[GiacomoHoldini]

I just got a pair of Sennheiser HD 598s – rated at 50 ohms, with a near-300 ohm spike around 100 Hz. I’m listening to them through two different receivers I have in my home: one is a Denon stereo receiver, the other is a Yamaha A/V 5.1 receiver. I am using the same model CD player with both, connected via analog output, so the differences I’m noticing should be related to the receivers only.

On the Denon, the upper bass/lower midrange sounds a bit bloated when played without EQ. Lowering the bass tone control by one click (2 dB) on the Denon helps, but generally the ‘phones sound better on the Yamaha – when tone controls are bypassed, bass is good and present, but not bloated; no EQ is necessary. Everything else sounds great on both setups. I can’t tell if the sound variance is due to the natural coloration of the circuitry, or if it’s a power issue. The Yamaha has a headphone output of 150 mV/100 ohms; the manual for the Denon provides no information regarding headphone output. Given that the bass “bloat” on the Denon is happening right around the frequencies associated with the impedance spike in the drivers, I’m wondering if the Denon is powering those frequencies more highly than the Yamaha. Similarly, according to Meier-Audio, higher output impedance, relative to driver impedance, results in lower attenuation of bass frequencies, which has the potential to result in a darker, muddier sound.

With that in mind, I am considering hauling my older Yamaha receiver out of storage and using it in place of the Denon with the hope of getting the same sound I have with my current Yamaha. However, according to the manual, the older Yamaha has an output of 0.34 V/560 ohms. I understand it’s using about double the power at about 5.5 times the impedance of my other Yamaha, but I don’t know what this means in any practical terms. How is this difference likely to affect the sound, assuming output is the only variable? Which output is a more appropriate match for the 598?

Thanks in advance.

 

[NA Blur]

The most likely thing is the colorization of the Denon amp circuitry.  There is probably a natural bass enhancement built into the amp end.  Can you hear this with a loud speaker setup using your Denon?
 
Controlling bass is hard to accomplish and having poor driver damping caused by impedance mismatching is a direct cause.  The classic 1/10 rule can be applied to see if the headphone output impedance is adequate on just about any amp.  Measure the impedance of the stereo jack in Ohms to get just the resistance load of the output.  Say on the Denon it is 100 Ohms.  If you drive a pair of 50 Ohm headphones with a 100 Ohm output impedance the damping factor will be just lousy ( 1/2 ).  This results in distorted bass which can sound boomy at times.
 
Most well-built headphone amps will have an output impedance for the headphones of less than 1 Ohm.  This will give you a damping factor of at least 25 or more for most headphones.  Good damping almost always equals good control of the speaker driver.
 
See "Output Impedance" here:
http://en.wikipedia.org/wiki/Headphone_amplifier

 

[OJNeg]

I'm familiar with the dampening factor calculations, but how exactly does amplification relate to a headphone's impedance curve?
 
I've noticed most headphones have very flat impedance curves. Meaning that if the headphone is rated for a 40ohm impedance, Z=40 for pretty much the whole frequency range. Other headphones seem to have wacky impedance curves, specifically Sennheiser's HD series. While something like the HD600 is nominally rated at 300ohm, at 100Hz, Z=500ohm+. People commonly recommend that to get the best out of your Sennheiser 'phones you need amplification. Is there a correlation between this sort of curve and the need for amplification? Is it solely based on the amplifier's ability to supply voltage for the sections of the frequency bandwidth that have a higher impedance? Or does it have to do with creating a high dampening factor that tames the unruly low-end response?

 

[GiacomoHoldini]

Thanks for the response. I was just about to post an update, since I dragged out my older Yamaha receiver (5540) and did some comparison listening. I noticed two things: one, the older Yamaha sounds identical to the Denon when tone controls are set flat, except that the Denon isn't as loud. The 5540 has a lot more volume. Two, what I first described as bass "bloat" is now sounding more to me like proper presence, depending on the recording. Some recordings sound fantastic when played flat on either the Denon or the 5540, and some sound like they just have too much upper bass/lower midrange happening. So, that could be an issue more with recording choice, rather than the equipment.

However, when I compared the 5540 to the newer Yamaha (5940), the difference became clear. The 5940, while it may sound cleaner when playing some recordings, is actually lacking significant bass impact through the 598s on tracks that demand it. When I played more sub-bass heavy tracks, the punch of the more powerful 5540 was really evident. (It's only more powerful in terms of its headphone output. In terms of loudspeaker output, the 5540 has substantially less power than the 5940.) Even when I increased bass EQ on the 5940 to roughly match the sound on the 5540, the impact of the bass was weaker.

