Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › DIY amp for ESL headphones?
New Posts  All Forums:Forum Nav:

DIY amp for ESL headphones?

post #1 of 19
Thread Starter 

I want to build an amplifier for ES headphones to do some experiments.

 

Tubes are a good candidate, but as I have no experience (or components) I would prefer using transistors.

 

My plan is to build a simple power follower running on ~60V with a VAS in front and a 1:10-20 transformer on the output.

 

Is this a good way to go or should I look at a balanced amp with high voltage MosFets ?

 

Roger

 

 


Edited by McGyver - 3/23/11 at 11:41pm
post #2 of 19
post #3 of 19
Thread Starter 

cobaltmute: Looks very interesting, might just give it a try! smily_headphones1.gif

 

Roger

post #4 of 19

You need to do some research. This is a whole world in itself. The link above is a good place to start. Be aware of the safety issues involved. All electrostatic amps run voltages in the 600 range. This is not a good place  for beginning DIY.

post #5 of 19
Thread Starter 

I think I have done quite a bit off research, but most information I seem to find is for tube amps when ES headphones are involved! 

 

I build ESL speakers so I'm used to high voltage and this will definitly not be my first DIY project! wink.gif

 

I have a special liking to no feed back class-A SE amps, so my first thaught was to build a small version and use a step up trafo to get the needed voltage. As the small ESL element in a headphone should be a quite easy load for a trafo!?

 

My question was meant to find out if some off you have been down that road and would like to share information on how it worked out for you!

 

Roger


Edited by McGyver - 3/23/11 at 11:39pm
post #6 of 19

actually no, es headphones are terrible loads for xfmr, the headphone's  C is smaller than almost any high step up ratio xfmr winding internal parasitic C

 

the efficiency is hopelessly low, you still need large Lm to hold up the high V at low frequency so the core cross section is still large

 

you probably end up accepting some low frequency distortion from core nonlinearity and fight to get flat HF response even a half octave beyond 20 KHz

 

compared to direct drive, xmfrs are technically poor for driving ES headphones

 

post #7 of 19

Yup, from a technical standpoint direct drive is the only real solution.  Doesn't mean we don't like transformers for the same reason people like other flawed ways of reproducing sound.  I just went from a BHSE to a Firstwatt F5 driving a custom transformer box using Lundahl LL1627's and it certainly is an interesting comparison. 

post #8 of 19

Transformers get a bad rap because of the poor quality transformer boxes that Stax built to enable a traditional speaker amp to drive electrostatic headphones. Transformers really aren't suited for large step-ups at high voltage levels. However, a purpose designed transformer based amp with a low step-up ratio can be very competitive with direct coupled designs and has some advantages of it's own. Direct coupled designs will always win the measurements "war", but may not be to everyone's sonic preference. People tend to forget that classic tube gear like McIntosh used transformers and to some ears still crushes modern solid state designs.

 

The key to any transformer design is the transformer itself. There are no high quality "off-the-shelf" models. I built a transformer based electrostatic amp with custom Electra-Print 1:1+1 high level phase splitting transformers. These were speced for 20Hz-30kHz (-1dB) and a 1kv swing. In order to maintain bandwidth, the primary is limited to about 3k, and a 1:2 step-up is about the realistic limit of this type of design.

 

Some  advantages of a transformer based design are reduced parts count, more power supply design options, two stage designs equivalent to direct coupled 3 stage (transformer step-up), and perhaps most importantly, the ability to easily implement DHT's into the design. Disadvantages, besides technical measurements, are the increased size and weight of the amp, the lack of much published design information, and tthe substantial cost of the transformers.

 

As far as the sound of equivalent designs goes, you get the same arguements that divide the larger audio world into modern solid state "neutralists" and SET/vinyl classicists. I have a modest but well respected direct coupled solid state amp as well as a fairly ambitious transformer based tube design. I prefer the transformer design. YMMV.

post #9 of 19
Thread Starter 

Thank you for your answers!

 

From what I read, building a transformer coupled amp does not have to be a complete waste off time.

 

I'll build a scaled down version off my no feedback class-A amp and give it a try!

 

Roger

post #10 of 19
Quote:
Originally Posted by McGyver View Post

 

I'll build a scaled down version off my no feedback class-A amp and give it a try!

 


You'll probably need to scale your amp up rather than down. Ideally you'll want a gain of about 1k with  a swing of about 1kv. A more modest design, similar in output to the lower end Stax offerings, will still need to swing about half those numbers.

 

post #11 of 19
Thread Starter 

By scale down, I was thinking of the power needed!  bigsmile_face.gif

 

At 25% efficency, things need to be rather beefy if designed to drive speakers, but a headamp it will be "smaller" considering cooling etc!

 

Roger

post #12 of 19

for headphone amps ES drive can be pretty high power, 6-10 W per stator in Class A + up to ~ 1/2 again in predrive/voltage translation for some of the recent SS/hybrid CCS bias projects can add up to 50+ W, then some regulate the Hi V too

 

we should hope to see some newer designs with the past few years smaller IXYS HV mosfets to replace the disappearing CRT bjt in existing designs

 

a SS direct drive amp parts for 4 channels of stator drive w/ps could cost less than one of the Lundahl step up xmfr (less case, "audiophile" balanced vol pot, "jewelry"...)


Edited by jcx - 3/24/11 at 1:30pm
post #13 of 19

For me, building a direct drive electrostatic headphones amp using vacuum tubes is so much fun.  You don't even need a PCB.

 

Wachara C.

 

 

post #14 of 19

It's just as easy to do SS without a PCB.  smily_headphones1.gif

post #15 of 19
Thread Starter 

The PCB is not a problem - I have a small router like you have Wachara! wink.gif

 

I'm thinking off using small standard toroids as step-up. This works well for ESLs from ~200Hz and up, but not for the bass.

 

But for the power needed to drive ES headphones I hope it is enough!? ( 2x12V/240V "backwards" => 1:40  and ~480V (+margin) at 50Hz before saturation)

 

 

New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › DIY amp for ESL headphones?