At the risk of offending everyone...

[Christopher]

meaningless numbers?

This is starting to look like the old DBLT amplifer thread. I think MBK is making a valid point. How do the numbers hold up? After all, last time I checked, if we are talking fidelity then the numbers indicate how close your device approaches the "wire with gain" and thus mean quite alot. I wasn't into hifi during the numbers war of the '70s but I'm assuming they only published the favorable numbers and usually they were taken out of context? Just a guess. Anyway, MBK's approach of looking at the amp WRT where and how it's going to be operated is solid. Perhaps instead of bickering over how good this or that sounds, we actually look at the strengths and weeknesses from a theoretical POV? I already have a good OTL pair, so my interest is in other rooms of the house, and frankly using my amps during the summer.

I have done zero research into GC's but I like the convenience/cost issue, and, if they sound even close to my OTL's then I'd love to build a few.

A lot of people (obviously DRK is one) have a hard time beleiving that something this cheap can sound good. You're not going to prove it can by saying it sounds good, put some numbers up, I'd love to be a fly on the wall for that discussion.

Chris

 

[tcpip]

Where can I get details to build one?

Having read all these posts, I wanted to know where I can get full details to build what the chip-amp lovers among you call a "good" chip amp. It appears that there are significant audible differences between just the amp from the NS appnotes, and its fine-tuned version. I believe this implies I need to look for something called an "inverting LM3875"? Right?

Where can I get schematics and PCB? I believe I can get the other parts here in Bombay. And unlike some other DIYers, I'm not a real man... I am perfectly happy to buy a PCB and populate it. 😀

I'd also like to be pointed to sites where I can get details of tweaks, etc. And as I said earlier, I'm wimpy enough to use things like PCBs, so I'm not interested in trying P2P wiring. I hope I can get what you guys consider "good" results even with these constraints?

Any pointers, please?

TIA,

Tarun

 

[Andrewbee]

Where can I get schematics and PCB? I believe I can get the other parts here in Bombay. And unlike some other DIYers, I'm not a real man... I am perfectly happy to buy a PCB and populate it.

http://www.scott-nixon.com/

I'd also like to be pointed to sites where I can get details of tweaks, etc. And as I said earlier, I'm wimpy enough to use things like PCBs, so I'm not interested in trying P2P wiring. I hope I can get what you guys consider "good" results even with these constraints?

Any pointers, please?

http://greggbaker.net/cgi-bin/yabb/YaBB.pl?board=news

Have fun.

 

[Steve Eddy]

My comment on the numbers aimed at the argument that the laser trimmed accuracy of the integrated approach in chips beats accuracy of discrete designs achievable by the DIY'er. I don't see this advantage in accuracy in the data sheets.

True.

Also look at the production standard deviations. You can get around that by selecting the best out of a large number, but as with the PSU and resistor network tweaking: by the time you succeed, you might have built and fine tuned three discretes ;-)

Heheheh.

Interestingly the LM3886 for instance also exhibits phase shift below 800 Hz, p.12 of data sheet.

Well, phase shift in itself isn't necessarily a problem. Even a frequency independent delay, one which delays all frequencies equally, will cause phase shift, but group delay will be flat.

In general: good numbers may not suffice to make it sound good, but bad numbers likely don't help...
😉

Shhhhhhhhhh. Don't say that around the SET fans. 🙂

se

 

[Steve Eddy]

Re: meaningless numbers?

This is starting to look like the old DBLT amplifer thread. I think MBK is making a valid point. How do the numbers hold up? After all, last time I checked, if we are talking fidelity then the numbers indicate how close your device approaches the "wire with gain" and thus mean quite alot.

It can certainly mean quite a lot if your goal is purely objective perfection, or as close as one can achieve. But I'd like to think that what means the most to all of us in the end is how it all sounds to us. And improving objective performance doesn't universally result in better sound. Some will find it preferable, and some won't. And on those grounds, no one is any more right or wrong than another and there's simply nothing to argue.

se

 

[planet10]

Re: Where can I get details to build one?

put some numbers up

And we well know that the common numbers have little to do with how things sound.

Where can I get schematics and PCB?

The Not just another gainclone thread is a must read too.

dave

 

[DrG]

in ears we trust...

Seems to me some people would like a few LM number-specs to back up some of the miraculous achievements sometimes attributed to GC's... while others say 'who gives a toss what the numbers say, trust your ears!'

Broadly, there are two 'groups' in the last few posts: the former group should probably be aiming at excellent sound AND flawless measured parameters: if Mark Levinson is out of your league (as it is out of mine), then you guys would appear to have a good, discrete SS kit amp in your future. It's ultimately unavoidable. Even assuming a GC sounds magnificent, you're always going to wonder about those pesky data-sheet compromises.

