[jcx]

Quote:

The ESP-950 does very well with my KGSSHV. The extra power improves the bass response, but it still isnt as good as the orthos.
Dont feel that a better amp is a necessity tho. The Koss E-90 plastic amp aint no slouch. It sounds great! 

at 600 V supply for the E-90 vs the KGSSHV 450 V the E-90 will drive the ESP-950 to higher SPL at bass frequency, as well as the rest of the audio range by  ~ 2.5 dB (from: 20*log10(600/450) )
 
http://www.head-fi.org/t/582518/electrostatic-amplifiers-voltage-ratings#post_7924572

 

[Anaxilus]

Quote:

at 600 V supply for the E-90 vs the KGSSHV 450 V the E-90 will drive the ESP-950 to higher SPL at bass frequency, as well as the rest of the audio range by  ~ 2.5 dB (from: 20*log10(600/450) )
 
http://www.head-fi.org/t/582518/electrostatic-amplifiers-voltage-ratings#post_7924572

 
Sadly your numbers don't jive w/ the sound in this case.  The KGSSHV rocks the 950 in bass response and dynamics compared to the E90.  It's not even remotely close.  E90 is thinner and more flacid compared to the punch and richness of the HV.  Sounds like you are measuring the wrong thing to me.

 

[stuckonsound]

Yes I've had a good number of Ultrasones and also the K701. If the bass is better than the K701 I would be pretty happy. I don't expect monster bass from a stat.

On a side note it looks like the prices have been pushed up toward msrp. I used to see these in the $600 range on Amazon.

I'd like to add on the bass, imho the 950 does better than a lot of non-basshead cans in terms of musicality/extension - for example the Sennheiser HD 580; it extends deeper and hits harder. That doesn't mean it will compete with true basshead models (like what you might get from Ultrasone), but it's certainly not "lean" or "thin" along the lines of the K701 or MDR-SA5000. I don't know if you've heard any of those cans, but I commonly see people assuming (incorrectly) that the ESP/950 (as an electrostat) will have no bass whatsoever and they need an ortho in order to hear anything under 300hz. That simply isn't the case. Now if you want to rattle your skull apart, the 950s are probably not a good choice. I haven't tried the orthos (because tbh I don't like lots of bass or heavy cans), so ymmv. Regarding their natural-ness/musicality; they're among the best I've heard - but that's just me.
And imho their amp is perfect (I haven't even bothered looking at more expensive aluminum boxes) - does everything it needs to do, and they sound suberb. So what's the point in spending another $2000?

 

[obobskivich]

Yes I've had a good number of Ultrasones and also the K701. If the bass is better than the K701 I would be pretty happy. I don't expect monster bass from a stat.
On a side note it looks like the prices have been pushed up toward msrp. I used to see these in the $600 range on Amazon.

Yeah I've seen the prices climbing too - no idea what to tell you. I would regard the 950s as having at least as much impact as the 701s, with more extension. I might even go as far as saying they have more; given that they give the HD 580 a run for their money (and those are bassier than the 701). They are not PRO2900 level slam though. So if you want that, they don't get there.

 

[Magick Man]

The 950s do have a little more bass quantity and better quality than the AKG x701s. Not that the AKGs are awful, with the HE-300s they're the best in their range. ESP/950s are just superior.

 

[jcx]

Quote:

 
Sadly your numbers don't jive w/ the sound in this case.  The KGSSHV rocks the 950 in bass response and dynamics compared to the E90.  It's not even remotely close.  E90 is thinner and more flacid compared to the punch and richness of the HV.  Sounds like you are measuring the wrong thing to me.

all else being equal: "flat" amps, same headphones, the main determiner of relative perceptual frequency response balance is loudness - look up Fletcher-Munson curves
 
if you don't measure you don't know if you have matched SPL, "sounds the same loudness" with seconds of switching time won't get you even 1 dB "by ear"; 0.1 dB SPL, ~ 1% V match is required for rigorous subjective comparison
 
measurement is the only way to establish this minimum necessary condition for valid subjective evaluation - for ES amps direct V measurement isn't easy the output V, impedance makes special lab equipment necessary
 
cheap SPL meter accuracy is poor, reproducibility may be adequate, better than ears anyway - the positioning on the coupler gives more variability - don't move/remove the headphones on the coupler, just switch the amps back and forth until you get the volume settings that give equal SPL
 
of course blinding is also necessary to make sure the only info your brain is using is the sound reaching your ears
 
 
all of the above is a big threshold for subjective evaluation to meet so I do think reasoning from the published technical specs of the amps is likely more reliable for perceptual prediction in this instance
 
there is still the possibility that the E/90 has inadequate input coupling C - that I can measure, tweakers could easily upsize the caps - Koss isn't clear in the specs if the "8 Hz -35 kHz" includes the amp
 
I don't know how much to rely on purrin's measurements without more info on the setup, measurement details
 
old Stereophile review has specs - don't see why they would have changed:
"
Description: Electrostatic headphones with separate energizer unit. Frequency range: 8Hz–35kHz (no tolerance given). Sensitivity: 104dB at 100Vrms differential input at 1kHz. Diaphragm thickness: 1.5;um. E/90 Energizer/Amplifier. Frequency response: 1.6Hz–50kHz ±3dB at 100Vrms differential output. Input impedance: 100k ohms. Input level: 1Vrms for full output. THD+noise: 0.001% at 1kHz and 100Vrms differential output. Voltage amplification: 60dB. Channel separation: 80dB at 1kHz, 100Vrms differential output. Audio output voltage: 600Vrms differential, 2300Vrms push-pull with soft limiting. S/N ratio: 100dB at 100Vrms output. Bias voltage: 600V DC.
 

