Ground loop ('frying') noise


When idle or even in standby, I can hear a “frying” noise coming from any of the line outputs of the UR824. The noise only disappears when disconnecting the audio cables from the interface, or when disconnecitng the USB cable.

I rebooted the Mac and the UR824, but the noise is still there. It can be heard when no application is running, as well as when there are audio applications running. So, it does not come from some audio application.

I never noticed this kind of noise with my previous interface (MOTU 828mkII FW). Frankly, I’m not even sure I heard it during the first test I did immediately after having installed it. But today I was working with soft-sounding material, and the issue became evident.

The fact that disconnecting the USB cable solves the problem, makes me think there is some clocking conflict between the Mac and the audio interface. Is there any way of diagnosing it?

My Mac is a MacBook Pro Early 2008, 2.4. The UR is directly connected to one of the USB ports, with no hub between the Mac and the cable. The ferrite sleeve is three-four centimeters away from the USB connector on the side of the Mac.

Does anybody else notice this noise?


I suspect it is not a digital noise, but some power interference. Here is what I did:

  1. Disconnect the USB cable from the USB connector on the Mac.
  2. Let the USB cable connector just touch the USB connector on the Mac (or a hub).

The noise begins immediately after I touch the two metal surfaces. No need to insert the plug into the USB socket.

I tried a different cable with no success. I tried alternative passages, and the noise is still there. And - surprise: the noise is not in the headphones.

In this case, what do you suggest? The ferrite sleeve supplied by Steinberg/Yamaha is there, around two turns of the cable. What else can I check?


I tried with some different cabling configurations.

The power plug was connected to a different outlet (including the one the Mac is connected to).
The power supply was placed in a different place (including on the floor).
Power cables were kept separated by the other power and audio cables of the rack shelf.
I tried different USB cables. Made the USB cables follow a different path.

No way to get rid of the frying noise.

The only tried I didn’t do are using balanced cables, and replacing monitors. I must say that these are the same cables and monitors I used for my old MOTU 828mkII, with not a single hiss or buzz.

I will purchase new cables. But I fear I will not be able to replace my Macke HR824 anytime soon (not that I would be compelled to, unless I can afford a pair of good Genelec or Adams).

If someone else hears this noise (apart for me and Sunshy) I would be happy to know. I wonder if this is a defect of the interface, or a problem that can be solved someway.



Luckily, the humm noise does not enter the inputs. Line inputs (Pad inserted) measured -144dB with the input volume pots at 3/4 (3 o’clock), and the Mic inputs (Pad non inserted) measured -80dB with the pots at the same position.


So, I replaced the unbalanced audio cables, connecting the UR’s outs to the studio monitor, with a pair of balanced mic cables (Reference/Amphenol, 55€ each - ugh!). The ‘frying’ noise suddenly disappeared. So it was certainly a ground loop noise.

I guess this noise was generated by both the UR and the MacBook not being grounded; the ground signal had no way of discharging. This did not happen with the MOTU I had previously, because it was fitted with an integrated power supply with proper grounding.

I question the reason for using a consumer, external power supply in a professional audio system. The UR is very big, and should have room enough to host an integrated supply.


PS: Incidentally, the output signal even improved with the new cables. Have I already told it? Despite these annoyances, this thing sounds fantastic.

Funny. One of my friends has the old, trusty Yamaha iX88. It causes ground loops, in the same system where an RME FF400 (also on an external power supply) and various MOTU interfaces have never caused any issue.