You are here

Focal + RME gain structure

All about the tools and techniques involved in capturing sound, in the studio or on location.

Focal + RME gain structure

Postby Domien » Mon Sep 09, 2013 8:05 pm

After a long period of saving I've finally upgraded all my stuff and invested in an RME fireface ufx interface with a pair of Focal Solo6 BE monitors. There's no volume to be set on the focals since they always output at unity gain. I like this a lot since this takes the thinking about gain structure away concerning the speakers.

But now I'm a bit puzzled by the output options of the RME. I can set the output level to 4 options: -10dBV, +4dBu, High Gain and +24dBu. I always assumed that since all my gear outputs and accepts balanced signals that everything should be set to +4dBu. Is this a correct assumption, and when would one then use the High Gain or +24dBu settings?
Domien
Posts: 2
Joined: Tue Aug 11, 2009 11:00 pm

Re: Focal + RME gain structure

Postby rACed2 » Thu Sep 12, 2013 4:13 am

You might have more detailed answers to this at the RME forum. High Gain I cannot even give a clue to you over, +24dB is something to do with a broadcasting standard, Hugh might be able to shed some light.
rACed2
Regular
Posts: 97
Joined: Sun Apr 14, 2013 11:00 pm

Re: Focal + RME gain structure

Postby Hugh Robjohns » Thu Sep 12, 2013 10:36 am

Domien wrote: I like this a lot since this takes the thinking about gain structure away concerning the speakers.


Not really... you're still thinking and asking about the appropriate gain structure here... ;) and there is a toggle switch if memory serves to switch between +4dBu and -10dBV nominal input levels on the speaker which adjusts its internal gain structure by 12dB!

But now I'm a bit puzzled by the output options of the RME. I can set the output level to 4 options: -10dBV, +4dBu, High Gain and +24dBu. I always assumed that since all my gear outputs and accepts balanced signals that everything should be set to +4dBu. Is this a correct assumption, and when would one then use the High Gain or +24dBu settings?


It's not a correct assumption -- it's a bit more complicated than that -- although setting the +4dBu mode will often work adequately when connecting to a semi-pro desk, outboard, or analogue recorders etc.

The RME output level options are exactly that: they determine the analogue output level corresponding to a digital signal of 0dBFS. (This is all detailed on page 41 of the UFX manual, by the way).

The +24dBu option is the full professional spec, delivering +24dBu for a 0dBFS digital signal. In a professional situation you would expect to be working with around a 20dB headroom margin, and the equivalent of the analogue +4dBu reference level inside the digital equipment -- ie your average working signal level -- would therefore be -20dBFS.

The other three RME options provide lower peak output levels and assume smaller headroom margins. So the High Gain mode outputs a maximum of +19dBu and has a nominal 15dB headroom margin, so you would expect your average digital level to be around -15dBFS. The +4dBu mode provides a maximum peak of +13dBu and assumes a 9dB headroom margin (ie average mix levels around -9dBFS) -- and this mode works well with semi-pro gear that just doesn't have sufficient clean headroom to work nicely at proper professional analogue levels. And finally, there is the -10dBV mode which can output a maximum of +2dBV (which is about +4dBu) and assumes 12dB headroom.

This chart shows how the relative output levels and headroom margins compare for the different modes. The +24 mode is the same as the SMPTE standard. The junction between the green and yellow sections represents the nominal reference level and where the average signal level should hover. The 9dB wide yellow band represents the normal range in which controlled peaks should reside, and the red bit above is the remaining safety headroom margin.

Image

So.... that's the techy stuff... the question now is how to apply that to your monitoring situation. From a gain staging point of view, the limiting condition is that the maximum analogue output from the ufx shouldn't cause the speaker to go into overload. Unfortunately, I couldn't see a maximum input level figure for the Focal in the handbook, so that remains unknown. I doubt it is as high as +24dBu, though...

So the best way forward is to set the Focal speakers to their +4dBu input mode, which is the least sensitive condition (the -10dBV mode makes it 12dB more sensitive -- ie, louder!), and set the RME monitor outputs to the -2dBV option which provides the lowest maximum output level. You can then play some of your typical working material, raising the RME's volume control slowly to its reference position, and assess the volume in your room.

If, with typical mix levels and the volume control at the reference position, you feel the listening volume is too low from the speakers, then lower the monitor fader on the RME, change mode to +4dBu and try again. If you want more still, then try the Hi mode... or even the +24dBu mode if necessary.

