Well, no, 110 ohms is the nearest E24 standard value...
Sure!
But why would they pick the same starting points I happened to use? I picked 32 ohms and 400 ohms because the fellow I was trying to help happened to have phones with those two impedances. Another person I helped with a guitar amp headphone attenuator some years ago had 220 ohm 'phones. It seems consumer phones come in lots of different impedances, so why would the IEC standards body pick exactly 32 and 400 ohms?
I understand the reason for Pano's comment - I wondered the same thing myself as I was typing my post about it. It is possible, certainly, but by no means certain. Not until we find some solid evidence to support the hypothesis.
Incidentally, the IEC standards number quoted in this thread is wrong - it applies to phantom power specifications, not headphone amp output impedances.
In today's era of cheap microcontrollers and digital audio, a proper headphone standard could do much better, even for the same kind of analogue headphones we're all using now. For example, the headphone amp could have a 0.1 ohm resistor between headphone return and ground; applying test tones at two or three frequencies and monitoring headphone voltage as well as voltage across the 0.1 ohm resistor, the drive amplifier could calculate headphone impedance over a range of frequencies, and set drive voltage accordingly.
If we're willing to take the extra step and have a small cheap PIC or other microcontroller in the 'phones, things become even better: the headphones could report their impedance and sensitivity to the driving amplifier, which would then configure itself to provide a reasonable maximum SPL at the listener's ears.
-Gnobuddy
Not quite - the optimum source impedance shifts a bit, to 138.5640646... ohms. Call it 140 ohms, rounded to the nearest E48 standard resistor value.
Is 600 ohms a popular headphone impedance? I've never encountered it, but I've only owned a handful of headphones over the years, and never an expensive one.
-Gnobuddy
I agree, and I've never read it, but there it is on Rod's site as IEC 61938...
My first headphones way back when were AKG K240M (600 Ohms), and I can appreciate if in a studio you're stringing together several in parallel, having a high impedance is very useful. But today I would not be looking for high impedance headphones (much over 300R) unless there's a very good reason.
Lots of HP measurements, courtesy EUVL:
DIY-Audio-Heaven | Enjoy music through headphones on a low budget
Jan
DIY-Audio-Heaven | Enjoy music through headphones on a low budget
Jan
It's a bit off topic (sorry for that) but the hd25 in 600ohms version are my favorite headphones to use with smartphones (noise cancelling Bluetooth phones set aside). It's a bit counter intuitive but they are very sensitive and so they get loud enough even from a weak source. But they still allow me to use the digital volume control at about 80% most of the time, giving me a finer control in quieter backgrounds. And since they require so little current, they don't tax the electronics much. There is also less risk of a sudden jump in volume as they operate closer to the limits of the device.
End of the OT.
LOL, that might well be the one I used to maintain for a while, over a decade ago.
Today, we've got slightly more convenient and comprehensive online resources.
Characteristics of headphones - Reference Audio Analyzer
As you can see, headphone sensitivity varies wildly, from the mid-80s to over 140 dB / V. That said, some (but not all) of the most sensitive in-ears are actually optimized for a 10 ohm driving impedance. Still, we're easily talking a delta of 40-50 dB. I think about the minimum you'll want to cover well is 95 to 125 dB / V, that still is 30 dB.
For what the state of the art is, take a look at ASR's reviews - here's the latest one:
RME ADI-2 FS Version 2 DAC and Headphone Amp Review | Audio Science Review (ASR) Forum
Generally speaking, if you can hit 9 Vrms well on the top end and less than 2 µV of noise on the bottom end (both at the same time not required - I don't see much of a point in more than about 90-100 dB of instantaneous dynamic range at any given time), you're doing very well already. Bonus points for reaching up to 16 Vrms or so, so you can get the last bit out of HE-6s and the like (5 W @ 50 ohms), or <=1 µV of noise for the grass growing detectors out there.
A good headphone amp is going to be an exercise in dynamic range management (gain staging), with multiple gain settings and perhaps some variable attenuation to boot.
Last edited:
I don't see how, it would invariably influence output impedance, which if in doubt is something you'd want to control separately. You could probably measure impedance briefly after headphones are plugged in if there's a contact in the socket to signal such (preferably trying not to make too much of a racket in the process), which would require a micro and such.
Impedance isn't a terribly good indicator of sensitivity anyway. Granted, few things 16 ohms or under aren't on the sensitive to very sensitive side, and beyond about 150-200 ohms sensitivity tends to be moderate at best, but around the 32 ohm mark it could be all over the place. Could be a BA IEM at >130 dB / V, could be orthos at <100 dB / V.
I would consider taking a page out of RME's book and including a dedicated separate 3.5 mm output for IEMs with inherently lower output level and noise. Then each output could also be equipped with toggle switches or jumpers for 100-120 ohm (1/4") or 10 ohm (3.5 mm) series resistors. I would also research what kind of balanced connectors have gained traction in the respective headphone / IEM worlds, as I know that there are some and it's literally just some more output jacks to be wired up in parallel. (I think some IEMs use a 2.5 mm TRRS, there's Sony's rather neat 4.4 mm Pentaconn, and it seems in the full-size headphone world a 4-pin XLR seems to have mostly superseded the 2x 3-pin XLR outputs found on early balanced amplifiers.)
Speaking of output jacks, make sure the output stage can tolerate intermittent shorts without issue either way, as phone jacks are likely to cause these. In case of a discrete output stage, a few ohms in series with the collectors may already prove sufficient.
In any case, the last thing the world needs is yet another nondescript fixed +16 dB box. (From consumer line level, maybe +26..30 dB tops, -20 dB min?)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.