M32 DA (non-pro) clocking issue with UFXiii

I'm running a brand new Mac mini M4. No issue with the UFXiii. Also just tried ADAT and that is working perfectly although I'm limited to 16 channels. Yes, UFX is set to 'Internal' as the clock source.

Thanks

Yes, have tried all of those things but the clicks and wild metering persist. Interestingly, I switch to try 96k frame and back again to 48k and got my first sync message as shown here. However if I try to change the sample rate it starts switching around and ends up on 32k...

https://drive.google.com/file/d/1e67WpS … sp=sharing

I think he has the non pro

That screenshot clearly shows sync set to MADI input, means slave, means the setup can not work and the clock loop ends in exactly the described behaviour.

As said already UFX III should be on internal clock, the M-32 should be set to sync to MADI input. Sample rate changes then should be working at least between 44.1 kHz and 48 kHz. 32 kHz is a clock loop error and not supported.

Changes to 88.2 and 96 kHz should work with 96k Frame set to on.

Thank you guys, I do really appreciate your help. However, I think in my quest to resolve this I have tried and posted things that have just led to confusion.

My initial setup (the setup I would expect to work based on my experience) is shown in the attached screenshot.

I am not trying to run anything in 32k, it was automatically configuring it this way when I used the suggested 'MADI optical' as the clock source. I had not expected this to be the correct source but tried it to see if I could establish a lock (which I sort of did).

I had also tried a 96k frame but the unit itself will not allow switching to 96k for some reason.

Please note, I am getting sound through all 32 MADI channels so there is audio connection, and I'm getting a sync message on the Fireface USB Settings but still a green flashing sync light on the M32. I wonder if there is some compatibility issue with the UFXiii and the older convertors...OR, I have a faulty unit. I will be able to try a wordclock config tomorrow but can't get hold of the connector until then. It does seem odd though that MADI isn't able to sync the clock, not at all what I expected.

Thanks again

https://drive.google.com/file/d/1O7Df1D … sp=sharing

Hi MC, could you elaborate on the ‘tons of USB errors’ please. I’m not entirely sure what you mean as the screenshot states CRC errors as 0.

I have changed the frame to 96k in the usb settings and activating DS State on the M32 does now lock at 96k but I still have a flashing sync light (MADI STATE section of the M32 front panel) and a flashing green 48k LED and constant orange 96k LED in the Frame output section.

I have tried another USB cable and different ports but I’m still experiencing the same clicks, pops and wild metering. Really not sure what else I can try.

I think MC refers to this number.

https://www.dropbox.com/scl/fi/lw1dsiwd … g&dl=1

If you were using a Windows PC you would try all USB3 ports with the goal to find one which is more compatible to the USB3 transfer modes that are being used by the driver. Preferrably ports from the chipset or at least from 3rd party USB3 controllers on the mainboard, that are compatible, see manual. Alternatively to deploy another USB3 PCIe card with compatible USB3 chipset like FL1100 and others.

On Apple it might be a little bit different (HW, OS dependend), sorry, in that regards I have no other recommendation for you than to try all USB3 ports on your Mac mini M4. But I just see, this machine has only two USB-C ports in the front. So it is likely that the two are from the same chipset or USB3 controller. But simply try .. this can be done quickly.

There's no reason whatsoever for this not to work... One thing you don't really need in this setup is the MADI connection from the converter to the interface. This is just a DA converter, there's no return path. The unit should happily and easily sync to the incoming MADI signal. Have you got a BNC cable to test this with coaxial? Can you provide a picture of the converter's front panel? USB ports have nothing to do with this if performance is ok otherwise across all channels and if the errors don't reappear.

Thank you Matthias, I have indeed read the manual extensively over the weekend trying to work out why a simple MADI connection is taking up so much of my time. I understand it is easy as ‘experts’ to assume inexperience or simply the wrong button being pressed is the obvious solution but I have enough experience to spend a good amount of time troubleshooting and trying other scenarios (even if they don’t seem logical) before reaching out for help. I understand sample rates, doubling etc, my post clearly states that the only way I can get any resemblance of a 96k frame indicator is in DS mode but this will obviously not resolve my issue.

So…let me explain again my initial setup (how I would expect the convertor to operate) and the fault I am experiencing.

I have a current Mac mini m4 running OS 15.6 with USB C connection to the current RME FIREFACE UFX III. Firmware is v21, driver v4.27.

I am running an SC optical cable from the OUT on the fireface to the IN on the M32 DA. My fireface USB setting are:

MADI Output: Optical
MADI Format: 64 (32) ch.
MADI Frame: 96K Frame

Sample Rate: 48000 Hz
Clock Source: Internal
Current Clock: Internal

I was always getting audio with this config but the clicks and pops led me to believe there is a clock issue/lack of sync via MADI (only now do I start experimenting).

I have a flashing green sync STATE indicator. The FRAME select button does not allow me to switch to 96k (dispute my fireface being set to 96k frame). CLOCK is set to MADI on the M32 DA CLOCK SECTION.

No further settings should be relevant. This should work but it doesn’t. I will try word clock when I can get a cable but there is no other reason I can see why this is passing audio but not clock unless someone can tell me otherwise?

Thanks

P.s ADAT is working without trouble but I need the adat outs on the fireface for other convertors so not a long-term solution. It does however show that a clock sync is possible, just not through MADI currently

Like the optical cable...this is a bit unclear now, so - was Coax already tried instead?

As I wrote 96K Frame is for 96 kHz, not 48 kHz. While you can pre-select that for the MADI output in the Settings dialog, AFAIR on the unit it can only be activated with DS mode (aka 96 kHz) active. So this then would not be an error.

Optical cables can be bent and with this partly damaged. Also the sticking out ends can get damaged. MADI optical is a very thin optical fiber that needs to be positioned precisely to work. Dust/dirt in the MADI sockets can also cause trouble.

Of course the optical transmitters/receivers might be defective too. None of this is widely spread or 'typical', but it could happen.

Point was to test MADI coaxial (BNC), so set the M-32 DA to sync on that MADI input.