• 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!

Electrostatic Headphone Amplifier Review & Comparison

kevin gilmore

Member
Joined
Feb 29, 2020
Messages
54
Likes
259
i was more thinking along the lines of the nfb-1 amp, and the kx-r model 20 preamp. and the momentum-hd preamp.
but if .0000000x thd is your thing, then you will never find an electrostatic amp that can do that.

you are probably going to love the raal headphone amp. -18db second harmonic.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
i was more thinking along the lines of the nfb-1 amp, and the kx-r model 20 preamp. and the momentum-hd preamp.
but if .0000000x thd is your thing, then you will never find an electrostatic amp that can do that.
My "thing" is transparency on what equipment is about. When I spent thousands on these Stax tube amps as I went up the line, I had no idea they had unleashed distortion on me. Stax's reputation is about fidelity and that is why I bought the line way back when.

With data in the open, people can make more informed decisions. If they like the glow of tubes, and know it is creating more distortion, and still buy it, more power to them. But to buy blind thinking it is higher fidelity is not right.

And no, I am not looking for zero distortion. I am just unhappy that the tube amp has tons and tons more distortion than solid state version. This should have been clearly documented by the company so I knew to avoid them and stay with solid state.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
My "thing" is transparency on what equipment is about. When I spent thousands on these Stax tube amps as I went up the line, I had no idea they had unleashed distortion on me. Stax's reputation is about fidelity and that is why I bought the line way back when.


Could you do a comparison measurement of one electrostatic amp and one headphone attached to it?

Showing equivalent measurements for both?

Measuring the amp at the same input level at which the associated headphone is tested?

Preferably at a listenable (not screamingly loud) level?

Amp measured while feeding the headphone and not an arbitrary load?

And any other practical considerations I may have missed?

In a new thread?

If you did and I missed it, sorry.

---

Extra Credit - repeat and compare for a conventional amp and headphone from your pile.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
Could you do a comparison measurement of one amp and one headphone attached to it?

Showing equivalent measurements for both?

Measuring the amp at the same input level at which the associated headphone is tested?

Preferably at a listenable (not screamingly loud) level?
How would that resolve the argument? The AP has the lower impedance so hooking up the headphone as well to it, would just add to the load of the amp, not reduce it. Did you mean something else?
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
How would that resolve the argument? The AP has the lower impedance so hooking up the headphone as well to it, would just add to the load of the amp, not reduce it. Did you mean something else?

Hmm...

Oh well.

Do the Stax sound good or bad now that you know?

What distortion levels do the Stax create at a usable listening level? (I don't listen extra loudly to headphones - same level as moderate speaker listening - say 75dB average, 95 peak, maybe less)
 
Last edited:

Tom C

Major Contributor
Joined
Jun 16, 2019
Messages
1,501
Likes
1,370
Location
Wisconsin, USA
I’m confused. Wouldn’t the idea be to remove the resistive load and substitute the headphones as the load? Might not make a material difference, but wouldn’t that reduce the overall load faced by the amplifier?
 

kevin gilmore

Member
Joined
Feb 29, 2020
Messages
54
Likes
259
So there are a bunch of problems here and the 100kohm dc load causes an unintended issue.

First stax amplifiers with tube output stages. Every single one of them except for the T2 have plate resistors as the current source loads.
plate resistors as current sources absolutely suck as they are clearly not linear over the intended voltage range. Now if you have one of the amps
that can take the constant current source modification board, then things get much much better. Get a mod board and find out for yourself.

In fact the only real difference between all of the solid state and the tube stax amplifiers is this plate resistor vs current source.
really early stax solid state amps also had resistors as the current source. If you find one that still works and test it with the current method you
will find that it also sucks. (The stax D10 is very different, and would likely test very well but is voltage swing limited)

A srm-007t modified with 6s4 tubes and the constant current source will perform MUCH better.
additional issues are that 6fq7 or ecc99 just suck as output tubes. Another reason why the T2 is much better.

So if you want the best currently manufactured stax amp, the srm-700s is going to be it, as the srm-700t still uses plate resistors.
Although a srm-700t with the current source board might be even better because the tubes are 6sn7b.

assuming you want to make measurements that are amplifier friendly, then the ap design note is it. With ne-2 neon lamps as arc supressors
directly across the input to the ap. even better are the ceramic versions (as used in the stax d50) which are nano-second fast supressors. also cheap and available. or get a pair of the tektronix probes and power supply, total ebay cost about $500.
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
I’m confused. Wouldn’t the idea be to remove the resistive load and substitute the headphones as the load? Might not make a material difference, but wouldn’t that reduce the overall load faced by the amplifier?