 

[WilCox]

Quote:

Yeah I've seen the prices climbing too - no idea what to tell you. I would regard the 950s as having at least as much impact as the 701s, with more extension. I might even go as far as saying they have more; given that they give the HD 580 a run for their money (and those are bassier than the 701). They are not PRO2900 level slam though. So if you want that, they don't get there.

 
Price on Amazon has dropped to $721. 
 
It's not surprising that Koss is playing around with the price point of these.  I think I paid ~$600 when I bought my first pair in the early to mid 90's.  I bought another pair last year at about the same price.  Not too many products can survive for almost 20 years without a price increase.  They are still a great bargain at $1,000 but cheaper is always better.

 

[obobskivich]

all else being equal: "flat" amps, same headphones, the main determiner of relative perceptual frequency response balance is loudness - look up Fletcher-Munson curves

if you don't measure you don't know if you have matched SPL, "sounds the same loudness" with seconds of switching time won't get you even 1 dB "by ear"; 0.1 dB SPL, ~ 1% V match is required for rigorous subjective comparison

measurement is the only way to establish this minimum necessary condition for valid subjective evaluation - for ES amps direct V measurement isn't easy the output V, impedance makes special lab equipment necessary

cheap SPL meter accuracy is poor, reproducibility may be adequate, better than ears anyway - the positioning on the coupler gives more variability - don't move/remove the headphones on the coupler, just switch the amps back and forth until you get the volume settings that give equal SPL

of course blinding is also necessary to make sure the only info your brain is using is the sound reaching your ears


all of the above is a big threshold for subjective evaluation to meet so I do think reasoning from the published technical specs of the amps is likely more reliable for perceptual prediction in this instance

there is still the possibility that the E/90 has inadequate input coupling C - that I can measure, tweakers could easily upsize the caps - Koss isn't clear in the specs if the "8 Hz -35 kHz" includes the amp

I don't know how much to rely on purrin's measurements without more info on the setup, measurement details

old Stereophile review has specs - don't see why they would have changed:
"
Description: Electrostatic headphones with separate energizer unit. Frequency range: 8Hz–35kHz (no tolerance given). Sensitivity: 104dB at 100Vrms differential input at 1kHz. Diaphragm thickness: 1.5;um. E/90 Energizer/Amplifier. Frequency response: 1.6Hz–50kHz ±3dB at 100Vrms differential output. Input impedance: 100k ohms. Input level: 1Vrms for full output. THD+noise: 0.001% at 1kHz and 100Vrms differential output. Voltage amplification: 60dB. Channel separation: 80dB at 1kHz, 100Vrms differential output. Audio output voltage: 600Vrms differential, 2300Vrms push-pull with soft limiting. S/N ratio: 100dB at 100Vrms output. Bias voltage: 600V DC.

 

Didn't Gilmore measure the E/90 for one of his chart/list posts? I vaguely remember the 2300Vp-p spec was found to be conservative, and the thing had like 4mA total output (I think it was 2 per side) - but I don't know what that explicitly does (performance wise) in comparison to other 'stats or if the 950s are adequately driven or what by that circuit. I would assume that they are, since that amp is designed from the ground up just to drive them (and vice versa), and much like the other Koss ESPs, it should be a good match as a result. But maybe I'm mistaken.

Anyways, I think the E/90 is a suitable amp, and I like the overall package - it's compact, it doesn't suck down power, both pots work properly, etc. And from that same Stereophile review (and the user manual, which is where those are pulled from - pretty sure you can still DL it from Koss, if not, I'll send it to you if you're interested) they spec the max continuous output at around 130 dB. I don't think the system wants for power...:basshead:

 

[spritzer]

Those power ratings are the theoretical maximum voltage swing of a given circuit and they are based on the voltage rails.  The E.90 runs at +/-600V and the only DIY amp to use that high a voltage is Kevin's own take on a stacked amp which can take +/-750V.  How the amp behaves into a load is a very different discussion as the limited current will cause problems.  There is also the small matter of electrostatic drivers should never see more voltage than bias*2 so 2300V is utter overkill. 

 

[jcx]

ouput current limit in ES amps determines the speed which it can charge/discharge the capacitive load = slew rate, which means the limitation comes into effect at high frequency, not bass
 
most music is estimated to have ~ 3-5 kHz "power bandwidth", the max frequency that the max V swing just starts to give slew rate limiting
 
 
ES amps Voltage rating specs can be a mess, mixing of units: pk-pk stator-to-stator vs Vrms on each stator - I think the Koss E-90 spec is confused, using Vrms where they have to mean Vpk-pk
+/-600 Vsupply alows ~1200 Vpk-pk on each stator, the 2300 V spec has to be Vpk-pk considering both stator's total difference - the absolute V never exceeds 600 V
stator-to-stator pk-pk is used to pump up the numbers for ad copy
 
but differential Vrms is used for sensitivity
 

-->20*log10(2300/2^0.5/2/100)
 ans  =
 
    18.203657 
 
so 118 dB SPL is a fair estimate of max possible ESP-950, E/90 max SPL from the specs

 

[BoogieWoogie]

How can I use the esp with a stax amp, there's a conversion cable ?

 

[spritzer]

You need a Koss extension cable and one of these.  The wiring setup can be found with a simple search. 

 

[BoogieWoogie]

Thanks, helpful as always. 

 

[obobskivich]

You can also buy them premade from ALO or APS (I forget which one, I think APS) for like $150. Or have the headphones completely re-cabled with an SR-007 cable (I don't know who does this as a service though).

I'd personally build it at home, but some people don't like to do that.

 

[spritzer]

I believe APS sold these but I'd urge people to stay away from them.  Never heard about any issues but the list of troubled transactions is quite long.  ALO and electrostatics, now that is a frightening thought....