• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of Denon AVR-4306

Joined
Feb 8, 2019
Messages
22
Likes
20
Location
Australia
Just to finish off on the matter of headphone output resistance, one last test and some brief listening impressions. I tried out the Audiofly AF140 in-ear monitors. (RRP $AU399) These use three drivers: dynamic for bass, balanced armature for mids and balanced armature for treble. Nominal impedance is 38 ohms, but with that driver arrangement, you'd have to expect some variation across the range. If driven with 470 ohms of in-line resistance, they are presented with a pre-screwed frequency balance:

Audiofly AF140.png


They also produce sound at a much lower level since more than 90% of the power is being dissipated over the 470 ohm resistor.

Right now I'm listening to some music using these IEMs using the Pro-Ject Pre Box S2 at a nicely loud level. The volume on the Pro-Ject is set at -25dB. When I put the 470ohm load in line (I've measured the Pro-Ject's output resistance at somewhere between 0.24 and 0.37 ohms, so it is insignificant), the volume drops enormously. To get something like the same subjective level I have to wind up the volume control to 0dB. The bass is very prominent with the line resistance in place, the treble very recessed. Anyone using these earphones with a Denon or NAD or Marantz home theatre receiver would probably find them quite unsatisfying.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,587
Likes
38,287
Location
Gold Coast, Queensland, Australia
These use three drivers: dynamic for bass, balanced armature for mids and balanced armature for treble. Nominal impedance is 38 ohms, but with that driver arrangement, you'd have to expect some variation across the range.

Based (no pun intended) on the plot above, X/over about 2.7K and 5.5K for the mid/treble if those kinks are on the money, with the main driver likely running full range. According to the company, they are:
"For artists who are all about that bass. The triple hybrid-driver AF140 features one dynamic bass driver and two balanced armature drivers to deliver a rich, full-bodied bass with well-defined mids and highs."

So, we are not talking an accurate headphone are we?

Consider that high quality dynamic headphones have been made for many decades and for all those decades until just recently, all headphone drive circuitry was conventional and used protective dropper resistors. The most expensive digital sources, reference CD players and preamplifiers all employed >100ohm headphone sources.

So the question becomes, are headphones really designed en-masse for zero output impedance drive circuits? I personally don't believe so. Were they tuned in the first place for a typical source impedance of several hundred ohms to give their target response curve or were they designed to only give their 'perfect' response on a zero z-out impedance instrumentation amplifiers that no consumers or audiophiles possess?

What really matters is the acoustic output of the drivers when fed with various source impedances across the frequency range, and, as most of the variations are occurring right on the resonant peak of the driver in the bass region where electrical input and acoustic output are not necessarily correlated, more investigation is in order.

As an aside, I remember collecting lots of cool 1970s and 1980s headphones for nothing or close to it, that I used to see at garage sales and op-shops. That was well over 25 years ago, back when any type of over-the-ear cans were seen as Grandpa and old-skool. Everyone had silly buds jammed in their ears. I amassed quite a collection of mint condition vintage cans, many of which were absolutely terrible to listen to, but looked fantastic. Some however were quite good and made me realize, there were some jewels in the 1970s in the headphone market.

At the time I had a new pair of Sony MDR-777 and AKG-600s which I still have in my headphone drawer, but rarely use. The AKGs of 25+ years of age, are virtually indistinguishable from the AKG-601s and the AKG-702s are similar again. None care in the least about source impedance. I hear rubbish about them being 'hard to drive'. As long as one has an amplifier with a decent voltage swing (which any decent HiFi amplifier should have), they sound great.

Anyway, I could plot impedance curves on headphones today or go to the beach with my gal. I think I know which will be more fun. Cheers.
 
Joined
Feb 8, 2019
Messages
22
Likes
20
Location
Australia
Based (no pun intended) on the plot above, X/over about 2.7K and 5.5K for the mid/treble if those kinks are on the money, with the main driver likely running full range. According to the company, they are:
"For artists who are all about that bass. The triple hybrid-driver AF140 features one dynamic bass driver and two balanced armature drivers to deliver a rich, full-bodied bass with well-defined mids and highs."

So, we are not talking an accurate headphone are we?

