Driver and enclosure for 4 watt amp, 12x12 room, moderate volume chamber music?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
:soapbox:
sreten - have you actually heard one (or more) of the Decware amps, or researched the design?

I have - more recently than the model and vintage of the review in your link - of course the price has increased in the 10yrs since (now at $775 for the factory built unit - but still available in PCB kit form for $300 USD)

It was an audition to one of these approx 8yrs ago that opened my eyes to the appeal of flea powered SET amp, and I've built/owned more than a few kits and scratch build DIY amps since - including Bottlehead Paramour 2A3s, George Wright Mono7 300B, and a variety of SET and P/P EL34 and EL34s. Several of the scratch builds were as close as possible to the freely published Zen schematics a possible without the custom OPTs.

This is quite a nifty little piece, and as Dave mentioned the combination of Steve's in-house built OPTs, and the resultant operating conditions are a bit out of the ordinary. It really is quite more comfortable with lower impedance loads than one would normally suspect. No, it's not a Krell arc-welder, or an ARC Reference 610T, and doesn't pretend to be the last word in triode-ness, but it's a very sweet, fast and musical little gadget that I'd wager could fill the bill for the application that zenga describes.

If he has space to accommodate a speaker system with more efficient driver(s) than the Mark Audio /CSS EL70 that has been discussed for the past few posts, so much the better. ( i.e. Fostex FE126E in "Horns" or DIY designs is widely enjoyed by a lot of owners of the Zen amp)

sorry if that sounded a bit petulant
 
I don't know if it helps, but I'm listening to a low power SET with full range speakers. I'm using the Fostex 127E, the now discontinued sibling of the FE126. My amplifier puts out somewhere between 1W and 2W. To get an amplifier twice as loud would require 10W, so although the Decware Zen amp is more powerful, it's still likely that my amplifier gives some useful insight.

With this roughly 90dB of full range speaker I'm using, in a room that is 12' x 15' it is loud enough on full volume to listen to music providing the kids aren't disturbing me. It's on the edge of being loud enough and it's an amplifier I routinely operate at full volume. With some 88dB Fountex speakers I made the amplifier is only used at full volume (this speaker is destined for a SS amp).

A SET amplifier couples the output tube to the load with a transformer designed to ensure the tube is operating under the right conditions - with good power and low distortion. It's possible to match the tube to different speaker loads by having different taps on the transformer; if there's a dedicated 2 Ohm tap then you should be OK. If it doesn't have a dedicated 2 Ohm tap then you had better experiment. As far as I can see, if you have an opportunity to match your speaker to the amplifier you already own and therefore, should choose a speaker with an impedance that matches one of the taps on the output transformer. If it doesn't have any taps and is described as being suitable for 8 Ohm, then use 8 Ohm nominal speaker impedance. Multiple drivers per speaker are potentially an added layer of complication.

Hope this is useful, but bear in mind that my experience is limited - I know Dave's a big user of Mark Audio drivers, he has listened to them extensively as well as Fostex drivers and knows their performances well. His advice has always be useful for my projects.
 
I don't know if it helps, but I'm listening to a low power SET with full range speakers. I'm using the Fostex 127E, the now discontinued sibling of the FE126. My amplifier puts out somewhere between 1W and 2W. To get an amplifier twice as loud would require 10W, so although the Decware Zen amp is more powerful, it's still likely that my amplifier gives some useful insight.

With this roughly 90dB of full range speaker I'm using, in a room that is 12' x 15' it is loud enough on full volume to listen to music providing the kids aren't disturbing me. It's on the edge of being loud enough and it's an amplifier I routinely operate at full volume. With some 88dB Fountex speakers I made the amplifier is only used at full volume (this speaker is destined for a SS amp).

A SET amplifier couples the output tube to the load with a transformer designed to ensure the tube is operating under the right conditions - with good power and low distortion. It's possible to match the tube to different speaker loads by having different taps on the transformer; if there's a dedicated 2 Ohm tap then you should be OK. If it doesn't have a dedicated 2 Ohm tap then you had better experiment. As far as I can see, if you have an opportunity to match your speaker to the amplifier you already own and therefore, should choose a speaker with an impedance that matches one of the taps on the output transformer. If it doesn't have any taps and is described as being suitable for 8 Ohm, then use 8 Ohm nominal speaker impedance. Multiple drivers per speaker are potentially an added layer of complication.

Hope this is useful, but bear in mind that my experience is limited - I know Dave's a big user of Mark Audio drivers, he has listened to them extensively as well as Fostex drivers and knows their performances well. His advice has always be useful for my projects.

Very helpful, thanks Bigun. Both you and Dave suggest using one driver, which sounds like the simpler solution.
 
frugal-phile™
Joined 2001
Paid Member
Current version of the amplifier.

