Hi all,

I own a pair of BM5A's monitor's which I've had for a long time, and I like them a lot. They have three input sensitivity settings: +4 (for low output sources), 0 (default), and -10 (drops the input by 10 db for high input sources). With the sensitivity dip switch set to 0, the monitor can generate 85 db SPL @ 1m with a -16 dbu rms input signal. My audio interface is a fireface 400 which can generate +13 dbu @ 0dbfs in the outputs. For normal listening, I usually have the master volume level on the software Totalmix set around -38 to -30 db for mastered material (peaks close to 0dbfs). Otherwise, the monitors get very very loud if I set the level fader higher due to the high sensitivity of the speakers.

I was using the monitor's sensitivity setting at it's -10 setting, which would cause a -6 dbu signal to output 85 db spl @ 1m. This lets me set the totalmix fader setting at a higher level. However, I notice a slight change in the sound quality. Like for example, I was listening to the EBU SQAM CD's Male Speech track and the voice sounds more smoother as far as dynamics when I keep the speaker's sensitivity at it's default 0 db setting. When I attenuate the input signal using the input sensitivity set to -10, then I can perceive that the dynamics are not as smooth and sound a little bit more constricted. This leads me to believe that the input attenuation of the speaker is degrading the sound a little bit. Has anyone experience this before?

- Daniel

--------------------
# Daniel #

The input attenuator on an active speaker shouldn't affect the quality in any way at all, other than defining the listening level and, potentially, the headroom margins.

In general, you will get the best results if you set the speakers to accept the loudest input (+4dBu), and run the total mix output as high as possible. That will maximise the signal to noise ratio and use the greatest conversion word length. Running the total mix at -36dBFS means you're converting at 6 fewer bits than intended, and if the processing isn't dithering the output correctly you may well hear some quantisation granularity.

I prefer to run the d-a as hot as practicable, and use an analogue attenuator (such as the SM Audio nano) to set the listening level to the speakers. That approach also pays dividends if the computer should go mad and generate full level noises while locking you out of the software level control!

Hugh


--------------------
Technical Editor, Sound On Sound

As a follow on to this I have a question about the attenuators on active speakers:

At the moment I run my DA at unity which then passes into my monitor controller and then into my speakers. At the moment I use the speaker attenuators to their maximum (most attenuation) to allow me to have my monitor controller in the 60% position for normal use. Is this good practice or should I be attenuating the speakers at source as little as possible?

Cheers

BA

No, you are working in exactly the right way with a sensible gain structure through the converter and monitor controller.

Hugh

--------------------
Technical Editor, Sound On Sound

Hi Hugh,

Thank you very much for your input. I did a little more experimenting and followed your suggestions.
On my previous experiments, I think I wasn't matching the levels exactly enough during the -10 / +4 sensitivity experiments (Even though I was using an spl meter). The psychocoustic effect of the louder signal made me believe that the speaker's attenuator was messing with the sound quality. With classical music, and the lowest sensitivity on the speaker, my digital level fader has to be around -10 or -5 and it sounds quite lovely

Thanks,

Daniel