More work for the wizard

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
Post Reply
Fragletrollet
Posts: 16
Joined: Sat Oct 31, 2020 9:55 am

More work for the wizard

Post by Fragletrollet »

Hey Wayne,

so this tactic worked before (the wet/dry pcb, that I still haven't gotten around to ordering :? ) so I'm just gonna shamelessly throw it out there:

I think you could design a very nice headphone monitoring system for use with high end cans. You already have the Dual class A, smack a couple of those together for balanced outputs, implement an EQ so one can alter the response, add crossfeed functionality, a nice switch for volume control, maybe some stereo/midside processing functionality (stereo width adj?) etc...

I like the headphone amp I use with my HD800's (the one in my Dangerous Music Monitor ST), but they are awfully powerful with my 800's, resulting in having to use the lower range of the volume potmeter, where L/R tracking is pretty terrible. The adjustment range becomes very small, easy to hurt your ears and easy to use a range that skews the panorama.

And I would like one for the bedroom. But I need that eq adjustment (based on response files of the 800 that I like, that I now do digitally with a generic parametric eq), the 800's are just so thin and treble'ey without it. And since I got a nice set of expensive balanced cable throwed in along with the cans when I got them (that the seller seemed intoxicated about), why not go balanced. Even tough from my knowledge, I don't really understand why this is a thing with headphones and their short cable runs in generally interference free environments... but double is always better, right? :lol: Or just double outputs, so I can impose my music on my spouse.

some inspiration for units I have in mind :
https://spl.audio/en/spl-produkt/phonitor-3/

https://rme-audio.de/adi-2-4-pro-se.html (well sans the digital part I guess. Or, if you by any chance are an amazing digital designer as well, do go ahead :lol: wouldn't mind having building blocks to a high end converter either, but I presume that's a tougher nut to crack)

In my head such a processor fits right into your product line, that's generally directed towards mastering needs. Maybe just a processing board to be added to the ClassA project?

Here's hoping....

-magnus
User avatar
mediatechnology
Posts: 5497
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: More work for the wizard

Post by mediatechnology »

Thanks for the input Magnus!

I think there may be two reasons for balanced headphone wiring.
The first is to provide extra voltage swing for hard-to-drive-phones.
The second is to avoid the shared I*R drop in the TRS sleeve connection.

The I*R drop in the common sleeve connection reduces the Mid component slightly thereby increasing width.
Increased width would be more obvious with lower impedance 'phones.

The more Mono the source the greater the sleeve current is because Mono is a common mode signal.
L-R difference "Side" components are differential with current flowing from tip to ring.
Insert an impedance in the common conductor and it will sound wider.

I found the EEQ Elliptic Equalizer provides a nice low end cross feed for 'phones that for some mixes cleans up muddiness and provides more "point" to the low end.
You can hear it somewhat on speakers but on 'phones its a very obvious improvement.
When warped vinyl is played EEQ helps a lot with headphone image stabilization along with reducing Doppler shift particularly on speakers.
Warp is mostly vertical L-R and, when presented to the ears using 'phones, I find nauseating despite being infrasonic.
It's a pressure change effect.

I think you should integrate a Monitor Controller, Stereo Width Controller, EEQ and DCAO into a single unit and ditch using fully-balanced drive.
Or just go with the superb RME interface with some DSP and be done with it.
(My only reservation about the somewhat expensive RME gizmos are how long their OLED displays will last and be supported.)

Some of the DCAOs sold have been made into performer cue systems but most are used for general-purpose monitoring.
I can't recall anyone using it balanced except for line or high level transformer driving.
I know balanced headphone drive is a thing but I'm not sure how appealing it would be to no-nonsense commercial users with easy-to-drive 'phones.
It would be something I'd have to think about but the pieces are there for anyone who wanted to do it now.
User avatar
JR.
Posts: 3712
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: More work for the wizard

Post by JR. »

Back in the 80s I designed a headphone amp while at Peavey (the HB-1) that drove the sleeve with opposite polarity to generate 2x the voltage output from +/-15v rails.

I designed it after hearing how loud a drummers headphone mix was during a session. 8-)

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5497
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: More work for the wizard

Post by mediatechnology »

JR. wrote: Sat May 25, 2024 10:15 am Back in the 80s I designed a headphone amp while at Peavey (the HB-1) that drove the sleeve with opposite polarity to generate 2x the voltage output from +/-15v rails.

I designed it after hearing how loud a drummers headphone mix was during a session. 8-)

JR
So if I were to guess the sleeve was driven with L+R aka Mono?

I was going to point out to Magnus that 4 wire connections are beneficial even if the return conductor is undriven and just grounded.
On the DCAO-II the ground returns between left and right Phoenix connectors are split and telescope directly to the DC 0V power supply connection.

3.5 mm TRS connectors in my experience can develop high contact resistance in the sleeve over time when left mated.
I often have to occasionally twist the male in the female to break the oxide on sound card outputs or they'll develop hum from small leakage currents.

I also have to do the same thing on the Fluke leads' banana connections to measure low Ohms.
They'll often drop from 10-12 Ohms to sub-Ohm with a couple of twists at the meter and probe ends.
User avatar
JR.
Posts: 3712
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: More work for the wizard

Post by JR. »

mediatechnology wrote: Sat May 25, 2024 10:33 am
JR. wrote: Sat May 25, 2024 10:15 am Back in the 80s I designed a headphone amp while at Peavey (the HB-1) that drove the sleeve with opposite polarity to generate 2x the voltage output from +/-15v rails.

I designed it after hearing how loud a drummers headphone mix was during a session. 8-)

JR
So if I were to guess the sleeve was driven with L+R aka Mono?
Yes something like that... with the sleeve driven L+R the left output was driven with L-R so they combined to cancel out the R. Alternately the right output was driven with R-L so the left cancelled out.

This was not an audiophile product but a loud headphone monitor system that could melt the wax in muso's ears. The circuit output stage was conservatively designed so it could actually drive loudspeakers. Playing speakers too loud would melt the 1A wall wart's internal thermal fuse, so I added a resistor in series with the half wave PS rectifier so high current draw would cause the PS rail to sag enough to protect the wall wart. These were not designed to actually drive loudspeakers, but I routinely used them that way to drive loudspeakers for trade show displays.

JR
I was going to point out to Magnus that 4 wire connections are beneficial even if the return conductor is undriven and just grounded.
On the DCAO-II the ground returns between left and right Phoenix connectors are split and telescope directly to the DC 0V power supply connection.

3.5 mm TRS connectors in my experience can develop high contact resistance in the sleeve over time when left mated.
I often have to occasionally twist the male in the female to break the oxide on sound card outputs or they'll develop hum from small leakage currents.

I also have to do the same thing on the Fluke leads' banana connections to measure low Ohms.
They'll often drop from 10-12 Ohms to sub-Ohm with a couple of twists at the meter and probe ends.
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5497
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: More work for the wizard

Post by mediatechnology »

JR. wrote: Sat May 25, 2024 11:57 am Yes something like that... with the sleeve driven L+R the left output was driven with L-R so they combined to cancel out the R. Alternately the right output was driven with R-L so the left cancelled out.
JR
That sounds right and you end up with 2L and 2R which doubles your voltage swing and squares the available power at a given load impedance.
I played around with a similar thing maybe 35 years later.
Post Reply