0030.jpg


Original Zen Triode Amplifier

model SE84C+

2 watts x 2 or 6 watts x 1
Single Ended Triode Class A1
SV83 or EL84 Output tubes
For speakers between 1 & 8 ohms.

Features 2 inputs and Volume Control,
no preamp is needed, but you can still
use one.

note the 1>8 ohm bit...

Further, from here Audio Tube Amps by DECWARE(tm) Zen Triodes (noteL Steve is a master at turning a phrase)

An 8 ohm load yielded around 1.5 to 1.9 watts RMS. A 4 ohm load gave us around 2.38 watts RMS and a 2 ohm load gave us around 3.21 watts RMS. In the same tests, the amplifier was bridged using our series bridging technique made possible by floating the outputs. In this configuration the now mono amplifier exactly doubled all the figures above.

more: http://www.decware.com/newsite/articles.html

dave
 
You get more power by steepening the load-line for the output tube, it swings over a wider range of current. if you adjust the tube loading through different transformer winding ratios (taps) you can keep the same load-line but the power will not then increase into lower impedance loads. If I understand this correctly. So if you are operating the amp with higher output power you must therefore be exposing the tube to a larger current swing at the anode. This also increases distortion - there's no free lunch. Performance will be better when the amp is operating at lower power and this means the most sensitive speakers.

I'm not the designer of the amp - not an expert by any means. But as you say, people with commercial interests always have an incentive to promote what is good for their business whether it be through mastering a turn of phrase or otherwise.
 
SET design

Someone asked if I could look at this thread and maybe clear up a few things.

I designed the Zen Triode amp (SE84C) with an unusually high primary impedance of 9800 ohms into a 6 ohm secondary so that when a 4 ohm speaker with less than 2 ohm dips is used, the reflected impedance back to the output tube causes the primary to drop and couple more power to the loudspeaker. This is just the opposite of almost all SET amplifier designs which would put about a 4000 ohm primary into an 8 ohm secondary (with this tube) so that when the impedance of the loudspeaker drops the tube puts out less power and control. This is why most SET designs enjoy a bad reputation in the bass department.

When choosing speakers I try to keep it as simple as possible, meaning little or no crossovers. NOT because the amp won't drive a complex crossover, it will and with excellent finesse. But because crossovers spend a portion of the first watt into heat before it reaches the voice coil and I only have 2 watts. And because crossovers inject phase and timing errors into the playback.

I also try to get the most bass response possible from a given choice so that I don't get suckered into running a sub. Most people make the mistake of assuming a small amp like this should run small speakers, but it's just the opposite really. Use the largest speaker you can without killing the potential for good imaging and soundstage. Giant cabinets will sound big, but do not disappear as well.

Steve
 
Hi p10,

I'll also note Steve D. is a master at turning a phrase, but its his site, for
everything sensible he says I've tend to have found something ridiculous.

Given his claims your right and I'm wrong. How he arrives at the fact
(if its true) optimising single ended windings for 2 ohms is best is beyond
me, and so are the claimed power figures, single ended either runs out
of current (assymetrically) or voltage clips, whichever first.

I'd guess voltage clippping might be less onerous than assymetric current.

Still I'm having difficulty with the power figures, the claimed profile implies
some current and voltage limitations, but for single ended current is fixed.

It should increase in power with load until voltage clipping is hit. I'll also
note he says lower impedance with the same efficiency will be louder,
not warning that the opposite for the same sensitivity will be true.

Basically I don't trust what he says as far as I could throw him, which
is not very far at my age, but have to agree 4 ohm speakers apparently
would seem to be a better option than 8ohm, and might be in real life
more advantageous than indicated, depending in the hyperbole.

Certainly for single ended the assymetric current clipping limit should be
very nearly approached with the nominal load impedance, and for higher
load impedances it should produce more power until the voltage clips,
i simply don't believe the current is set that high i.e. that the turns ratio
of the transformer is set that high, though its always possible to do that.

(The optional higher impedance transformer is even more smoke and mirrors,
recommended for 16 ohms loads, why isn't that best into lower impedance ?)

If it true, fair enough, but the consequences of such an oddball choice
are IMO fairly severe in lots of other related performance terms.
(That is what defines "oddball", more pragmatic options exist.)

But have you any idea what the damping factor of such a feedbackless
amplifier would be with 2 ohm speakers ? Makes the sound a crapshoot.

Personally I simply don't believe what he saying, that it goes against
the grain of all other valve single ended amplifiers behaviour is to me
no surprise, the surprise would be (like some other claims) that its true.

