I have tried to search for info on whether damping factor matters for headphones, and, not surprisingly, found contradictory answers. On this forum I found some claims that it does matter. Somebody said that Sennheiser design their headphones assuming a zero-Ohm output impedance. Somebody referred to somebody elses measurements, but the link was unfortunately dead. On the other hand, according to this link
www.epanorama.net/circuits/headphone_attenuator.html
there is an IEC standard that headphone outputs should have 120 Ohm output impedance (or do I misunderstand?), so then most headphone manufacturers ought to design for that, one thinks. It is also claimed that damping factor anyway shouldn't matter for correctly designed headphones. So what should one make of this?
The reason I was wondering is that a very common and practical way to implement headphone protection is to have an output series resistance and a relay that can short the output to ground via this resistor. For practical reasons one doesn't want a too small resistor. However, the epanorama link additionally raises the question whether one even should have a 120 Ohm resistor on the output?
www.epanorama.net/circuits/headphone_attenuator.html
there is an IEC standard that headphone outputs should have 120 Ohm output impedance (or do I misunderstand?), so then most headphone manufacturers ought to design for that, one thinks. It is also claimed that damping factor anyway shouldn't matter for correctly designed headphones. So what should one make of this?
The reason I was wondering is that a very common and practical way to implement headphone protection is to have an output series resistance and a relay that can short the output to ground via this resistor. For practical reasons one doesn't want a too small resistor. However, the epanorama link additionally raises the question whether one even should have a 120 Ohm resistor on the output?
Christer,
The guy inn the link sounds like he knows what he is talking about. If indeed phones are NOT mass controlled but resistance controlled, damping would be a non-issue. That would mean output driving Z is a non issue.
The 120 ohms appears to be selected to be about right for different types, but not a specific value for acooustic reasons, so you would be at liberty to vary it depending on your needs.
Maybe you could do a quick check of the IEC pub, if you have access to it?
Jan Didden
The guy inn the link sounds like he knows what he is talking about. If indeed phones are NOT mass controlled but resistance controlled, damping would be a non-issue. That would mean output driving Z is a non issue.
The 120 ohms appears to be selected to be about right for different types, but not a specific value for acooustic reasons, so you would be at liberty to vary it depending on your needs.
Maybe you could do a quick check of the IEC pub, if you have access to it?
Jan Didden
Damping factor has some influence. Check this . As you can see the frequency response gets not flat, more like a boost in the base.
I have also listened with 120 ohms out and 10 ohms out and the difference is huge may the golden ears say but I'll find it microscopical, almost zero in fact. I used 150 ohms Sennheiser HD545.
Christer if you have 100-120 ohms out you will get both short circuit protection aslo attentuation if you use low-Z cans.
I have also listened with 120 ohms out and 10 ohms out and the difference is huge may the golden ears say but I'll find it microscopical, almost zero in fact. I used 150 ohms Sennheiser HD545.
Christer if you have 100-120 ohms out you will get both short circuit protection aslo attentuation if you use low-Z cans.
But Per-Anders, that seems to be a measurement of the amplifier under different load conditions. I want to know how and if the headphones are affected by the output impedance. Designing the amplifier to handle different load is important, but a different issue.
The only thing you can see from measurements is that the voltage is increased a bit in the bass. Is the headphone affected? Don't know, I'll just listen in this case.
According to Jan Meier, the output impedance matters.
For the maths, see here : http://www.meier-audio.homepage.t-online.de/tipstricks.htm
For the maths, see here : http://www.meier-audio.homepage.t-online.de/tipstricks.htm
peranders said:
How can you see that? The headphones are only connected in one case. There are no curves for headphones and different output impedance. Besides, I am interested in a more general answer, if possible, not only what happens to one particular pair of headphones.
Jan,
Yes, the guy at epanorama certainly sounds like he knows what he is talking about. On the other hand, there are many people who sound like that without actually knowing. 🙂 However, he has the IEC standard to back it up, so the question the would be if headphone manufacturers bother about that standard?
00940 said:
Thanks, I think that is the same person that was referred to in the link that was broken. It is obvious of course that output impedance matters if the headpone impedance is more than marginally dependent on frequency, and it seems from this article that that is usually the case. Interestingly, he also refers to the 120 Ohm standard, but seems to indicate that most, but not all, headphones sound better with a lower output impedance. However, he seems not to discuss anything lower than 50 Ohms, so maybe that is safely low? I guess the sensmorale of what Meier says is that a headphone amp should ideally have a choice between different outpur impedances (series resistors, that is).
