The Best Half Watt Amp I ever heard

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
When I heard about the ZEN amp designed by Nelson Pass, I thought it was a neat idea, but had doubts. Rod Eliot, considering it, echoed my thoughts exactly when he said it wasn't hi fi. When I saw the distortion specs, I blew my top. I couldn't understand how or why anyone would create a monstrosity like that. I was actually disgusted.

I remember the night a friend of mine invited me to hear his new amplifier. It was a 40 W Marantz, and it was expensive ($400 in 1979). I brought my favorite albums over. He was being very coy and practically apologized for the amp's inferiority, before I even heard it. The best I heard up til then were similar to my brother's Kenwood system he bought in '71 in the Phillipines. It was real hi-fi, very clean and powerful.

But the sound of this 40 Watt Marantz amp with an ortofon cartridge, cheap turntable and cheap speakers Stunned me. I remember several thoughts in my mind at once. One was "how can it sound so good?" Another was was a Realization this was an Order of Magnitude Better than anything I ever heard. And that the information on those discs (albums) was far greater than I had known. I never realized the music sounded THAT good, and wondered whether it was the talents of the producer or musicians I was hearing.

I couldn't believe what I was hearing.

The speakers were junk, little 5 inch drivers from cheap compact stereos, with single capacitors for blocking the bass out, and a couple miserable Morse brand speakers, some horn tweeters. They were stacked up on opposite corners. It had sorry bass. But it didn't matter.

Every time I heard that thing I could close my eyes and the music was RIGHT THERE in the room, and NOT inside the speakers. The word " Effortless " immediately came to mind. I could HEAR the instruments. It was unbelievable to me.

I haunted a lot of hi fi shops, heard great speakers like the Dahlquist DQ 10. It is a six hundred dollar box with 6 speakers inside. It sounds like what it is : speakers. So many systems that should have blown my friend's system away. None of them did, I can't remember one that quite equalled it, though obviously it wasn't the last word in hi fi. I vaguely remember some pretty darn good ones that might have been as good or better. VERY FEW.

I tried to duplicate that sound, but always failed.

I want to create a system that good, or better.

So when I looked at the ZEN amplifier, with it's hideous distortion specs, compared to that Marantz (0.04 %, huge damping factor, 40 V/uS slew rate), I couldn't see the logic. I love music, it seemed a crime to create an amplified clock radio like the ZEN amp.

For a gag, I decided to build the Zen. I allowed maybe it wasn't horrible. But I didn't have all the parts, and didn't want to dissipate all that wattage. So I created a low power class A amp to drive headphones. The amazing thing is it didn't sound terrible. It sounded like any other hi fi amp I heard. That is a real accomplishment, to be able to create an ultra simple amp that sound every bit as good as some commercial unit.

But it lacked bass and treble. So I created a tone control network, violating the Law we all learned : Never use an inductor in the tone control network. I made the crossover freq. about 75 Hz, the treble around 5 kHz or so.

This introduced some loss, so I increased the gain of my zen style preamp. I wondered if it could drive speakers. I hooked it up to a pair of JVCs with miserable tweeters and midranges, having to add a piezo tweeter.

These speakers are rated 140 Watts, and there is NO way this flea powered amp could power them. But it did. Not only that, with the tone control network in there, it made the amps I have here sound like some kind of Mickey Mouse affairs. The reason is their inferior bass tone controls. On the scope I verified the power was one half watt maximum. My cheap old Cerwin Vegas didn't like the amp, but they might be damaged.

There is no way this amp can sound that good. It never blew me away. But it equals and betters in sound a LOT of run of the mill hi fi amps, including all I ever owned, due to my little tone controls.

I next created a modified JLH amp, removing the input transistor, thereby lowering the gain. I only got 5 watts due to my power supply, but it gave a good account of itself. So far I prefer the sound of the Zen type amp. I kept checking things over trying to verify it was really only a half watt. It must have been more than that, due to overdriving it, but the distortion was so un-noticeable, still, it must have been producing a distorted couple of watts.

I couldn't hear the distortion. I know it's mainly musically related even order harmonics. There was No Hint of any kind of harshness or sibilance, no flabby bass, no squawking, it just sounded like music. Probably no Zen type amp could stand up to any serious competition ? But I have yet to build a true Zen. I am considering the Penultimate version. That these amps would even be considered at all as hi fi contenders or competitors is interesting.

Nelson Pass introduced the Zen concept to encourage experimentation, and that I have done. I'm rebuilding it all now, with much lower hum in the supply. It was a kind of joke, an amp even simpler than the Zen. I can hardly keep a straight face looking at the schematic. No spec-head such as I was could take it seriously. Eleven parts. Yet it may be that a Zen type amp with the right speakers and a good program source can be the heart of a really good system.

My little joke of an amp, by itself, given what it needs, sounds like a decent hi fi amp with the volume turned down, and one that NO apologies need be made for. The Zen can only be better, and I think anything better than that would be Very good indeed.

