yodas will just tell me how to connect it properly.

Thanks in advance for your help!

-b
Old 15th May 2015
  #2
Lives for gear
 

hey bgood,

I haven't worked with the exact same devices, but worked quite bit with RME devices and various external ADAT converters. Setup should be very similar.
Basically, there has to be one common sync source (without external generator, one device is chosen as a sync master) and other devices are set as sync slaves. Synchronization can be carried either over ADAT output from master or dedicated wordclock cable (bnc plugs). Finally same sample rate has to be set on your card and converters.

There are couple of options..
first ADAT sync without S/MUX, RME as a master
- connect both converters using four ADAT optical cables (2 cables per converter, connect each first socket for input and output section at Digimax back panel) to the Digiface
- set Digiface as Master at its software panel section Clock Mode and set your desired sample rate (48k) at your DAW project settings (setting will propagate to the RME card via its ASIO driver)
- power-up your DigimaxFS converters and set both units to external sync via ADAT input (underlighted switch should be red and LED rate should indicate 48k)

next daisy-chain Wordclock, RME as a master
- ADAT connection is the same as previous, maybe you can omit ADAT outputs from Digiface, if you need to use just A/D
- connect two BNC terminated wordclock cables in daisy-chain fashion. First cable from RME WC-out to the first converter WC-in and second cable from first converter WC-out to the second converter WC-in. Pay attention on termination button at Digimax, first unit should be unterminated (button isn't depressed), second should be terminated as there isn't any WC out connection from it.
- software settings for RME is the same as previous incl. DAW setup for rate
- hardware setting at DigimaxFS is same as previous except of external sync switch set to BNC choice (blue)

of course, it can be also set other way.. for instance one Digimax as system sync master with internal clock, its WC-out running to the second converter and then to RME's WC-in, 75ohm BNC terminator plugged to its WC-out.. RME then set to the AutoSync mode with preferred sync source from its Wordclock input. But you've got general idea about setup.
And final option is to use common external sync generator (like Art SyncGen or Apogee Big Ben) and set all devices to get sync from wordclock inputs.

I hope it don't sound like total mess

Michal
Old 15th May 2015
  #3
I was hoping you were going to give me a diff answer...

That's how I've set it up... I'm only getting pops out of my number two digimax so perhaps it's the unit...

I notice in the hammerfall settings that it's not showing BNC sync... Adats yes.

RME has great gear but I can't find updated docs on the updated drivers/etc
Old 15th May 2015
  #4
Lives for gear
 

Quote:
Originally Posted by bgood View Post
I was hoping you were going to give me a diff answer...

That's how I've set it up... I'm only getting pops out of my number two digimax so perhaps it's the unit...

I notice in the hammerfall settings that it's not showing BNC sync... Adats yes.

RME has great gear but I can't find updated docs on the updated drivers/etc
Excuse me, what from my post did you actually used. I've described at least three possible variants.
Also, what do you mean by "settings that it's not showing BNC sync" ?.
Choice for preferred sync source is called "Wordclock", do you miss that choice at control panel? Or did you connected correct wordclock input signal to Digiface and Sync check section there isn't showing status "Lock" for Wordclock item?

Michal
Old 16th May 2015
  #5
Tried the first few same jitter... Although I realized that it's only jitter from digimax 2....

I did some testing and found that the jitter I hear as I record is captured on the resulting program... Does actual jitter end up getting recorded? If not maybe its a bad trs cable

I'm going to try your second to last bit with the digi max 1 as master into the rme as slave...

Dedicated word clock is not an option

I'm going to double check cable fidelity... I'm hoping on finding a bad cable and have a quick fix. I'm suspecting one of my bnc cables is crap as I don't get rne telling me I have word clock sync regardless of which config I use.

On daisy chaining... If I add a 3rd digimax to the system --rme as master -- it's this correct:

Rme wc out > digi 1 wc in >
digi 1 wc out > digi 2 wc in >
digi 2 wc out > digi 3 wc in.
Digi 3 term button pressed in

Last edited by bgood; 16th May 2015 at 06:35 AM.. Reason: Sp correct
Old 16th May 2015
  #6
Lives for gear
IMO msmucr makes an assumption about how the FS's handle WC: they may not regenerate WC from in to out.

