[maxxevv]

There are a few "Class A" portable headphone amps to be found on Aliexpress. But whether they are any good is unknown unfortunately. Saw one sometime back that lists it suitable for 8~600 Ohms. Not sure if I'm able to find it anymore though.

But most of these are pretty battery hungry, as such a huge 4000mAh battery may not last 8 hours. Versus a tiny Fiio that maybe half or 1/3 capacity and good for double that !

 

[vector84]

Would using an impedance adaptor the ZS5 widen our potential options?

A voltage divider based one like the iFi IEMatch or EarBuddy would widen your potential options enormously though they require significant volume headroom. They allow you to ignore output impedance (as it will be the 1 or 1.5 of the voltage divider) and raise input impedance to a flat 16 ohms, making coupling capacitors much less of an issue as long as they are sized properly (undersized ones will still produce predictable sub-bass roll-off based off 16 ohms input impedance).

Honestly it's shocking to me that iFi doesn't market them for such purposes...


Series resistor based ones would color the sound heavily as it's the same as having elevated output impedance, but should largely avoid coupling capacitor issues, I believe?

 

[young59]

Yeah I planned to make one but got the wrong resistors I will order the right one and will make an impedance adaptor

 

[vector84]

Yeah I planned to make one but got the wrong resistors I will order the right one and will make an impedance adaptor

Maybe someone with more knowledge of things could say for sure, but I think with a voltage divider based attenuator setup, you'd be more concerned with clean driving power - able to put out something like 10x (depending on attenuation factor) your desired power value into 16 ohms (or whatever input impedance your adapter has) without significant distortion, something that small portable players and phones might struggle with depending on implementation, but a real amp should do better.

 

[young59]

I intend on making an impedance adaptor as stated here .
https://diyaudioheaven.wordpress.com/headphones/headphone-attenuation-adapter/

But I plan to use lower values 15 ohm& 1.5 ohm instead of 33 ohm & 3.3 ohm.

 

[Slater]

Q1 MkII just showed up on (US) Amazon at $77 fulfilled by Amazon actually...

And cheap gear won't drive the ZS5 within specs...

Are you saying the FiiO won't drive the ZS5 within specs?

 

[vector84]

Are you saying the FiiO won't drive the ZS5 within specs?

FiiO cites recommended headphone impedance as 16-100 ohms SE and 16-150 ohms balanced so 4.8 ohms isn't really within specs right?

I'm pretty sure it will work just fine though

 

[xrk971]

Guys, really there's no need to avoid coupling caps. When implemented well, many of the top world class amps (like many SE Class A) use coupling caps and well, sound world class. Your DAC probably has a cap in the signal path. And you know what, in the studio where all the music is recorded using balanced line mics, preamps, mixers, all that XLR cabling with balanced lines to reduce noise pickup that pros use... well they have coupling caps (one ea) for the 600ohm line drivers. Anyhow, if you fear them for the sake of fearing them, you are reallly limiting yourself without good cause.

 

[xrk971]

measured again zs6 with different tips, trying to match same distance between BAs and mic (~2.55cm). used 192kHz 24 bits this time.

noticed differences between a narrow foamie and a bigger one; the narrow cuts off less 11kHz peak than the wider.
following @xrk971 hypothesis ("An 11kHz peak would imply something at 1.55cm deep - which may be the hard leading edge of the silicone earplug tip. So changing to foam tips may remove the 11kHz peak") which seems to explain the foamies effect, maybe more surface or more compacted foam in the wider ones help.

anyway, based on last measurements i got, have tried some simple EQ settings for silicone, small and big foamies. and i'm enjoying the zs6 very very much.

silicone
-2.5 dB 8000Hz q=6.0
-9.0 dB 11000Hz q=4.0

foam S
-6.0dB 11000Hz q=4.0

foam M/L
-4.0dB 11000Hz q=4.0

if you don't have a parametric equalizer, you'll get similar eq by using a band equalizer, if you can access to 11kHz band (and to 8kHz for silicones).

you can play with your desired gain (the above gains provide a bit more treble than zs5, but that's my preference), and with the frequency offset.

Spoiler: New graphs of ZS6 with different tips + EQ filters

(click to enlarge)



EQ filters action:

Silicone:


foam S:


foam M/L:

Super work Hakuzen!