Anyone who says the Zen is not hi fi must not have heard it, imo.
It changed my way of thinking.
 
so you build a headphone amplifier with tone controls... putting out
maybe 1/2 a watt of power... hook it up to bad speakers... and you
think you know how a Zen amplifier sounds?

Considering you were horrified after looking at the distortion specs,
and reading an article by Mr. Elliot, i wonder how you came by this
change of heart?

Maybe you should buy some BIG heatsinks and build the Zen amplifier
without modifications! (well, i'd reccomend the modified PSU
as written up in the Zen Revisited article).

Then get back to us on how you think it sounds :)

(btw, i think you are on the right track)
 
Sounds like you've found a home. I stumbled on this forum a couple of years ago while trying to repair my Adcom 555. The guys here helped me out but it was too late; "Hey", I said..."a Pass forum...cool." I chuckled when I saw there was someone here using a 'Nelson Pass' nick. Ha Ha funny...but...WOW! it's really the guy, and he'll talk to you and help you and on and on...and on. This all out of the goodness of his heart. I don't get it but I certainly appreciate it! I got the 'bug' right away.

Really and truly. I honestly believe there is not one single thing on the Pass forum you could build and have it sound bad. I look forward to your reaction after you've built an Aleph..any Aleph. You simply can't go wrong here!

BTW, I haven't so much as plugged my Adcom in since. It will gather dust until I sell it or whatever. Even my friends Krell KSA250 doesn't attract me much anymore...I think my 10 watt Aleph sounds better!:smash:
 
same

Thanks for all replies. :)

>... and you think you know how a Zen amplifier sounds?

The Zen MUST sound better than my amp; if it didn't, I could create a sensation by offering it as The Next Big Thing. So it must be pretty good. I don't have the skills of Nelson Pass, otherwise I'd have my own line of products. The difference between the Zen and my amp is mainly the current source. If you draw out the schematic from my description, you'll see it borders on the ridiculous.

IRF520 mosfet. (originally an IRF 610 ; (Rds (on) = 3 ohm (?) bad choice, my computer was down, data sheet files unavailable then). source to gnd. 3K3 resistor gate to gnd. 10 uF input cap. 8 K resistor gate to drain. collector of a PNP trans. to drain. base to gnd thru resistor (select for beta/ current drain ). emitter to V+ thru 1 ohm w.w. resistor. 1N4002 diode in series with resistor (select for current drain; I used 39 ohm). diode MUST be thermally coupled to PNP trans. same diode anode to V+, resistor to base. 10mF cap drain to spkr.)

I used about 750 mA current, got 8 V pp into 8 ohm dummy load and the scope showed that into the spkr. (1 watt). Supply : ~22 V. Originally a half watt, I bumped it up.

Designing (?) the thing was hard to take seriously. "The Zen's too complicated", I decided. I wondered how far minimalism could REALLY be taken. Pretty far, I think.

>Considering you were horrified after looking at the distortion >specs,
>and reading an article by Mr. Elliot, i wonder how you came by >this change of heart?

I listened to it, for hours. The music coming from the speakers was - musical. I enjoyed it, a lot. My amp must have several % THD. But no blurring, no harshness. I can't believe those speakers are efficient, how can a flea pwr amp drive them ?

The speakers are not the worst. The woofers are ok. 12 inch, heavy ribbed cones, good looking surrounds. V.C. dia. = not sure (2 " maybe). The dust caps allow some venting. (Read : NOT designed for to be driven by tiny amps) It was my cheap Cerwins that sounded bad, I suspect some damage, they sat around for years.

>Maybe you should buy some BIG heatsinks and build the Zen >amplifier
>without modifications! (well, i'd recommend the modified PSU
>as written up in the Zen Revisited article).

I think that's good advice. As I said, it seems obvious the Zen could only sound better.

>let us know if you need any advice, whatsoever.

Thanks, I shall.

>Welcome to the Pass zone.

Also thanks. I looked at the PDF files from the site, it took about 10 milliseconds to decide to d/l them.

>I chuckled when I saw there was someone here using a 'Nelson >Pass' nick. Ha Ha funny...but...WOW! it's really the guy, and he'll >talk to you and help you and on and on...and on. This all out of >the goodness of his heart. I don't get it but I certainly >appreciate it! I got the 'bug' right away.

I browsed the forum for a while, this is right up my alley. I spent a lot of time mulling a lot of issues over. Came to the conclusion class A bias was the way to go, and many nights trying to conceive an efficient circuit (textbooks always shows the n = 4 % model as grounds for dismissal). I note Nelson has explored the potentialities of class A to a great degree, and providing free info - the equation in my mind is : Nelson Pass = Santa Claus. "To what do I owe this unexpected boon ?" is the thought I still have.

I like the way Mr. Pass thinks, it's the same conclusions anyone must come to. I like the fact he has the guts to create horn spkrs, and lots of - I have to use the word "neat- o" type things, caring more about possibilities / actualities / realities and the fun of it than conventional 'wisdom'.

If I could create something to provide the level of hi-fi I have glimpsed, I would be content with it, no matter what it is. I love music, and I see how some commercial products skimp, and how it degrades the music.