Consider that high quality dynamic headphones have been made for many decades and for all those decades until just recently, all headphone drive circuitry was conventional and used protective dropper resistors. The most expensive digital sources, reference CD players and preamplifiers all employed >100ohm headphone sources.

So the question becomes, are headphones really designed en-masse for zero output impedance drive circuits? I personally don't believe so. Were they tuned in the first place for a typical source impedance of several hundred ohms to give their target response curve or were they designed to only give their 'perfect' response on a zero z-out impedance instrumentation amplifiers that no consumers or audiophiles possess?

What really matters is the acoustic output of the drivers when fed with various source impedances across the frequency range, and, as most of the variations are occurring right on the resonant peak of the driver in the bass region where electrical input and acoustic output are not necessarily correlated, more investigation is in order.

As an aside, I remember collecting lots of cool 1970s and 1980s headphones for nothing or close to it, that I used to see at garage sales and op-shops. That was well over 25 years ago, back when any type of over-the-ear cans were seen as Grandpa and old-skool. Everyone had silly buds jammed in their ears. I amassed quite a collection of mint condition vintage cans, many of which were absolutely terrible to listen to, but looked fantastic. Some however were quite good and made me realize, there were some jewels in the 1970s in the headphone market.

At the time I had a new pair of Sony MDR-777 and AKG-600s which I still have in my headphone drawer, but rarely use. The AKGs of 25+ years of age, are virtually indistinguishable from the AKG-601s and the AKG-702s are similar again. None care in the least about source impedance. I hear rubbish about them being 'hard to drive'. As long as one has an amplifier with a decent voltage swing (which any decent HiFi amplifier should have), they sound great.

Anyway, I could plot impedance curves on headphones today or go to the beach with my gal. I think I know which will be more fun. Cheers.
These are specifically designed as live performance in ear monitors. They tend to be a bit bright for my taste, but I can see that as an advantage in that situation.

The problem, as I see it, is that there is no widely used standard. If the norm was to use, say, 100 ohms output resistance, then headphone/earphone designers could design to that. But there isn't. So if you are buying headphones, you don't know the internal resistance of your headphone amp, you don't know whether the impedance of headphones varies over the audio band or is the same, so you don't know what the result will be.

I agree that what really matters is the acoustic output. That will be largely determined by the performance characteristics of the drivers (read large: including enclosures and other treatments that have an effect on sound) PLUS the frequency balance of the signal that is fed into them. We agree that with some headphone models, their impedance characteristics cause the balance of that signal to be modified.

So, given the lack of a standard, perhaps a low impedance on would be a good one to implement. Especially as various headphone dongles supplied with some phones (I've tested the one for the Google Pixel phone as well) manage this at low cost.

I couldn't agree with you more about 'hard to drive'. As with so much in hifi, there's a huge amount of bull with regard to headphones. 'Hard to drive' is a concept imported from loudspeakers, where some very exotic models could impose very strange loads indeed. But in my experience, 'hard to drive' means 'of course our very special device is going to sound lousy with your mediocre gear ... you should buy our other special stuff to make it sound right'.
 

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
4,643
Likes
2,812
  • Power output. There is not a single multi-channel, single box AVR, regardless of price or pedigree that can deliver the sheer amount of power that dedicated 2 channel (or monoblock) amplifiers can deliver. Especially power into very difficult loads for extended periods.
  • Dissipation. Put 5,7,9 or 11 channels of amplification in a box, attached to inadequate heat-sinks and things get very hot, very fast. Long term heat damage in AVRs in their Achilles' heel. I see many <10yo AVRs with heavily heat damaged components, traces and PCBs. Consider, an AVR is only likely to run flat-out in the most extreme situation for one feature movie- maybe 2 hours tops.

Hi John

Regarding the parts I quoted here.

Do you think @amirm 's 4 ohm test here is done for sufficiently long enough time period?

For example, if one were using this unit's 'EXT IN' analogue inputs, to directly power 4 ohm AMT drivers (which have a pretty flat impedance plot) would this 4306 do the trick? Based on amir's 4 ohm measurement?

Say for a 2 hour music listening session at 85 dBA?

Amir how long is it held for at 4 ohms?
 
Top Bottom