The proper and typical way to connect a master to multiple slaves is like this

Master WC ------T-connector---------/other slaves/-------T-connector +Terminator

...where the 3rd perpendicular arms of the T's are connected to the WC input of each slave device and where each of these is set to lock to WC input. The T connector and terminator isn't necessary if the last device has a switch to enable termination; it sounds like your FS's do. Other slaves should have termination off.

If the above doesn't work, then there's something wrong with the cabling, the connectors or the devices. BTW try not to skimp on the BNC connectors: cheap ones often have connectivity problems and can fall apart easily. You might also want to try adding and testing one slave at a time to troubleshoot.

BTW, you're using the word "jitter" to describe the wrong phenomenon. You should look it up. In fact, to minimize jitter, it's generally suggested that your A/D converter be the master clock... But that's another chapter...

Last edited by undertone; 16th May 2015 at 06:20 PM..
Old 16th May 2015
  #7
Thx...

Even assuming that the converters can pass along wc, is it better to t-split anyway?
Old 17th May 2015
  #8
Lives for gear
Quote:
Originally Posted by bgood View Post
Even assuming that the converters can pass along wc, is it better to t-split anyway?
It's a debatable point based on the quality or capabilities of the equipment you're using; for example if there's any delay between input and output, and you're wiring several devices from in to out to in to out, there could be a cumulative delay between the source and the last device that may prove significant; many will say that any delay is significant. Using T-connectors eliminates the possibility of that in-to-out delay so it's the safer option.

WC distribution from a master clock usually can be wired by sending a separate cable from the one of several outputs of that clock to each slave (each slaved device is terminated) so there's no possibility of cumulative delay; using T-connectors from one clock source as I described is the next best thing. BTW, it often gets cramped behind racks to connect the perpendicular part of the T to the BNC jack of the slaves: I've found using short male-to-female BNC extenders very useful for this.

Last edited by undertone; 17th May 2015 at 05:42 AM..
Old 17th May 2015
  #9
Noob question... But am I correct that I don't have to terminate if I'm using only adat to sync?

Another user here reported clicks/pops on the digimax fs when tracking above 44... When he worked at 44 there were no issues...

I tried switching down to 44 and the problem disappeared!

Got busy this weekend and have yet to pull the racks and switch out cables, but will report back when I do.

Thanks to all for responses!
Old 18th May 2015
  #10
Lives for gear
Quote:
Originally Posted by bgood View Post
Noob question... But am I correct that I don't have to terminate if I'm using only adat to sync?
If you're using ADAT to sync, you don't need to use WC at all, so termination is irrelevant; it's one or the other. But depending on your equipment, you might get lower jitter using WC; ADAT jitter is usually worse than WC jitter. In either case, you should also start considering making your main A/D converter your clock master as that will typically get you lower jitter in the conversion.

These decisions are based on knowing the specs of your equipment and their capabilities. I don't know the FS at all: is it rated for 48kHz operation? If it is and your cabling is right and your settings are correct from master to slave, then it should work.

When in doubt about clocking, try the different options for yourself and listen to the results. It could be that given your equipment and listening environment, ADAT sync is all you need.
Old 18th May 2015
  #11
Lives for gear
 

OP, any chance of uploading your Digiface's HDSPe Settings window when you have it setup for ADAT, like this:
Attached Thumbnails
Digiface + 2 digimax FS-digiface_settings.png  
Old 20th May 2015
  #12
Lives for gear
 

[QUOTE=bgood;11050268]Tried the first few same jitter... Although I realized that it's only jitter from digimax 2....

I did some testing and found that the jitter I hear as I record is captured on the resulting program... Does actual jitter end up getting recorded? If not maybe its a bad trs cable

I'm going to try your second to last bit with the digi max 1 as master into the rme as slave...
[QUOTE]

So this can be also problem of one particular DigiMax unit, I would try to swap first one with second daisy chained unit.