Christer,
I must say that Jan Meier doesn't really give any arguments why the phone imp would be freq deopendent. The story of break in also detracts from his credibility.
The measurements shown above certainly do NOT agree with him. They seem to indicate that only minor changes result from different damping (different Zout). There is SOME effect, but orders of magnitude less than with speakers. So, my vote goes to the first guy. 😉
Jan Didden
I must say that Jan Meier doesn't really give any arguments why the phone imp would be freq deopendent. The story of break in also detracts from his credibility.
The measurements shown above certainly do NOT agree with him. They seem to indicate that only minor changes result from different damping (different Zout). There is SOME effect, but orders of magnitude less than with speakers. So, my vote goes to the first guy. 😉
Jan Didden
Compare the soundcard itself, with resistive load and with complex load. With complex you will not get a flat response. I have also tested with 120 ohms out and the difference is minor compared to 10 ohms.Christer said:
Christer
Build a CMOY or similar and connect a trimpot in series and test different values. You will be surprised how little difference it makes. When you have done that yopu can also compare what Dr Meier says. Are small things big or are they small?
Jan,
Meier did after all have measurements for two different headphones, showing the effect of varying the output impedance, which suggests a frequency dependent headphone impedance. If by "the other measurements" you mean what Per-Anders linked to, then note that they didn't measure headphones with different output impedances, so those measurements tell nothing interesting.
As for break-in, I agree with you that for most electronic components break-in seems to file under mysteries from a performance point of view. However, I am not so sure about electromechanical devices. It seems much more plausible that break in can have an effect there.
Meier did after all have measurements for two different headphones, showing the effect of varying the output impedance, which suggests a frequency dependent headphone impedance. If by "the other measurements" you mean what Per-Anders linked to, then note that they didn't measure headphones with different output impedances, so those measurements tell nothing interesting.
As for break-in, I agree with you that for most electronic components break-in seems to file under mysteries from a performance point of view. However, I am not so sure about electromechanical devices. It seems much more plausible that break in can have an effect there.
Toni has made a pratical observation, the same as mine. 120 ohms will be good for any headphone and the sound quality isn't affected especially much if he increases the impedance to 470 ohms and my observation is the same if I'll decrease it down to 0.1 ohms or less.janneman said:
In my QRV-04 I have a feature of including feedback _after_ the series resistor. Check the J2 jumper. Even this feature affects the sound zilsch!
Christer, you could surf over to www.head-fi.org I'm pretty suree they have strong opinions about this.
Christer said:
Well, I don't want to start an off-topic rant here, but if break in has such effects (sometimes "day and night" differences as he says), it MUST be measurable in freq response or impedance changes. Never seen any meassurements to back it up though.
Jan Didden
janneman said:
Yes, or at least somehow measurable. Has anyone measured headphones or speakers before and after break-in? Of course, his claim that the difference is "night and day" should be taken with great pinch of salt.
I've had a set of headphones with a rub/buzz on high amplitude low frequency signals out of the box, I was going to return them - after 24 hrs of "burn in" playing the noise went away - presumably the supension stress-relieved and centered itself better with the workout (or maybe the high spot just wore off)
it is reasonable to expect speaker/headphones to "burn in"/change/age with mechanical exercise given the universal use of plastics with hugely nonlinear viscoelastic-creep behavior
Apple and other portable manufacturers are certainly ignoring the IEC standard - you can't put 80 mW into 16 Ohms (iPod) with 120 series R and only a single cell Li Ion battery
it is reasonable to expect speaker/headphones to "burn in"/change/age with mechanical exercise given the universal use of plastics with hugely nonlinear viscoelastic-creep behavior
Apple and other portable manufacturers are certainly ignoring the IEC standard - you can't put 80 mW into 16 Ohms (iPod) with 120 series R and only a single cell Li Ion battery
Headphones have diaphragm and enclosure resonances which are reflected in their impedance curves and thus they may be damped or not. Wheather it REALLY maters is another question.
Interesting, googled this : http://www.audiocircle.com/circles/viewtopic.php?p=25677&sid=f538e92fcedfbfa1a0cb8c70b9be0e67
the man (woman ??) called JaZZ writes interesting things
the man (woman ??) called JaZZ writes interesting things
- Status
- Not open for further replies.