back again - is this the solution?

Clocking Multiple Audio Interfaces with a Mastering Clock
https://www.youtube.com/watch?v=N-j54LteSGg

using wordclock cable...

Its quite obvious that you need the 802 to get some analog ports and preamps, as the RayDAT is a pure digital card.
I would keep the 802 connected through USB to keep it operational in TotalMix FX.

You can either operate RayDAT and 802 in one TotalMix FX instance and then switching between them.
Another possibility is to open a new TM FX window (2nd one) and to select RayDAT in the 1st and 802 in the 2nd window.
If you save the workspace, then the window positioning will be saved as well along with it.

Usage of ports on the RayDAT.
For me it is not 100 clear what I/O modules you have on the M5000 and whether they support
- AES, coax SPDIF, ADAT or optical SPDIF (optical SPDIF is only possible on ADAT4 port of the RayDAT)
- whether the M5000 is able to be clock slave or not, if not then you need to take this device as clock master
  and all other RME devices will be the perfect clock slave due to steadyclock technology (802 and raydat).

A basic layout of the setup would be this:

PC------PCIe---------RayDAT<--------ADAT1----------802 (e.g. for analog ports of the 802)
                                          ----------ADAT1-------->802
                                          <--------ADAT2----------802 Rest of the ports: -mic preamps
                                          ----------ADAT2-------->802                              -headphones

For the M5000 the following ports would be free
                                          ADAT3
                                          ADAT4 (optical SPDIF)
                                          AES
                                          SPDIF (coax)
                                       
PC------USB3/2------802 (only for operation (TM FX) and to perform firmware upgrades; for that USB driver is needed)

Clock distribution (again):
- there can be only one master clock in a setup
- the master clock has the setting clock source = internal
- on the rest of the devices which are clock slave you need to configure the port where the clock signal will arrive or WC (word clock)

You do not need word clock and it also has no better clock quality.
You achieve the same clock quality by using the clock signal in the data stream of ANY digital port (AES, ADAT, SPDIF, MADI).

As your M5000 seems to have digital inputs and digital outputs it should have the possibility to get the clock source from its digital input, then you would not need word clock at all or as I said, make the M5000 clock master.

It might be an advantage to use the RayDAT as clock master in the setup (if the M5000 can be clock slave), because then all devices (clock slaves) in your setup would follow the RayDATs master clock if you change samplerate in your DAW between 44.1 and 48 kHz. Then the application/DAW sets the sample rate in the driver for the RayDAT and all slaves automatically would learn it. Surely an easier handling ...

EDIT:

#1 if 8 ports of your 802 would be sufficient, then you could even use ADAT2 additionally for connecting the M5000.

#2 I didn't find a good product description, I am currently not sure, whether all four I/O modules in the M5000 work independend from each other. It seems they own have their own digital inputs and outputs.
So it could maybe even be, that each of them has its own clock, but I am not sure. It was not fun to read some of the manuals through external manual providers, maybe you (should!) have more clue about your M5000 device. Sorry here I am out of time and fun to investigate deeper into the M5000 manuals.

Your current cabling / setup is still unclear to me, too many words/information, too few overview for my taste.

One little drawing is easier to understand the information in your posting and the screenshots.

Make an ASCII drawing that shows the connections, how you set it up now.
Add information which device is clock master and detail the settings of each clock slave ("gets clock from digital input xyz").

> I cant choose internal in 802, because its slaved to raydat
As I said, only one device can be clock master, fully normal.

i was thinking...maybe its the old fireface settings thats messing things up..802 was connec to that "back in the days" :-)

it is now connected to usb, only to see its totalmix

Hi
As far I can see, the digital (dsp) hardware option cards have an input and output pair of each XLR (which probably is for AES), chinch (SPDIF?) and optical. Now my question is: what protocol can be used on the optical connection? And to what of these protocol is it set on your units? Could it be there is some confusing with different kind of connectors and digital audio protocols?
The other thing which is not clear to me is: are you sure to have the signal routed correctly within the Fireface 802. If I understand it right, the signal should make this way:
Computer DAW out > Raydat optical (Nr?) out > 802 optical 1 in > internal Totalmix routing > 802 optical 2 out > M5000 optical 2 in
—Processing—
M5000 optical 2 out > 802 optical 2 in > internal TotalMix routing > 802 optical 1 out > Raydat optical (Nr?) in > DAW
If so, please check the routing on the 802 and also make sure to have matching settings for the audio transmission between all the devices. RME products normally can switch automatically to the right one given from an input. But some other products can’t (for example I have a very cheap interface (Zoom U-44) which has optical in/out only for SPDIF protocol. Sending an ADAT stream to it simply doesn’t work). How does the M5000 behave? And of course for the outputs it always must be set manually.