 

[hakuzen]

Would using an impedance adaptor the ZS5 widen our potential options?

the impedance adapter was one of the solutions to tame the highs of zs6, due to its special impedance curve. but this is an exception: impedance adapter usually alters frequency response when applied to iems with BAs, which isn't desirable. this alteration, when applied to zs6 specifically, was a benefit casually.
in the case of zs5, you'll notice harsher upper mids and treble, and we don't want more sibilance in zs5. higher the impedance, harsher the effect. check the effect in one of my graphs, when applying a source output impedance of 15 ohms or something similar. i wouldn't consider it a solution.
https://www.head-fi.org/threads/kno...essions-thread.698148/page-1553#post-13777918

 

[hakuzen]

Super work Hakuzen!

thanks! you stimulated me to redo the measurements.

Guys, really there's no need to avoid coupling caps. When implemented well, many of the top world class amps (like many SE Class A) use coupling caps and well, sound world class. Your DAC probably has a cap in the signal path. And you know what, in the studio where all the music is recorded using balanced line mics, preamps, mixers, all that XLR cabling with balanced lines to reduce noise pickup that pros use... well they have coupling caps (one ea) for the 600ohm line drivers. Anyhow, if you fear them for the sake of fearing them, you are reallly limiting yourself without good cause.

there is no problem with output coupling caps, when you plug a high load, 600ohm phones, or let's say 60ohms, or an amplifier, to the output, or when the capacity of the caps is big enough.
but check what happened to bass when connecting zs5 (4.8ohms) to a source with output caps of 470uF (value you find in many daps or amplifiers):
https://www.head-fi.org/threads/kno...essions-thread.698148/page-1553#post-13777918

 

[xrk971]

thanks! you stimulated me to redo the measurements.

there is no problem with output coupling caps, when you plug a high load, 600ohm phones, or let's say 60ohms, or an amplifier, to the output, or when the capacity of the caps is big enough.
but check what happened to bass when connecting zs5 (4.8ohms) to a source with output caps of 470uF (value you find in many daps or amplifiers):
https://www.head-fi.org/threads/kno...essions-thread.698148/page-1553#post-13777918

I agree that 470uF cap is not good. It's too small. That's why I said, when well implemented.

The cap needs to be sized so that the f3 cutoff frequency is about 2 octaves below what the actual cutoff will be. So, for a 16ohm headphone, the equation for first order RC filter cutoff frequency is: f3=1/(2 pi R C)

Set f3 at say 5Hz (2 octaves below 20Hz), solve for C with R set to 16ohms and we get 2000uF. So in his case, a 2200uF cap (or two 1000uF Panasonic OSCONs in parallel) would give very deep bass extension down to 20Hz with almost zero attenuation and down to 5Hz with -3dB. The OSCONs have advantage of ultra low ESR and are what I use in my amps. Sometimes I add 10uF Silmic II bypass cap for folks who like a smoother sound.

 

[hakuzen]

I agree that 470uF cap is not good. It's too small. That's why I said, when well implemented.

The cap needs to be sized so that the f3 cutoff frequency is about 2 octaves below what the actual cutoff will be. So, for a 16ohm headphone, the equation for first order RC filter cutoff frequency is: f3=1/(2 pi R C)

Set f3 at say 5Hz (2 octaves below 20Hz), solve for C with R set to 16ohms and we get 2000uF. So in his case, a 2200uF cap (or two 1000uF Panasonic OSCONs in parallel) would give very deep bass extension down to 20Hz with almost zero attenuation and down to 5Hz with -3dB. The OSCONs have advantage of ultra low ESR and are what I use in my amps. Sometimes I add 10uF Silmic II bypass cap for folks who like a smoother sound.

love to know the maths/physics behind any thing, thanks!

 

[groucho69]

My ZuperDAC worked wonders on my laptop and drives the ZS5 easily. Even drives my HE-400i.

 

[skajohyros]

FiiO cites recommended headphone impedance as 16-100 ohms SE and 16-150 ohms balanced so 4.8 ohms isn't really within specs right?

I'm pretty sure it will work just fine though

My nx1s, which is supposed to be 16-300 ohms, works fine with the zs5 to my untrained ears. And even better with the cap bypass mod. I think these entry level amps are fine for us beginners.