Jitter isn't correct description of the issue as undertone already mentioned. Jitter generally happens when clock frequency is modulated by another one (eg. it is unstable) and fluctuating around its nominal rate. In most cases (unless it is really extreme, so downstream device can't lock) it doesn't exhibit as dropouts or glitches.
What you're most probably experiencing is simply unsynchronized device from the rest of system - eg. its oscillator has slightly different frequency and when this difference exceeds point, where downstream device lose sync, audio dropout will occur.
In that case, if you can hear dropout in your monitoring, it will be also recorded.
Of course TRS cable can be also problematic, but it usually shows as complete lost of signal or some hum, due to bad contact at connector or intermittent connection at broken wire or shield. According to your description and later report at 44.1k it is very unlike to me.

Quote:
Originally Posted by bgood View Post
Dedicated word clock is not an option
I wouldn't be so steep with it. It is usually my preferred variant, direct connection to each device, no issues with improper loading with multiple clock receivers. Also some devices (experienced it several times) requires proper wordclock signal before powering-up otherwise there are issues with its reliable syncing. Usually generator is always-on or at least powered first, so no-issues with proper startup sequence of other devices.
It isn't be necessarily very expensive thing (eg. that small ART box or some eBay purchase).

Quote:
Originally Posted by bgood View Post
On daisy chaining... If I add a 3rd digimax to the system --rme as master -- it's this correct:

Rme wc out > digi 1 wc in >
digi 1 wc out > digi 2 wc in >
digi 2 wc out > digi 3 wc in.
Digi 3 term button pressed in
Yes, that should work for daisy chaining.

Quote:
Originally Posted by undertone View Post
IMO msmucr makes an assumption about how the FS's handle WC: they may not regenerate WC from in to out.

The proper and typical way to connect a master to multiple slaves is like this

Master WC ------T-connector---------/other slaves/-------T-connector +Terminator
....


I didn't blindly assume, but actually checked Digimax FS manual.
http://www.presonus.com/downloads/pr..._Manual-EN.pdf
Page 7 and 11. Where daisy chaining is recommended over T splitters.
There is also mention about equivalent conversion quality of master/slave setup.

Anyway, I don't see T splitters as the "proper" way for word clock distribution.
It really depends on particular devices and input impedance of its clock input and it is usually worst (if one check it with scope).
It such a device has non-defeatable internal 75 ohm termination (lots of computer interfaces or cheaper converters equipped only with clock input for instance), it will present excessive load to clock source, when T splitters are used. Result will be really compromised signal quality (level, skewed edges due to loss of bandwidth). On the other hand, most of devices equipped with WC input and output, where I had chance to see its guts, has also switched source to its WC output. When device operating in internal sync mode, buffered clock output was fed by internal oscillator, in external sync mode, input of this buffer was switched to external source. So propagation delay of this circuit is really minimal, definitely not enough for any sync issues. Buffer of course helps with loading of source with cables and downstream devices.

I don't argue with you, for some devices and setups it is good way, how to do it. But I usually try to avoid or limit amount of downstream devices fed by one splitted source in larger setups. Of course direct line from dedicated distributor/generator is best, but IME daisy chaining is commonly "second best" .

Quote:
Originally Posted by bgood View Post
Thx...

Even assuming that the converters can pass along wc, is it better to t-split anyway?
I don't think so.

Michal
Old 20th May 2015
  #13
Lives for gear
 

Quote:
Originally Posted by bgood View Post
Another user here reported clicks/pops on the digimax fs when tracking above 44... When he worked at 44 there were no issues...

I tried switching down to 44 and the problem disappeared!

Got busy this weekend and have yet to pull the racks and switch out cables, but will report back when I do.
It would be very odd, Digimax FS don't work correctly at 48k. There could be some bug with particular model, I've used Digimaxes D8 few times at 48k and never experienced any difference with its reliability over 44.1k.
Possibly you'll really find BNC cable issue or one unit with some hardware trouble (as indicated by your previous post).
After you successfully exclude those simple issues by some swapping and If problem at 48k persists, then contacting of Presonus support can be helpful maybe (at least for asking about possible 48k issues reported by other users).

Good luck!

Michal
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump