Jump to content
Pirx

Small amplifier for Stax

Recommended Posts

Hello

My little amplifier for my Stax headphones. First stage is based on ECC82. One triode per channel as cathodyne phase inverter.

The second stage is 6N1P-EW with CCS, and power stage is with E80CC with CCS too.

It was designed to work with my PC to drive Lambda Pro. So output voltage is not so big 308 V peak-peak, but it is enough for Lambda Pro.

 

 

schemat.jpg

02.jpg

03.jpg

3_DxO.jpg

DSC00095_DxO.jpg

DSC00101_DxO.jpg

04.jpg

6h1p-eb.jpg

DSC01598_DxO.jpg

DSC01608_DxO.jpg

  • Like 1

Share this post


Link to post
Share on other sites
9 hours ago, swt61 said:

Can I drive my Beats with that?

I am afraid you can't drive your Beats with this amplifier. It was designed only to drive electrostatic headphones.

Share this post


Link to post
Share on other sites

I tried to drive my KGSSHV once, but I found the lack of a steering wheel or gearbox very disconcerting........   :o

Sitting on heatsinks in mid-summer is also not to be recommended.........      :rolleyes:

Share this post


Link to post
Share on other sites

Some comments about the original post.

 

The overall topology is similar to the back end of Kevin Gilmore's original all triode amp, with CCS substituted for plate resistors.  The use of a CCS on the output stage will make it more efficient in driving the headphones compared to a plate resistor, which is beneficial.

 

But, the output tube is not a good choice.  Its max DC plate voltage is 300V, but more important, its max plate dissipation is only 2 watts which significantly limits the amount of current it can supply without burning up.  Consider substituting a 6CG7/6FQ7, which appears to be pin compatible.  It has a max DC plate voltage is 330V, and its max plate dissipation per plate is 4 watts (total plate dissipation is 5.7 watts).  Then you can increase the output CCS  current to set the output plate voltage 1/2 way between ground and B+.  The downside of this substitution is that this tube has a mu of 20 vs 27 for the E80CC, which is about 2.6 dB less gain,  You may need to decrease the feedback resistor R5 to maintain some overall feedback to stabilize the gain, but the increased current drive is useful, as devices seldom sound their best when driven close to their limits.

Share this post


Link to post
Share on other sites

I could imagine clipping if volume is a little above normal listening 89db or so or driving heavier loads. 

Share this post


Link to post
Share on other sites

Hows about whacking in some 6S4A's and calling it a mini KGST...?

Share this post


Link to post
Share on other sites

Thanks JimL, DefQon and wink for your posts.

 

It is not state of art to make amplifiers with tubes like everyone around :) I know 6CG7/6FQ7 are better in last stage, but I have about 40 pieces of E80CC and lot's of 6H1P-EB.

I love the way Tungsram E80CC looks like, only Tungsram's E80CC have holes in anode.

You are right about limitations of voltage swing and deliver current to headphones. This helps me to keep volume on healthy levels ;)

I designed this amp to drive my Lambda Pro while I am working on my PC, and Lambda Pro are easy headphones to drive.

With Lambda Pro max volume levels are still above levels I like listening to the music with pleasure.

Stax SR-007 are harder to drive and volume level is to low then. I am going to build bigger amp for SR-007 in the near future.

E80CC won't be in the last stage :) But 6CG7/6FQ7 and 6S4A won't be there too :-)

 

JimL - I have absolutely no idea about Kevin Gilmore's topology, but in my opinion 3 stages like this can't be called anyone's topology.

I ended in topology like this after many tries, it seems obvious to me, that many people could end their research with similar schematics.

 

Edited by Pirx

Share this post


Link to post
Share on other sites

The whole easy to drive is a bit of a loaded statement as while something works, it doesn't mean you can't do it better by adding some headroom to the design. 

Also grounded cathode is the bog standard of all bog standard designs so no copying here. 

Share this post


Link to post
Share on other sites

That amp had to be small and stronger tubes and higher voltages with CCS meant more temperature.

Sometimes you need to make decision - to make amp that folks on forum will call "awesome" or amp that you need but it won't impress the others :-)

I like listening with low volume level but I know that many people like high levels.

My amp is not good for everyone but it is perfect for me and Lambda Pro.

The others of my Stax'a headphones will have stronger and bigger amp in the future.

Share this post


Link to post
Share on other sites

i'm pretty sure i was the first person to do an electrostatic amp with dual local feedback at the output stage which in the 1950's was called a composite amplifer.

taking the original ges design and adding cascoded constant current sources at the output would definitely be an upgrade.

