NEW Electrostatic headphone amp

[Doc B.]

PB has been working on an ES headphone amp for a few months now. The ES project was something I had in the back of my head for several years, but it just never floated to the top of the development pile. A few months back Sajeev Ranasinghe of Nectar Sound contacted me asking if we had plans to do an ES headphone amp, and after some conversation he sent a demo pair of his very interesting headphones for us to listen too. http://www.nectarsound.net/?i=1

So once the development backlog was clear PB set about building up a direct drive headphone amp design. The circuit is a single ended input running into a two stage differential amp using the very linear 5965s and it features a bias supply high enough to run the Nectar Sounds, which require a 500-700V bias supply and can be had with the same 5 pin plug used on the Stax "professional" series. The nice thing about the Nectar Sound bias requirement is that the 580V bias requirement of Stax professional headphone models fits in the same range, so he set about cooking up a 580V supply for both brands of cans and we can also derive the 230V bias voltage of the older Stax cans as well. With the two styles of jacks (5 pin "pro", 6 pin original) that Stax uses available on the front and with the appropriate bias voltage assigned to each one, one could have an amp that will run the Nectar Sounds and any Stax ES headphone too.

We have three evolutionary prototypes in house and we have listened to Nectar Sound, Stax and Koss ES headphones. They all work as expected. The chassis has taken a turn away from our usual flat panel in wood base approach. The bias voltage supply, while operating at a teeny current, is high enough voltage-wise that we felt it made sense to completely enclose the circuit in metal. PB designed the layout so that it could be done as a kit. It would be a pretty involved build with a lot of sub assemblies to build and test. We're spitballing the kit price at around $800-900. If we deliver the amp as a finished product we are probably looking at $1200+.

The audience here is of course heavily biased (pardon the pun) towards kit building. We wanted to get some feedback on whether there may interest in our offering this as a finished product rather than a kit, considering the reasonably high skill level required for assembly. We'd like to hear your preference in this thread.

[oguinn]

I’d emailed Sajeev a few months ago about buying one of his kits but I ended up not - mostly because I wasn’t sure about his instructions would be up to the Bottlehead standard I’m used to. I never saw them, by the way, so maybe they are.

Anyway, I’d be in to build one!

[Doc B.]

Sajeev's amp kit is of the board and parts variety where you supply your own transformer and cabinet. And of course it's also a solid state circuit. He has put all of the details of building one of his amps up on his site as an open source design. We felt that the our amps are pretty different and thus the two circuits could give people a chance to try some different approaches.  I'm impressed with Sajeev's putting his whole headphone/amp project together, seems like a very competent guy.

[oguinn]

Yeah, I definitely got that impression. My comment should be taken as a criticism of my skills and not Sajeev’s.

Regardless, this seems really cool.

[Deke609]

The audience here is of course heavily biased (pardon the pun) towards kit building. We wanted to get some feedback on whether there may interest in our offering this as a finished product rather than a kit, considering the reasonably high skill level required for assembly. We'd like to hear your preference in this thread.

I wonder whether Headfi might be a good place to gauge interest in the pre-built amp? And maybe send out a few prototypes to some of the regular testers/posters there? Although I'm guessing that would require some kind of sponsorship payment since it's a commercial product.  But maybe it'd be worth it as a way of testing market temperature? 

I don't think I belong to the target market for either electrostatics or a pre-built amp - I like a lot of low end weight and, like most everyone else on this forum, I like building stuff myself. But kit-builders and diy types make up a tiny fraction of the market out there.

 Very cool, though. I hope it's really successful for you guys.

cheers, Derek

[Raymond P.]

Doc, you mentioned a certain high skill level required to build the ES kit. How difficult would it be compared to something like a Kaiju or BeePre?

[Doc B.]

This is a rough idea of what the build would look like, but without the C4S boards that will be installed and a few other things. Not the most difficult thing to build, but certainly a lot more involved than a Crack with lots of p-t-p work that gets a bit tight. Derek is on the right track with his observation - how small a niche is DIY kitbuilders who use ES headphones and want a tube amp? Would the sales volume be worth it to make this only as a kit? If it is only sold finished we don't have to go through the huge amount of work that it takes to write an assembly manual. That is hard to justify if we are only going to sell half a dozen kits. But if a lot of bottleheads have ES headphones or plan to try them it's certainly worthwhile.

This stuff is hard to gauge. For example, one of the kits we stopped making for a while was our very obscure tape path modification kit. It sold quite well then tapered off after a while. I guessed that maybe the folks who wanted one all got one. But then we got so many requests for it we brought it back, and it continues to sell quite well. I guess I need to trade a kit for a crystal ball.

[frankg]

I built a Crack and looking at the picture I don't see anything that scares me.  Just more parts.  I lean towards buying a finished product.  The high voltages may be a deciding factor for me.

The Stax amplifier market has room for a good quality tube amp.  You can easily pay more for a poor sounding Stax brand amplifier and may times more for other offerings.  I would guess the market for pre built is 100 times that of a kit.

[Deke609]

I guess I need to trade a kit for a crystal ball.

Just ask Magic 8-Ball!

http://magic-8ball.com/
http://www.ask8ball.net/

I tried it myself, but I don't think the results are reliable if the question is asked on behalf of another. See attached.

[Paul Birkeland]

I would also mention that this circuit is different than the other electrostatic amp offerings that I have seen in that it does not use a solid state front end, nor does it use feedback around the circuit. 

The amp with the C4S upgrades installed is also a tube roller's dream, as you can use 12AT7s, 12AU7s, 12BH7s, 5965s, and all those weird 12AT7 variants with lower gain.  The current sink in the differential stage at the output ensures that the stator pins are always sitting pretty close to 0V. 

[baseonmars]

I've build a SEX and have a Crack cooking up - this doesn't look scary and I'd love to build an electrostatic amp, especially with a constant current source.

One question - will this require any specialist equipment to build? My DMM is limited to 600V - I'm guessing this thing will be pushing voltages (all be it low current) around at well over 1000V.

[Paul Birkeland]

Although the bias supply is up around 580V, if you tried to measure it with your meter, the meter would draw enough current that the bias voltage would sag way down.  That would be one of the challenging sections to write in the manual, as different meters will give wildly different DC voltages at the bias output, and none of them will be correct. 

Otherwise the amp runs off a bipolar supply that is about -380V and +430V. 

[Doc B.]

Although the bias supply is up around 580V, if you tried to measure it with your meter, the meter would draw enough current that the bias voltage would sag way down.

I guess it's time for that VTVM kit.

[frankg]

How long does it take to develop an amplifier from prototype to ready for sale as a kit? 

[Doc B.]

Depends entirely upon what the project is, how many other projects are ahead of it, how much of the circuit uses things we know and how much is new research. The MonAmours came together in a couple months, using circuits and parts we are very familiar with. At the other extreme we've been working on developing a high powered amp for several years, with several prototypes having been built.

All that said the average timeline is probably around a year or two long from initial design development to released product. Within that timeline the part from building a first prototype to product release might be six months to a year. We seem to average about three prototypes before finalizing a production configuration, with the third one being the prototype that is photographed for the manual.