Now I also had a look into the M5000 manual and can tell you:
M5000 DSP module optical in/out is meant for consumer format, so SPDIF.
This for, in Fireface 802 settings you have to set the second optical output port ('ADAT 2 out') to AES/SPDIF and 'AES output' to consumer (instead of professional).
In TotalMix for 802 make sure you really have the signal routed to SPDIF hardware output.

Ok, that’s one thing. But the question is: does sound from Raydat and/or FF802 reach the M5000 for sure?
Oh, there is more or less every input going to every output on both your RME devices. I don’t know how much useful that is…? Don’t you work with the mixer view? Always matrix?

In the M5000 try another preset a basic simple reverb

back again, i just tried a xlr AUDIOCABLE to test it, it was the same sync as with adat cable - no difference.

i dont think a dmx cable would change that, so back to change my setup again :-(

Its all documented in the manual, you only need to read.

Summary:

1st of all, you need to differentiate between coaxial and optical SPDIF.

The RayDAT has three ports which can handle SPDIF connections:

1. the coaxial SPDIF port (only for coaxial SPDIF connections)

2. the AES port which can be set to consumer in driver settings and with a special AES to SPDIF cable you can use it also for SPDIF connections.

3. (only) ADAT4 can be switched to SPDIF (2ch) protocol in the driver settings. Its also called "optical SPDIF" because TOSLINK cables are optical connections. Which is very nice because there you have galvanic isolation and the cables can be a little bit longer. ~10m according to specs, but usually RME has selected TOSLINKs which work up to approx 15-16m.

ADAD  = 8 channels at 44.1 - 48 kHz
S/PDIF = 2 channels at 44.1 - 192 kHz

Still waiting for a drawing giving a good overview about your setup.
Your thread is becoming quite lengthy (expected).
Therefore I requested this.
You can not expect that others do all the work for you.
If you do not deliver slowly, then sorry, I am simply out of this thread.
I am customer like you and its your environment / problem, so you should know what work is on your side...
Cheers.

Thanks kim, but if you make a drawing like this then a few things are still unclear / not visible:

1. there are no device names, the back of a device doesn't tell me what device this is

2. You should mark the device which is the clock master

3. in the cabling, the names of the ports are missing. It is better to name them like e.g. ADAT1 OUT or ADAT1 IN"
This provides the benefit if you check your clock slaves setting, that you directly can see from the drawing what's the exact name of the digital input port so that the device becomes a proper clock slave.

4. if ports can be configured for different protocols it would esp. in your case be very beneficial to write to such a connection, which protocol has been / or should be configured. To put two examples from memory:
a) RayDAT ADAT4 which protocol is in use: whether ADAT protocol (8ch) or SPDIF protocol (2ch) which was definitively needed in one case
b) RayDAT AES port: whether professional or consumer (SPDIF) has been configured in the driver settings. I think in one case you might have needed such a connection with a "XLR to Cinch" cable (the usual pinout for such a cable as outlined in the RME manual).

In such an (already quite complex setup) you should provide such an overview not only us.
Mainly for yourself how the setup should be.
Then such a drawing showing all relevant information can act for different things:
- design phase
- deployment phase and
- troubleshooting, if a connection does not work as expected.
During troubleshooting it may be the case that you find out that the design needs to be changed in certain aspects.
Then you update the drawing and have it much easier, in case of re-cabling to check each of the connections
and to configure the proper clock and protocol settings (if needed like told above).

Without such a clear plan you everybody is lost. I made already many of such drawings, but do not have the time anymore and in this case I had no fun to go through all the infromation again and again to provide this, therefore I asked you to provide, but maybe it was not clear enough what information is needed in such a diagram.

Therefore I want to provide you a few examples from threafs in the past, maybe you can refine beased on that information.
Many thanks for your efforts and patience.
With such a diagram it should be much easier for everybody involved to give you a good support, because the skill is here in this forum. But please be so gentle to accept, that also us customers have our private life and something like this is simply time sucking ..

https://forum.rme-audio.de/viewtopic.ph … 91#p181491
https://forum.rme-audio.de/viewtopic.ph … 39#p189939
https://forum.rme-audio.de/viewtopic.ph … 73#p190273
https://forum.rme-audio.de/viewtopic.ph … 23#p191423

While going through the devices and connections ...

If you could change the layout of your setup this way, you would not have the SPDIF problems (bridged) that Oli mentioned.
I may be wrong, but it looks to me like you lost track of the whole setup, because there was no overview and if you know the connection between the devices better, then you can make better suggestions, which may already be the better solution.

So please check, whether this could become your target setup with a few optimizations from my side as outlined below in the text.

