A Vented Box (AKA Reflex) actually has a fourth order response (24dB/octave), which you can design for any of the usual high pass filter functions (Butterworth, equal ripple Chebyshev, etc).
And yes, at sufficiently low frequency remote from the port and driver resonances the driver motion approximates that of an open baffle, so you absolutely need subsonic filtering, either active or passive.
As for whether balanced is necessary or not in a domestic environment, it all depends. I use Linkwitz LX521,4, a five way active speaker. So an analog crossover with 10 outputs (5L and 5R) and ten power amp channels in three chassis. Try plugging that lot together single ended and try chasing the hum. With balanced IO it really is a simple task to plug it together and it works with zero hum - true plug and play.
If you have a simple system, with a preamp and power amp, I might agree that balanced is not necessary, and if simple receiver circuits are used (an instrumentation-type circuit) it will be significantly noisier that a single ended input. You have to try quite hard to design a quiet balanced receiver.
You can of course use transformers, if you have deep enough pockets.
And yes, at sufficiently low frequency remote from the port and driver resonances the driver motion approximates that of an open baffle, so you absolutely need subsonic filtering, either active or passive.
As for whether balanced is necessary or not in a domestic environment, it all depends. I use Linkwitz LX521,4, a five way active speaker. So an analog crossover with 10 outputs (5L and 5R) and ten power amp channels in three chassis. Try plugging that lot together single ended and try chasing the hum. With balanced IO it really is a simple task to plug it together and it works with zero hum - true plug and play.
If you have a simple system, with a preamp and power amp, I might agree that balanced is not necessary, and if simple receiver circuits are used (an instrumentation-type circuit) it will be significantly noisier that a single ended input. You have to try quite hard to design a quiet balanced receiver.
You can of course use transformers, if you have deep enough pockets.
Pretty trivial if the design is correct. Note, I have been known to use XLR connectors all across the board because it allows me to separate Earth/Chassis and "Audio Ground" but running single-ended signals across this.
The problem is not the single ended connection per se.
Nope, balanced does not guarantee "no hum" either, you can get the classic "pin 1" problem for balanced.
Well, my "system" nowadays is connected by optical to the source and by speaker cables to my 2-Way Series crossover speakers that use a 10" Bass/Mid and a 1.75" HF unit, subs are active from the speaker feed.
DAC, Preamp, Poweramp etc. are a single box (Marantz PM-75).
So I have to not deal with ANY cables.
I don't. Why do you answer how difficult a task it is to me to design a quiet line input receiver?
If our driver handles 600 Ohm fine, our input can have 600 Ohm input impedance, using a standard line receiver design with 600 Ohm input impedance and -6dB gain allows ~130dBu Ein at a signal level of +20dBu output, yup, 150dB dynamic range balanced input receiver.
Our active circuit needs ~ 1nV|/Hz self noise to not degrade that.
Now I probably call 150dB dynamic range in the analogue domain a little excessive, so if we set the input impedance to ~ 6kOhm and use a signal circuit with < 3nV|/Hz we will still have > 130dB real world DNR with circuitry ANYONE can design.
For home use they are unnecessary. If you do need very long cables and cannot guarantee the earth potential between the two ends, I do not know any electronic balanced circuitry that does not cause significant SNR problems completely absent with transformers.
Mind you, in this day and age stage boxes and mixers are digital, so that old 1980's massive 50-100m multicore "snake" from stage box to Mixer no longer exists in the actual format we had back then.
Thor
If you use a first order crossfeed circuit at 80Hz (ask me one day why 80Hz is fine) will have 20dB attenuation at 4Hz and 40dB at 0.5Hz. I usually also have a traditional highpass set at ~ 10Hz for another -20dB at 1Hz.
The main component of record warp is 0.55Hz (33.3 RPM) to 0.75Hz (45RPM). With the crossfeed and highpass first order we get around 60dB down for rerod warp fundamentals.
Are we really concerned with signals around 60dB down?
I have not found them problematic with a first order crossfeed and open baffles or vented boxes. It's very simple and effective.
Thor
Now that is interesting. Does your first order crossfeed have a flat response, or does it peak before it cuts off?
Craig
Well, I am just calling BS where I see it... Like low noise balanced receivers being hard etc. I usually find that what is needed is to correctly define the problem and then to solve it. Now to walk around with a solution and to then look around for a problems it may fit and failing to find one, bodge it in anyway.
No peaking. It is literally two capacitors and three resistors. It is THAT simple.
Again, we come to the point where we need to correctly define the problem and the solve is in a simple, efficient and expedious way and not first create a theoretically but complex Rube Goldbergian contraption and to then try to apply it to the problem.
Thor
I'd use the circuit used in the Palette Preamp. That uses a National Semiconductor design. Section 2.17, pages 2.53 tp 2.55 National Semiconductor Audio Handbook 1977. All of the design equations are there.
Search the web and you can find a downloadable copy.
There are other ways of doing it. Using inductors, or a multiple feedback bandpass filter (which is what the Cello Audio Palette uses)
Search the web and you can find a downloadable copy.
There are other ways of doing it. Using inductors, or a multiple feedback bandpass filter (which is what the Cello Audio Palette uses)
The other way to do this is using a state variable filter. That allows independent adjustment of frequency and Q. With other filter topologies, frequency and Q are dependent on each other - in other words if you decide you want a different Q but want the center frequency to be the same, you have to redesign the filter.
And this is required reading if you are trying to design filters https://www.ranecommercial.com/legacy/pdf/constanq.pdf
And this is required reading if you are trying to design filters https://www.ranecommercial.com/legacy/pdf/constanq.pdf
XLR is the One True Connector and thou shalt solder no wires to blasphemous lesser attachments.
My home-built gear uses only XLRs though I've not gotten round to installing them on my CD and Blu-ray players yet. Soon...
Now why isn't it a surprise you worked in pro audio? From back when Dilbert was funny:
https://static.tvtropes.org/pmwiki/pub/images/dilbert_unix_5611.png
Although it should be "Here's a nickel kid, get yourself a better connector."
Unless you were a Mark Levinson product back in the day when they used three pin Lemo connectors. Then had to supply adaptors, or cables with Lemo on one end and XLR on the other.
Back in '76 I worked for several months at CERN, and built myself a preamp using discrete opamps. Now CERN used (and probably still use) Lemo connectors. So my now rather ancient preamp (in the attic now) has Lemo connectors.
Why discrete opamps? Well in the mid 70's it was the era of useless opamps - a choice of several awful products, which were expensive. In '72 I built a pre-power combination that used 741 and 748 opamps. And they were 60p each at the time - todays equivalent of UKP7 each.
Back in '76 I worked for several months at CERN, and built myself a preamp using discrete opamps. Now CERN used (and probably still use) Lemo connectors. So my now rather ancient preamp (in the attic now) has Lemo connectors.
Why discrete opamps? Well in the mid 70's it was the era of useless opamps - a choice of several awful products, which were expensive. In '72 I built a pre-power combination that used 741 and 748 opamps. And they were 60p each at the time - todays equivalent of UKP7 each.