Share this post


Link to post
Share on other sites

Dear Kevin

 The only thing I am saying is that my project is mine, I don't know what you have done and when and how.

The second and the third stage is multiply by 4 in my amplifier and if you look at this, there is nothing special. Just two stages with CCS and simply NFB.

The first stage is also simply and known.

Is there anything hard to develop and design what could be called it is mine, yours, theirs etc ... ?

I would not call this NFB "dual". Just four simply amplifier, each with its own NFB.

Share this post


Link to post
Share on other sites

So it sounds like you came up with the circuit independently.  Not the first time that has happened, nor will it be the last.  In case you are not aware of it, Kevin Gilmore is, IMHO, the best headphone amp designer around.  He designed the Blue Hawaii, KGSS, KGSSHV, KGSS Carbon and Megatron electrostatic headphone amps, along with an excellent series of dynamic headphone amps.

 

If you are interested in building a more powerful tube stat headphone amp, I suggest you consider my own humble contribution, which is discussed in the "SRX revisited" thread elsewhere in the DIY section, and is simple enough to be built point-to-point.  The few people who have built it so far seem to like it.

Share this post


Link to post
Share on other sites

JimL - thanks for explanations.

My future projects will be something with DHT with chokes as a power stage.

I have build one like this, but it was too big and heavy. I needed small small amp so changed topology and build small.

Here is a pic of the last stage with DHT:

 

 

6C4C_ESL.jpg

Share this post


Link to post
Share on other sites

Chokes and electrostatics are a very bad idea.  Just look at the piece of shit that is the Woo WES. 

Share this post


Link to post
Share on other sites

 

A couple things.  Put simply, a choke is an inductor and an electrostatic headphone looks kind of like a capacitor.  Look in any electronics textbook and you will see that when you connect an inductor with a capacitor you get a resonant circuit, which means you can have a peak in the frequency response at the resonance frequency.

 

I said the headphone looks “kind of” like a capacitor.  It doesn’t exactly resemble a capacitor because it makes sound, which means it uses power.  The power spectrum of music is highest in the mid-bass to lower mid-range areas (approximately 50-300 Hz).   The impedance of a choke rises with frequency, which means that it requires more current to drive it at low frequencies, exactly where the headphone is requiring the most current and voltage to make music.  This increases the distortion of the output device.

 

The largest commercial plate choke I have found runs around 200H.  This has an impedance around 25 kilohms at 20 Hz, rising linearly with frequency from there so at 40 Hz its impedance is 50 kHz, which is a typical resistance for a stat amp plate resistor.  So at the lowest bass frequencies it requires/diverts more current from the headphone than a typical plate resistor.

 

This is all based on an ideal choke, ignoring the possible resonances and other imperfections of a real world device.

 

With a resistor load, you don't have the possibility of a peak in the frequency response but the resistor requires curent to drive it through the whole frequency spectrum.  In fact, with a resistor load, the amplifier actually wastes more current (and power) driving the resistor than driving the headphone.

 

By comparison a really good current source load demands a negligible amount of current, which means all the standing current of the output device is available to drive the headphones, which is what you want.  As I said elsewhere, it converts an amp for driving output resistors to an amp for driving headphones.  It is simply a better technology.  This is why Stax has used current source loads in its solid state amps in place of resistor loads since the 1970s.  

 

 

Edited by JimL
  • Like 6
  • Thanks 1

Share this post


Link to post
Share on other sites

That's exactly it and why the reactive load plus a choke is such an utter clusterfuck.  I mean seriously, everybody has to listen to the Woo WES at least once to bask in it's awfulness. 

Share this post


Link to post
Share on other sites

Thanks Jim. You're so kind. Always informative.

Interesting thing is, Stax uses CCS loads in their SS amps and resistors in tube amps. I could imagine 007t would be a nice amp with CCS.

Share this post


Link to post
Share on other sites

Was the WES we listened to at Chicago canjam malfunctioning in any way? That thing connected to the 007 mk2 was one of the worst sounding setups I've ever heard, and I'm pretty sure I was buzzed while listening.

Share this post


Link to post
Share on other sites

Nope, I even asked Jack if it was working correctly and he looked at me with a puzzled look and said "this is the upgraded version".  Yeah.... 

  • Like 1

Share this post


Link to post
Share on other sites

Let's not forget the total pile of crap known as the Eddie current Electra.

center tapped output inductors that saturate due to the DC problems at the input stage.

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×