That's along the lines of my thought, but the AP apparently has 100k/200k(balanced) input impedance, which, when connected across the assumed very high impedance headphone load, changes (increases) the load seen by the amplifier.
 

sq225917

Major Contributor
Joined
Jun 23, 2019
Messages
1,360
Likes
1,610
Just measure them into the correct load and republish or accept that the test regime isn't optimum for the DUT.

You can hand wave all you want that its close enough, Kevin's clearly explained why its not correct. And frankly he's the expert in this instance.
 

pkane

Master Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
5,630
Likes
10,203
Location
North-East
Just measure them into the correct load and republish or accept that the test regime isn't optimum for the DUT.

You can hand wave all you want that its close enough, Kevin's clearly explained why its not correct. And frankly he's the expert in this instance.

Why is it so hard to understand that until you can produce measurements that show that Amir’s results are incorrect, all you’re doing is hand waving? Who is or isn’t an expert is not a way to judge the accuracy of measurements. Other measurements are the way to judge the accuracy of measurements.
 

kevin gilmore

Member
Joined
Feb 29, 2020
Messages
54
Likes
259
first, some have asked how stax measured their amps. this is it going WAY back
https://www.ebay.com/itm/Sound-Tech...542214?hash=item3fc005c106:g:eVAAAOSwWLBaPAz8
i'm sure they don't use this anymore, as neither they or anyone else can keep them working. 100+ opamps etc.
For a while after that they were using the panasonic 7000 thing. Both of these things are 100% analog and measure thd completely differently than the ap, and they generate significantly different numbers.

lets start with the +350v power supply and the 50k plate resistor. You now add the 100k resistor to ground, and that +350 power supply
instantly is now 233v. But the -350 power supply is still -350. The entire amplifier is now completely out of balance. It just can't work correctly
under these conditions. It was never designed to work correctly into anything but a pure capacitve load.

do the same thing with the solid state constant current source, and nothing changes. So while the 100k additional load is unintended, it still
causes little issue.

the issue with the koss is the same thing. The koss uses a regulated switchmode power supply that monitors total power consumption.
the 100k resistive load causes the amp to consume much more power, and when that happens the power supply shuts down.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
I’m confused. Wouldn’t the idea be to remove the resistive load and substitute the headphones as the load? Might not make a material difference, but wouldn’t that reduce the overall load faced by the amplifier?
The resistive load is inside the analyzer. It cannot be removed without surgery. It is not like headphone/regular amps where we use external resistors (of very low values).
 

kevin gilmore

Member
Joined
Feb 29, 2020
Messages
54
Likes
259
even with surgery you cannot remove the 100k resistor, you have to add an external attenuator.
to avoid ground loops there is some very tricky and patented circuitry inside the ap that compensates for all sorts of issues that would show up if the input was not fully isolated. There is nothing wrong with an external attenuator.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
Just measure them into the correct load and republish or accept that the test regime isn't optimum for the DUT.
The load is correct comparatively since the same is used for all four amps from two manufacturers. That aside, you don't seem to understand the measurement system. The analyzer itself has a cap on the highest impedance it can represent. There is no option of "just measure it with a different load."

You can hand wave all you want that its close enough, Kevin's clearly explained why its not correct. And frankly he's the expert in this instance.
He hasn't clearly explained anything. He has suggested me buying expensive, obsolete audio measurement gear with far lower performance to test his hypothesis that these amps should measure better. He hasn't provided any evidence that these amps can't handle the load as presented. The conclusions are bulletproof that the two tube amps hugely underperform the solid state ones. Your knowledge here is way too deficient to be judge of anything.
 

kevin gilmore

Member
Joined
Feb 29, 2020
Messages
54
Likes
259
adding the constant current source board to one of your amps will cost you no more than $50.
sounds pretty cheap to me. then it will measure the same as the srm313.

Or borrow a T2 and measure that. You will be much happier with the numbers.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
even with surgery you cannot remove the 100k resistor, you have to add an external attenuator.
to avoid ground loops there is some very tricky and patented circuitry inside the ap that compensates for all sorts of issues that would show up if the input was not fully isolated. There is nothing wrong with an external attenuator.
There is a lot wrong with me spending more time on testing these super niche devices instead of going on to other projects. As I explained, you have a theory of better performance. That's fine. If you care enough, rent an AP, build the attenuator, do the testing and show your results. Or buy those old analyzers to do the same. I am simply not interested in chasing other people's hypothesis with all the work and expense being on my shoulder. I have ton of other home and forum work to be done.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
adding the constant current source board to one of your amps will cost you no more than $50.
sounds pretty cheap to me. then it will measure the same as the srm313.
That's another theory :). But no thanks. These are presetine, low mileage amps that I should sell one day. On that day, I don't want to explain that I went in there and messed with them.
 
Top Bottom