Whatever happened to measured reviews ? The facts would be simple.

rgds, sreten.

Just like to know the actual facts for optimum speaker matching.
 
Last edited:
Fairly low i'd hope. From the published power transfer, i'd guess on the order of 1. (maximum power transfer when output impedance = load impedance)

dave

Hi,

Well hopefully its not 1, that it implies as much power is lost in the amplifier as
is delivered to the speakers, not good for low powered single ended amplifers.

The (maximum power transfer when output impedance = load impedance)
maxim is only true for basic electronics, you quickly find out its a pretty
useless rule, but true, for non ideal voltage sources (it implies huge currents)
and that in the real world in nearly all cases it works rather differently.

I designed the Zen Triode amp (SE84C) with an unusually high primary impedance of 9800 ohms into a 6 ohm secondary so that when a 4 ohm speaker with less than 2 ohm dips is used, the reflected impedance back
to the output tube causes the primary to drop and couple more power to the loudspeaker.

Does make sense initially, but essentially its a mismatch, and as its
single ended, not push pull, if the transformers are designed for the
correct 6 ohm primary standing current its also "artifice" nonsense.

The only way to build a single ended amplifier that works into 2 ohms
is to build one with very poor relative performance into 4 to 8 ohms.

(Not that I'm saying that, I'm implying 6 ohms is as low it goes for SE's.)


rgds, sreten.

The cohesive technical facts would be nice.
 
Last edited:
Thanks for the information Steve, nice to see you here.

That was a pretty negative assessment, sreten, not just of Steve's work but also of his character. I don't know anything about the electronics part of your post, which you seem knowledgeable about, but your conclusions seem a little over the top to me, considering that back when I was using the Zen amp, I had hours of enjoyable listening.

At the moment all I have on my list is the FE126En, 8 ohms and 94 db. From what has been said here maybe a better match would be 4 ohms at 94 db or up. Is that correct, and if so, is there such a driver that would be considered 'musical'?
 
frugal-phile™
Joined 2001
Paid Member
Hi,

Well hopefully its not 1, that it implies as much power is lost in the amplifier as
is delivered to the speakers, not good for low powered single ended amplifers.

The (maximum power transfer when output impedance = load impedance)
maxim is only true for basic electronics, you quickly find out its a pretty
useless rule

A piece of information that i 1st learned from an old Western Electric engineer -- and the 1st glimmers that i absorbed that made me realize that high damping is not necessarily a good thing. He couldn't see any sense in making an amplifier where the output impedance wasn't equal to the load impedance. This was the standard in all the systems WE installed in theatres.

Whatever may be said, these amplifiers have proven their worth in the field, and one need not choose based on theory, but on actual real-world results.

dave
 
frugal-phile™
Joined 2001
Paid Member
At the moment all I have on my list is the FE126En, 8 ohms and 94 db. From what has been said here maybe a better match would be 4 ohms at 94 db or up. Is that correct, and if so, is there such a driver that would be considered 'musical'?

It might be a better match, but in a single driver, none of the usual candidates apply. The FE126 (often in the EdHorn, Doorman above used his with Frugel-Horn) has been proven to work well with the Zen.

Note that my actual measured efficiency of FE126En is 93 dB, a bit less than a dB up from the FE126e.

Lets have a look at "moderate" levels. I'm currently listening at what i would consider moderate... SPL meter says ~55-60 dB at my listening position. At ~1m it says about 5 dB more. This in a rool 24x36 with a 16' vault. Speakers are 89 dB so average power levels of a couple or 3 milliwatts.

dave
 
A piece of information that i 1st learned from an old Western Electric engineer -
- and the 1st glimmers that i absorbed that made me realize that high damping
is not necessarily a good thing. He couldn't see any sense in making an amplifier
where the output impedance wasn't equal to the load impedance. This was the
standard in all the systems WE installed in theatres.

Whatever may be said, these amplifiers have proven their worth in the field,
and one need not choose based on theory, but on actual real-world results.

dave

Hi,

Optimum power matching is also the same as halfway between a voltage
and current source, and only really relevant to power transmission, where
essentially it controls relative power transfer in each stage, theoretically.

In practise the magnetic saturation aspects were designed to achieve
this, but primary and secondary resistances are irrelevant, the whole
power system runs on relevant impedance matching, not resistance,
it ran and still does on lower current high voltage low loss.

Your old WE duffer knew his onions, as did that period speaker
designers, that is a world away from today, noting the output
resistance of a valve output transformer is not remotely 8 ohms,
that is far too lossy, its the reflected impedance that was 8 ohms.

They knew the difference between losses and amplifier "impedances".

Matching speaker impedance with modern >0.1R amplifiers is nonsense.

rgds, sreten.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.