Balanced Attenuator for Stereo Monitors + Subwoofer

[Brian Roth]

Low impedance (I think 600Ω) main monitor outs from a Neve 51 series

OK.....in the OP's proposed setup, a passive attenuator would be between the Antelope and the monitors. I know little/nothing about the Antelope's output impedance (or drive ability) so I have to ASSume it would need something like a 5K-ish load to avoid "choking" the drive source. I'm ASSuming a wimpy output circuit in the Antelope....maybe a TL07x?

If the Antelope uses something with some "heft" like a 553x, then the attenuator impedance could be reduced to...hmmm.....1000 Ohms and not piss off <g> the Antelope's outputs. Then I believe the crosstalk effect from the summer (after the fader) could be academic.

Bri

 

[Brian Roth]

Further thought after I hit SEND.....bridging the "wiper" of each "pot" into 10K summers....6 dB loss. Input Z of the sub is specified as 10K. Too lazy to calculate the total loss.

But, add a loss pad into each LR speaker.

Bri

 

[RoadrunnerOZ]

If the intention is to use the sub and main speakers unfiltered then there may be no need for main speaker attenuation as the range of the sub is not far below the main monitors, indeed the sub may still need attenuation. It’s one of those try it and see scenarios.

 

[RoadrunnerOZ]

The output impedance of the Antelope is from what I can gather around 56Ω - it’s not specified in the user manual.

 

[Brian Roth]

If the intention is to use the sub and main speakers unfiltered then there may be no need for main speaker attenuation as the range of the sub is not far below the main monitors, indeed the sub may still need attenuation. It’s one of those try it and see scenarios.

Indeed! We are clueless about the sensitivity of both the sub and mains (and room acoustics). In THIS case, there is a adjustment pot for the sub.....assuming the summer loss doesn't cause too much loss driving the sub, which could require a pad into the mains.

We're just guessing for the OP's setup.

I'm trying to concoct some sort of "rules of thumb design scheme" which could be applicable for anyone else randomly reading this thread with "who knows what" speakers in the future.

Yes....like nailing Jello to the wall....lol!

Bri

 

[Brian Roth]

The output impedance of the Antelope is from what I can gather around 56Ω - it’s not specified in the user manual.

And we don't know how low of an impedance those outputs can drive with minimal distortion. That's why I had a 5K value in my mind. Most everything these days can drive that without "farting" <g>.

Bri

 

[RoadrunnerOZ]

Bottom line fix-all approach would be trimmers on the L/R/Sub feeds post attenuator for calibration

 

[Freddy G]

Sorry for the slow reply. Let me go back to an earlier question. Can you "afford" using a third analog output from the Antelope to create a sum (via your DAW) of the monitor L/R signal? That would add at least two more decks on the attenuator switch to handle the adjustment to the sub.

Speaking of which...what topology are you considering for the balanced stepped attenuator? The simplest would be a potentiometer (voltage divider) arrangement, but there are ladders, H, etc.

Bri

I would rather avoid using another Antelope output. I'd like to at least explore the other options. I was thinking about just going with a series arrangement for the attenuator resistors.

 

[RoadrunnerOZ]

If using series 12K resistors you could just have a 10K precision stereo mini pot as a trimpot wired hot-top/ bottom/bottom/top-cold and output fed from the two wipers. Same for the two main speaker feeds. The pots would need to be pretty accurate so as not to get a different level from the hot and cold respectively. You used to be able to get linkable trimpots and the linkage - these guys have only a handful left - 3 pots and four linkages (but to ensure accuracy I’d be using a miniature stereo pot):
https://www.time-tunnel.com.au/elec...cision-trimpot-log-10k-ohm-pcb-new-old-stock/
And the linkages:
https://www.time-tunnel.com.au/elec...n-trimpot-multi-ganged-linkage-new-old-stock/
Edit: not enough 10K pots to do this job

 

[Neperdun]

to be it really balanced You'll have to use precession class resistors

 

[RoadrunnerOZ]

I usually match values using a meter and selecting from a whole bunch of 1% - means buying more than needed for a job but then resistors are cheap.