I have to check out my remaining mosfets, I regrettably burned out many, and the next toplogy is a source follower.
 
You may have hit on the most important reasons for all this supprising enlightenment... First of coase, at least in our time, it's all the work of N.P. and the selfless sharing therof... Class A... That is were the specs and the reality do not coorelate... The distortion components from most other topologies today may be lower overall but, they are much noticeably more anoying to the human ear and mind... Electronic tools to measure such things do not work the same way as the human measurement system. They also do not measure based on music as a source. N.P. has explianed this better... www.passdiy.com
As for Rod Elliott... He has tons of usefull info and overall a great site... And, I don't even want to approach being so negative other than to say: What would you expect from something named Death Of Zen??? And, dosen't that design look strangly like a JLH circuit???
Happy Holidays
 
>.. The distortion components from most other topologies today >may be lower overall but, they are much noticeably more >annoying to the human ear and mind... Electronic tools to >measure such things do not work the same way as the human >measurement system. They also do not measure based on >music as a source...

I believe that's true. Many times I've worked up some little amp from 1 to 20 watts, and I know if the signal looks good on the scope, & no output d.c. offset,etc, the amp is ok, and it always is. If I see a wierd looking waveform, or it's dancing around a lot, it isn't. How it behaves with real music, I have no instrument to tell.

There's something about splitting a signal up into two halves as in class AB I don't like. Also, I've always had a nagging concern for that type of distortion I loathe the most, caused by power supply resistance ... a case of the signal modulating the power supply.

In test amps, I isolated each stage of an AB amp by feeding each stage V+ through a diode and decoupling cap. The improvement was dramatic. I reasoned that the resistance of the wire, tho very small, was still very real, and any change of V+ must cause a like current to develop in it. If the stage has good isolation, that resistance is just part of that stage and no concern. But it could be passed to the next stage and amplified. Even an emitter follower is technically a very low (negative) gain paraphase amplifier. The diodes & caps ensure V+ (and / if V -) to each stage is not modulated by the signal. Most of the time the effect is so small it adds up to nothing, and the cure is simple enough.

The only nagging concern was how to deal with the Large effect caused by the power stage current drain. Even idle current causes V+ to drop on amps I'm familiar with. Signal voltage will certainly modulate the supply. But this can't be a big problem since I know there's a lot of excellent AB amps. Only, I do wonder ... if that effect is ever really absent, short of a gigantic transformer, etc.... Nothing more I can say on that subject, it was always kind of a big concern of mine.

So I choose to abolish it with class A bias.

I like Eliot's site a lot also. Some good stuff there. He was, I suppose, approaching it all in the spirit of fun. I bet his little amp sounds pretty good; but if it didn't sound better than the Zen, to me it makes more sense to go with the Zen, due to it's simplicity.
Yes I believe he created a modified JLH amp.

I used to toy with the idea of experimenting with those cheap mosfets Radio Shack used to sell - (about the only ones I have left now.) I Never would've experimented with a Zen type amp if it wasn't for Mr. Pass, and that alone, to me, is - I put it this way : I listened to that thing for hours, and enjoyed every minute of it, and look forward to more experiments - a lot.
 
JCM, Basically what you describe is a need for low impeadance power supplys. N.P. has written about that. The suggestion to use 2 seperate supplies is always optimum. 1 thing I like to do also to that end, is to put the last set of Bulk, and of coarse bypass caps, right at the output transistor(s). With todays miniaturization it is easy to have a 10,000uF on your board, right next to each transistor. Many of them in paralell, lowers the total ESR.
That little bit of resistance,,, is also a little bit of inductance. With the last set of bulk caps at the transistor you are creating another small L or Pi filter with the wire and connections from the power supply... Another area of concern is the copper thickness and width on the board... I use at least 2oz. copper at the output area (usually a seperate board) and as much copper as possible...
As far as the distortion, I beleive we are sensitive to where in the signal the distotion is happening. Class A, I would think, has more distortion at the extreme top and bottom of the waveform, clipping being the extreme. These portions are almost always short in duration, with the transients of the music. Class B-AB has crassover distortion all the time! With any level! Of coarse that is were they try to optimize away what a class A dosen't even have...
 
Yes, exactly, a low impedance power supply.

It seems to me with class A, one could have a fairly high supply impedance and get away with it, which I like : substantial RC filtering.

I resolved some time ago not to use PC board traces to carry more than a small amount of current. The wiring I mainly use for the supply is 14 gauge, sometimes 18.

I remember an amp I repaired once - by resoldering the leads of the thick film hybrid power amp module - it had melted its own solder connections on the PC board. I won't mention the brand. Probaby very few (none) diyers would rely on such skimpy PC board traces. Probably it's overkill, but I figure I can't go wrong with thick copper wires instead.

I was hoping to hear a big difference between the Class A experimental amp I created and the AB hi fi amps I'm used to. Yet even though they all must have crossover distortion, I couldn't tell a difference. Only when overdriving them, my flea powered class A never had that horrible strained sound of the music getting buried, compressed. It simply distorted. So I now wonder if crossover distortion really is that bad after all, if it's commonly audible.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.