The rule of ten helps me nail it down. The 5540 has an impedance eleven times that of the 598s, whereas the 5940 has an impedance of only twice the 598s. So it would seem the lighter bass sound of the 5940 is due to the fact it's not powering those frequencies as effectively as the 5540; as such, the sound coming through the 5540 would seem to be a better reference.

Regarding the bass enhancement, I don't know if this makes a difference, but the Denon has a "direct" mode that allows bypass of all tone control ciruitry, so the amp is processing a (theoretically) pure signal. Then again, if the bass enhancement is built into the amp itself, that might not matter. I haven't noticed any abnormal bass on loudspeakers, but like I said, I'm revising my position on the "bloat" factor anyway, so I don't know how much that matters.

 

[GiacomoHoldini]

While something like the HD600 is nominally rated at 300ohm, at 100Hz, Z=500ohm+. People commonly recommend that to get the best out of your Sennheiser 'phones you need amplification. Is there a correlation between this sort of curve and the need for amplification?

I noticed that when I plugged the 598s directly into the headphone jack on the CD player - which has amplification about on a par with an iPod - the 598s were plenty loud, but very bass-light, particularly in the range where the impedance spike in the drivers reaches close to 300 ohms. If you didn't have the benefit of comparison, you might think the 598s don't have high amplification needs, but are just bass-shy headphones. They're not. They don't have the fullest sub-bass, but bass above 40-50 Hz is very present and well represented, as long as the 'phones are amped well. The light amplification of the CD player headphone out clearly wasn't enough to show the 598s to their full potential. This suggests, yes, there is a correlation between the impedance curve and the need for amplification - at least if my rather unscientific example is anything to go by.

 

[GiacomoHoldini]

I just read the Wikipedia entry NA Blur linked to, and now I'm totally confused. How does one reconcile the 1/10 rule with the following statement from the article?

"Many headphone amplifiers have an output impedance in the range of 0.5 - 50 Ohms. The 1996 IEC 61938 standard recommended an output impedance of 120 Ohms, but in practice this is rarely used and not recommended with modern headphones. In 2008 Stereophile published an article that showed that a 120-Ohm output impedance could cause a 5-dB error in frequency response with certain headphones. [2] The article concludes that the effect of output impedance on frequency response is "non-trivial". Some newer headphone amplifiers have output impedances that are less than one Ohm in order to ensure accurate frequency response."

Whether we go by the IEC standard of 120 ohms, or the contention that output impedance should be a lot lower than that, following the 1/10 rule would seem to lead to some out of whack matching. In the end, I'm just going by my ears.

 

[OJNeg]

As I understand it, the older generation of headphones, specifically studio headphones, were very high impedance. 600+ IIRC. This means you would get a respectable dampening factor if listening to such a headphone with a higher output impedance. Obviously there are those who believe that you must have a 1:10 factor and the closer your get your output impedance to 0 the better. But I've heard 1:8 and 1:4 called the ideal ratio, and there are plenty of OTL designs which have 100+ output impedance and are listenable. The driver must move farther out to reproduce low frequencies, so a poor dampening factor will manifest itself as what seems to be loose bass.

 

[obobskivich]

See proton007's impedance thread; mostly to save me re-typing 3-4 pages of information.

Anyways, the quick and dirty is, what you're hearing is potentially:

1. In your head.
2. The higher Zout on one of the receivers.
3. A combination of both.

DF is largely a concoction of marketing - don't worry about it too much. It does not correlate to "loose" or "tight" bass - that's a function of mechanical damping and the driver's motor design (so if it's crap from the factory, it's always crap, even if you dump a million dollars into gear). Oh and you can't measure Zout by tapping the jack; you have to measure it with a simulated load and a v-meter.

The output on receivers is generally competent for what you're doing (I should say, older receivers that drive the output with their power amp; newer designs it's a lot harder to get into, because some of them use the power amp, some of them just have opamps or similar arrangements (and some of those are good, and some are not so good)). And the IEC standard is ignored because portable players can't hack it. The whole argument over "error" is fairly contentious too - high Zout simply results in non-linear attenuation of the FR relative to the impedance curve. Sometimes it's a good thing, sometimes it's a bad thing. E.g. imho the HD 580 sound best from receivers/amps with high Zout (400-500 ohms), it warms them up. There is no simple/easy "consumer reports" answer here.