                                                                               
                                                                                - 1x AES IN||OUT Maybe for DAW to be used as external FX
                                                                                     Clock Slave through AES IN (CM: RayDAT)
                                                                                - 8x Anal. IN/OUT Don't know what the purpose of 8 Anal. Ports might be
                                                                                 Eventide H9000
                                                                                 || || |||||||| (1x AES I/O, 8x Analog I/O)
PC+--USB3/2-----[only for operation]-------------------RME Fireface 802
PC+--PCIe--RayDAT---ADAT1 OUT---[ADAT, 8ch]--->ADAT1 IN----/
                    (clock-     ADAT1 IN<---[ADAT, 8ch]---ADAT1 OUT---/
                    master)  ADAT2 OUT---[ADAT, 8ch]-->ADAT2 IN----/
                                   ADAT2 IN<---[ADAT, 8ch]---ADAT2 OUT--/
                                  ADAT3 OUT----[ADAT, 8ch]--->ADAT IN-----------ADA 8200 (Preamp and AD/DA converter)
                                  ADAT3 IN------[ADAT, 8ch]---ADAT OUT-----------/                                      | | | |       | | | |  2x2 Analog IN/OUT
                                  ADAT4 OUT----[SPDIF, 2ch]---optic. SPDIF-------TC M5000 [DSP1#1]    [ADA2#1]  | | | |
                                  ADAT4 IN -----[SPDIF, 2ch]---optic. SPDIF--------TC M5000 [DSP1#1]   [ADA2#1]  | | | |
                                  AES OUT-------[AES, profess?, 2ch]---->AES IN---TC M5000 [DSP1#2]               [ADA2#2]
                                  AES IN<-------[AES, profess?, 2ch]----AES OUT---TC M5000 [DSP1#2]               [ADA2#2]

802 only connected to USB to be operational (TM FX)

RayDAT is clock master, clock slaves get the clock signal through their digital inputs
- H9000 from 802 AES IN (who is slaved to RayDAT)
- 802 from RayDAT through ADAT1 IN
- ADA8200 from RayDAT through ADAT3 IN
- M5000 DSP#1 and DSP#2 from AES IN and optical SPDIF (ADAT4) IN

I would try to connect as much as possible to the 802 to utilize more of RME's excellent analog I/O (high quality, transparent)
Therefore use two of the ADAT ports between RayDAT and 802 to have in total 16 channels for connections.

On the RayDAT I decided to use a combination of AES and optical SPDIF port (ADAT4) so that you do not get the limitaiton mentioned by oli77sch that coax spdif and optical spdif ports become bridged.

I am not sure what the purpose is to use analog and digital ports of the H9000, I simply assume its the way it works.
In the DAW I would define at least three external FX
- H9000 via 802
- M5000 module #1 and #2
The DAW will send a "ping" through the device and find the latency for correct compensation. Keep in mind to disable any delay effect as this will lead to an incorrect measure, perform measuring with activated "bypass" (100% dry). The delay introduced by two digital channels (ADAT, AES) will be minimal and can be ignored (only a few samples).

The ADA8200 you can use as Preamp and AD/DA converter.
I decided to connect the analog ports of the TC M5000 to the Behringer, this leaved you 4 ports to be used as Mic Pre.

EDIT: AES connection between RayDAT and TC M5000 DSP1 module #2 assumed to have professional format (to be validated).

You/we can use this ASCII drawing now to get a better understanding how the devices will be connected.
Who is clock master, who is clock slave.
Whether the AES ports are running in professional or consumer protocol (SPDIF).
Whether the ADAT ports are running in ADAT protocol (8ch) or whether optical SPDIF is in use (only ADAT4 of RayDAT).

Please comment on my cabling proposal. If you like it, cable this way and perform the required routing through RayDAT / ADAT / 802 and required protocol settings in the driver settings dialog.

Either do so or please document your current setup in exactly the same way like I did (ASCII drawing containing all relevant information for cabling, ports and clock synch).

We need to get an easy / common understanding about your cabling / setup based on such a drawing.

[/quote=oli77sch]

I tried this, there is no sound in module 2. only sound from module 1 which is from raydat´s spdif adat 4 or spdif coax

it was the one where i wrote its like 802 are missing some settings, unlike raydat

yes i got sound from 802 AES! i does sound weird maybe because of wrong cable...what did i do, i be back

back, i atually dont know what I did since it suddenly worked, but i remember i did a full reset on both totalmix

I did mess with all that DO-TYPE* AES/EBU etc., with no luck, i wrote that earlier, ´with the "--"

I orderd 1 meter, and 12 meter AES cables which arrive this week, they wasnt expensive

latency ok, but 802 is older than h9000. h9000 is a lot more than just a converter ;-) but i listen and will think it over regarding connect to 802.
there is a madicard in my h9000

im very happy that i now can skip the analog patchbay and do it all OSD

yes i know how ridiculous it is to talk about latency :-)

"A millisecond (ms) is one one-thousandth of a second. To put this in context, the speed of a human eye blink is 100 to 400 milliseconds, or between a 10th and half of a second"