The appropriate mode setting will depend on how much headroom margin you work with when tracking and mixing, and how loud you like to monitor.

Hope that helps

H
User avatar
Hugh Robjohns
Moderator
Posts: 17515
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Focal + RME gain structure

Postby Domien » Thu Sep 12, 2013 3:09 pm

Hugh,

thanks a lot for all this information. What this mainly tells me is how little I actually know about all these technical details.

I do all my mixing in Reaper. Before starting to mix I always lower all the track volumes equally until the master output displays something between -12 and -10dB, this leaves me enough margin to start fiddling with things, which usually brings the summed level up until -4 or -3 dB.

Most of the time I don't mix at high levels, I rather would call my monitoring levels moderate. Like I said, currently I've got the speakers and the fireface set at +4dBu, which makes that when I mix I usually have to raise the volume level of the fireface until -24dB. Like this, when I feel like giving my ears a nasty treat I can raise it until -12dB (more than this and I'll end up with angry neighbours). But when listening to commercial mixes, I sometimes have to lower the main volume as low as -40dB (displayed on the fireface display).

I always assumed, apparently wrong again, that headroom was something which purely resided in the mixing stage, but now there's also headroom in the physical recreation side of things.

Ok, I'll keep your recommendations in mind and I guess I'll have to do a fair bit more experimentation and learning.

Domien
Domien
Posts: 2
Joined: Tue Aug 11, 2009 11:00 pm

Re: Focal + RME gain structure

Postby Hugh Robjohns » Thu Sep 12, 2013 3:49 pm

Domien wrote:I do all my mixing in Reaper. Before starting to mix I always lower all the track volumes equally until the master output displays something between -12 and -10dB, this leaves me enough margin to start fiddling with things, which usually brings the summed level up until -4 or -3 dB.

That's obviously working for you, but personally I'd suggest working with even more headroom and aim to have the final mix peak no higher than about -8 to -6dB, with most peaks lower than that (depending on the musical genre).

currently I've got the speakers and the fireface set at +4dBu, which makes that when I mix I usually have to raise the volume level of the fireface until -24dB

Hmmm... Given what you've said, I'd suggest that setting the UFX monitor outputs to the +2dBV setting would be the optimal arrangement for your specific situation. This will be 9dB quieter than the +4 setting, and thus you'll need to turn the UFX volume control up to -15dB for the same listening level, and -3dB when you want a blast!

Aside from putting the volume control in a more sensible part of its range, which will give a better signal-noise ratio from the converters, it will also make it safer for your speakers and your ears should something go mad in the computer and you end up with a full level howlround and no means to control it!

But when listening to commercial mixes, I sometimes have to lower the main volume as low as -40dB (displayed on the fireface display).

Yes, that's entirely normal (and very annoying!) because commercial mixes have been mastered, and part of that process involves stripping out the headroom margin. Commercial music routinely peaks full level 0dBFS (and often actually 3 or 4dB above 0dBFS, in fact... ).

With the revised output level setting you'll need to turn the level down to around -30dB for commercial music, but I would suggest a better way of working would be to attenuate the commercial music by 12dB when you import it to the DAW. That way it will have a similar headroom margin to your own mixes, and thus be easier to compare and use as reference material.

I always assumed, apparently wrong again, that headroom was something which purely resided in the mixing stage, but now there's also headroom in the physical recreation side of things.

All analogue equipment is designed to accommodate signals in such a way that they sit well above the noise floor, but well below the clipping level. So everything is designed to work with a nominal reference level (typically -10dBV for semi-pro gear, 0dBu for broadcast equipment, and +4dBu for pro studio equipment), and that reference level is normally between 12 and 20 dB below the clipping level -- the amount being referred to as the headroom margin. Most semi-pro equipment has about 12dB headroom, while broadcast works with 18dB and pro music studios, 20dB.

Employing the right reference levels is important because if you drive the levels too high you risk increasing distortion as the headroom margin runs out. A lot of supposedly 'pro' equipment actually starts to sound pretty hard and edgy above +12dBu -- only the really high end stuff is comfortable with signals routinely kicking around +20dBu or more.

H
User avatar
Hugh Robjohns
Moderator
Posts: 17515
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound



Who is online

Users browsing this forum: No registered users and 8 guests