non-Pass amps youv'e built/heard

[BrianL]

Jocko, et al,

thanks for the info.

Why is it that we're in the 21st century and still no one seems to be able to quantify "it just doesn't sound right" on something that measures excellently on at least a necessary, though obviously not sufficient, set of "objective" criteria?

I'm surely not saying that these "subjective" conclusions aren't real as I've heard such "differences" reliably, but I just don't believe that one has to resort to psuedo-scientific explanations to "quantify" the differences

Borbely has picked up on Cordell's tri-tone intermod test and that sounds promising to me since he states it correlates well to his subjective impressions. Maybe I need to build up one of these test setups.

[elektratig]

I happened to see your post above re the AKSA and thought I'd say hello.

Although I'm not good at the Stereophile description stuff, I agree with Paul that the AKSA is a fine, fine amp -- and I say that although I'm generally biased (no pun intended) against ss amps. The 100W AKSA that I put together (which is the amp that pmkap heard) is basically bone stock and had only a bare minimum of breakin when we listened.

That said, if you're going down to 2R, Mr. Feedback, you may need to look elsewhere. I think the 100W AKSA is fine with 4R, but 2R (!) is quite a load. No doubt your countryman Hugh Dean can you provide you with further information on this score.

[capslock]

Originally posted by Jocko Homo
Built several permutations of the Alexander amp, and they all sounded like every other current-feedback project I've worked on.

As Mr. Horse would say...."No, sir, I don't care for it at all."

Can't describe it, but it just doesn't sound right. And everyone who heard any of them said the same.

Jocko

Hi Jocko,

integrated CFB op amps usually have 10-50x the settling time to 0.01% compared to a good VFB op amp that has similar specs (i.e. bandwidth and settling to 0.1%). I have yet to find an intelligable explanation why that is so...

I have toyed the thought of building the Alexander amp for a long time. Its greatest advantage is that the buffer that converts the input voltage into a current is not an open loop design as in almost any other CFB amp. The type of VFB op amp used for this and the way it gets its supply leave room for improvement, though. Also, I think the big IGBT enthusiam has dwindled, wonder why...

What modifications of the Alexander design have you tried?

On a side note: rather than trying A/B tests that can be confounding, I am swapping the op amps in my DAC on a weekly basis, trying to find out which one I like to live with. The LM6172, which is a CFB design in disguise, has its strong points, but so far I allways seem to like pure VFBs better in the long run...


Eric

[Jocko Homo]

I suspect a lot of it is psychoacustics. I read an article in JAES about 22 years ago that made the point we still have no way of determining how and where a given distortion mechanism will affect a musical note. And we still don't.

As for the Alexander thing, I can't remember all the things we tried, but they all sounded the same. Even the CFB amp built from scratch with discrete parts. And all the line stages and I/Vs we messed with.

If you do build any, the non-inverting version is harder to stabilize.


Jocko

[dice45]

Eric,

give me some time, i am currently building an AKSA 100W amp.

I wanted to have a 100W subwoofer amp matching to the tube amps i like best. Due to recommendation by my friend Manfred who was about the AKSA's subwoofer preformance, i didn't dillydally around with other choices and ordered it.

The kit is already sitting here and the housing gets built within the next week or so. I decided to make this amp usable both for subwoofer use (built-in open-baffle subwoofer XO) and as fullrange amp as i really want to know whether i can be happy with a SS amp after so many years with tubes. Moreover, i equipped the amp with a proper 600 R input trannie to make it usable both for 47k and 600R input impedance. Driving an open-baffle subwoofer, the amp will have to work quite hard. But power should be enuff. I come back with results as soon as the thing is up and running.

All,
the best SS amp i heard so far are the 15 year old 30W Hiraga from L'Audiophile, Paris and the 8 watt "Le Monstre". Unless the Monstre is fed by batteries (170Ah) i could not decide which to prefer (probably the Hiraga) but a battery monster is really something,

Apart from that i heard solme very nice DC-coupled amps, but none was as vivid and juicy as the Hiraga 30W. Hadn't i needed more power, i would have built a Hiraga.

But i must admit i have not cared much for SS amps in the recent years.

[HarryHaller]

I have heard good reports on this design. How about an some suggestions for different output drivers and outputs. I wouldn't change the front end.

http://www.jls-info.com/julien/audio/index1.htm

Monster:

http://www.infomaniak.ch/~bonavolt/monster.htm
H.H.

[FBJ]

At first view I thought of changing the input transistors to Jfets.
Then I went to the monster link.
I like the design. Simple and straight forward. I sure Mr. Pass would have ideas about this design but he likes opamps.

[Geoff]

Harry

If you are going to post links related to the Hiraga designs, why not include the site that has the full articles for both the 20W and 'Le Monstre' instead of sites that just have a schematic and a few photographs?

The site in question can be accessed from the icon below (ie it's mine :-)

Geoff

[HarryHaller]

Quote from grataku:

Yeah Harry,
These pictures are all well but when are you going to post something that is actually relevant to this thread and DIY in general?

I thought the pictures were relevent in terms of layout and parts choice. It is one of the best non Pass amps that I have heard which I though was the theme of the thread. If I had drawn a schematic would it have been relevent? If anybody else feels this way let me know so I can stop posting and everyone can go back to talking about heatsinks and device matching.

Thanks,

H.H.

[pmkap]

Bernhard,
On the 100w AKSA , you would be well advised to put a heatsink on the center BYQ28E rectifiers (if you bought the ps from Hugh). The center ones are the ones where both sides of the dual cathode commected diodes are both used and therefore subject to higher thermal stresses. Unheatsinked ones have been known to self destruct.

Hope you enjoy the AKSAs. Their bass performance is quite astounding, as well as the rest of the range. Very interested to hear you feedback.

Regards,
Paul