[adolphe]

Good evenin' Y'all (I feel that's the appropriate way to start a conversation now that I live in Kentucky )

I've been working on a PCB design for a headphone amp called the "Zeny". I've been working on it over at diyaudio.com, but I thought I would bring it over here for some more exposure so that maybe I can get some more input and suggestions.

The Zeny is based on the teachings of Nelson Pass, with the schematic done by jh over at diyaudio. I've simply taken that schematic, done a pcb layout for a single channel and a psu, and also made a digikey parts list. I'm new to PCB design, so I'm looking for tips on routing, ground planes, and any other "constructive" criticism you might have.

I'll likely order a proto board in a few weeks. If I like the results, I'll likely have a small run of boards made available for people to purchase if there's interest, and of course after getting the approval from Mr. Pass and jh. It seems like there's a lot of the "same old" solid state amps being built and I figured this one might be a nice change.

I'll post up the schematics first. PCB artwork will be posted in a bit.

Mono channel schematic:

PSU schematic:

Mono channel pcb artwork:

PSU pcb artwork:

-Scott

 

[adolphe]

A couple notes about the design...

Namely, the output resistor that's listed at 500 ohm's, that's what jh used with his high impedance headphones to avoid having his children "accidently" turn the volume up too loud while listening. For my Grado's, I'm going to leave it out.

Also, I'm designing the grounding scheam around using the psu board as a ground plane or star. I've added extra spots for the signal grounds to connect to it. I also left a couple spots on the regular amp board for the signal grounds for the purpose of the prototype so that I can experiment to see which gives the quietest result.

Finally, the output and input cap value I'm not sure I agree with at this time. It's what jh used, so my first trial will use that, and then I'll begin to experiment with different values and types.

Here's the Digikey parts list that I threw together.

For a singal channel:

For the PSU:

-Scott

 

[peterpan188]

Very nice work! First time I have seen a simple SS design at this forum (just may be I am too new

). Would you mind explain a little on the topology of this amp?

Regard,
Peter

 

[robzy]

Are you taking into account heatsinking requirements for the PCB layout?

Rob.

 

[adolphe]

I haven't done exact heat dissipation calculations yet, but I placed the FETs at the edge of the pcbs so that just about any sized heatsink can be used. The heatsink that I listed in the digikey BOM is inadequate for the PSU (pointed out by a fellow at diyaudio).

-Scott

 

[adolphe]

Peter,

For a good background on the theory behind the Zen series of amps, you can start with this this article, http://www.passdiy.com/pdf/zenamp.pdf

Then if you have more interest in this type of design, you can continue to read the Zen articles available at www.passdiy.com

Note that these amps are designed for driving speakers under normal circumstances.

The the Zeny is a modifed Zen amp. It uses cascoded transistors to create a single ended Class A amp with unity gain.

 

[ggta]

Very interesting!

 

[mb3k]

Pardon my ignorance but does the Zeny have any connection to the "Death of the Zen" (DoZ)?
http://sound.westhost.com/project70.htm

 

[adolphe]

No this is not related to Death of Zen. I believe Death of Zen was created by Rod Elliot because he didn't like the Zen amplifiers and wanted to design something "better" using the same principles.

-Scott

 

[ggta]

Do you have some news?

 

[MASantos]

Looks great. With a parts cost of about 80$ for parts it would compete with low cost pimeta.

With my very limited knowledge of amp design here are some opinions and questions , please take it with a grain of salt.

1. I am not familiar with The Zen amplifiers, but I always heard that capacitors in the signal path are not desired. And that if any capacitor must be used, then electrolitics are the really bad. I don't want to argue about this subject as it is very controversial but could you please develop the use of the input and output caps.( I will read the articles about the zen amplifiers, maybe it is explained there.) Furthermore, in your parts list you have a Pana FC capacitor for the output cap. These caps are considered by many around here as unsuitable for signal paths. maybe you could try a so-called "boutique" cap such as Elna's Silmic II or even blackgates. Same goes for the input cap, maybe provide more area in the board for bigger(axial) caps.

2. In the PSU, you may want to add snubber caps in the rectifier section. those murata diodes are quiet already, but snubber caps seem to help in almost any situation.

3. Is this suposed to be a two channel amp or is it possible to use a 3 channel configuration such as the one in the PPA or M3?

4. Is this amp design specificaly for any kind of headphones? Low or high impedance?

good luck with this project, it coud be my 2007 winter project!!

 

[Uncle Erik]

Can you give us a little background on the IRF6201 and why you chose it?

Also, would it be possible to bridge the design? This could be an inexpensive DynaMight competitor if you could build it bridged and balanced.

 

[adolphe]

Glad to see there's some interest in the design. I haven't yet built it, but I do still plan on it.

To answer MASantos' questions in order:
1) Yes you are right about the capacitors. Something other than electrolytics should be used. From my readings, polypropolene caps seem to be the best in general before exploring the boutique caps. Output caps are required in this design because of the type of power supply used. I'm not knowledgable enough in amplifier design at this point to determine if a bipolar supply could be used instead (this would allow for no coupling caps I *think*. Someone please correct me if I'm wrong).

2) I agree with the caps across the diodes.

3) This is a two channel design. No "ground channel".

4) The person who I have borrowed the design from said that it can be used with any impedence headphone essentially. I believe he used it most with higher impedence ones (300-600ohm), but I plan to use it with my HF-1's.

As for the IRF620...again, this was used in the design by jh. There's likely a few other IRF power mosfets that could be used in its place if you felt like it. But also again, I'm not experienced enough with amplifier design to start deviating much from other people's designs.

I'm away from my usual computer with all my files on it, but I will update the drawings and BOM's soon.

-Scott

 

[ATAT]

Quote:

Originally Posted by MASantos Looks great. With a parts cost of about 80$ for parts it would compete with low cost pimeta. With my very limited knowledge of amp design here are some opinions and questions , please take it with a grain of salt. 1. I am not familiar with The Zen amplifiers, but I always heard that capacitors in the signal path are not desired. And that if any capacitor must be used, then electrolitics are the really bad. I don't want to argue about this subject as it is very controversial but could you please develop the use of the input and output caps.( I will read the articles about the zen amplifiers, maybe it is explained there.) Furthermore, in your parts list you have a Pana FC capacitor for the output cap. These caps are considered by many around here as unsuitable for signal paths. maybe you could try a so-called "boutique" cap such as Elna's Silmic II or even blackgates. Same goes for the input cap, maybe provide more area in the board for bigger(axial) caps. 2. In the PSU, you may want to add snubber caps in the rectifier section. those murata diodes are quiet already, but snubber caps seem to help in almost any situation. 3. Is this suposed to be a two channel amp or is it possible to use a 3 channel configuration such as the one in the PPA or M3? 4. Is this amp design specificaly for any kind of headphones? Low or high impedance? good luck with this project, it coud be my 2007 winter project!!

he doesnt have a VGND. so obviously it needs an output coupling cap.. as the signal is riding on the power supply..

Also, There's no sense in using it in a 3 channel config as far as I know since again, there's no vgnd.. I cant think of how you'd do it atm..

In response to adolphe - I think you can float signal ground at whatever the DC offset is.. and that would probably work.. Then you could also run a ground buffer for 3 channels.

It looks like it'll work with low imps since the FETS will give more current than anything needs.

 

[digi01]

Quote:

Originally Posted by mb3k Pardon my ignorance but does the Zeny have any connection to the "Death of the Zen" (DoZ)? http://sound.westhost.com/project70.htm

Doz is BJT based amp.zen is MOS-FET based.the circuit is no connection anymore.

Z