And the latter group seems content with magical musical sweetness, totally disregarding any numbers whatsoever. My guess is there's probably a class A PP or SET tube amp in your future, if not yet in your living-room. Good, I like tubes.

Then there are those who think this is a pointless exercise in unconvincable (is that a word?) one-upmanship while yet others introduce an element of guilt toward my 'callous' disregard for the delicate sensitivities and possibly depleted wallets of the diy-fi newbies (all relevant apologies). Only these latter factions seem to have unambiguous support for, and unwavering faith in gainclowns

Good for you too! Why bother reading the whole newspaper when you can memorize verbatim one little classified ad. Hell, narrow-mindedness never killed anyone - unless your head got caught in a vice in the process. Exactly who is this DrG pest who dares to flout the Godclone?! What does he know (built a couple...) about hi-fi (sold it for 3 years...) or diy-fi (built plenty, tubes and trannies all) any bloody way? Which does NOT make me any kind of authority! Not even close.

But I know enough and have heard enough (Painclone included) to realize there are plenty of better-sounding, more load-flexible and more entertaining-to-build options out there.

Liberate tute me ex inferis audio...

 

[Beggar]

Re: in ears we trust...

Liberate tute me ex inferis audio...

So are you trying to say gainclones (I hate that name) are like the film 'event horizon' where once you go there, there is no return?

Nik 😛

 

[MantramAudio]

Re: in ears we trust...

But I know enough and have heard enough (Painclone included) to realize there are plenty of better-sounding, more load-flexible and more entertaining-to-build options out there.

Liberate tute me ex inferis audio...

You are right on track. Just as there are devoted groups of people that think the Beetle was NOT the greatest car ever made, but continue to pour $$$ into their '64 Beeltes, because it makes them happy. They could have fell in love with AMC Pacers, or Gremlins, or MG's, but they didn't; They love Beeltes.

It is the same here. That's why there are Pass forums, and ChipAmp forums, and tube forums.

Cest la`vie! (French, please excuse my abuse of your elegant language)


P.S. I hate the term GainClone as well. My amp is an LM1875DataSheetClone. I do NOT give 47 labs any credit; it all goes to National. The Gaincard is hype and marketing at its best.

 

[patriz]

for myself I don't care to evaluate cable and resistors, I'm not an electronicphile, I'm in search of electronics that allow me to enjoi my vynil collection. I sold all the expensive commercial stuff and made a multiamped horns system. Horns are easy load so gainclone are just right and as they are cheap I made one of them for each drivers. So even easier load. This is my best system up to now. For the electronics even the cheapest. Probably if you use Wilson Watt that's not so funny. After many years in audio I don't believe in "The best amp" thread or "The Class A components list" this is for magazine that have to be on the shelf every months. Audio is made of compromise, influence and small details. We have to think at the audio system as an unity not as single components chain.
I have to thanks Mr.Sakura for his contribution to give the good sound to the poor and not EE people, ok wasn't voluntary but it went this way.

 

[RAYSIMMONS]

Amen ! again

 

[Steve Eddy]

I have to thanks Mr.Sakura for his contribution to give the good sound to the poor and not EE people, ok wasn't voluntary but it went this way.

Mr. Sakura? You mean Mr. Kimura, as in Junji Kimura of 47 Labs? What did he contribute beside a contemptuously high pricetag and marketing doubletalk masquerading as eastern philosophy?

Shouldn't the credit go to the folks at National Semiconductor and other companies which have been producing inexpensive monolithic power opamps for decades now?

One thing I do admire Mr. Kimura for is having the balls to so blatantly thumb his nose at the prevailing status quo and go completely against most of its major tenets.

se

 

[Variac]

He did actually do 2 pretty big things:

1.

One thing I do admire Mr. Kimura for is having the balls to so blatantly thumb his nose at the prevailing status quo and go completely against most of its major tenets.

2. Rocognize how good chip amps can be AND recognize one of the very best chips. It's easy to know this now that he discoved it.
He had the confidence to believe in what he heard

But we should have a plaque made with the names of the 3875 design team and present it to NS

 

[Steve Eddy]

2. Rocognize how good chip amps can be AND recognize one of the very best chips. It's easy to know this now that he discoved it.

That's rather like saying Columbus discovered America, which came as a bit of a surprise to the folks who greeted him on the shore. 🙂

I first started using chip amps over 12 years ago. Long before I'd ever heard of anything called a Gaincard. Does that make me any sort of great discoverer? I don't believe so. Others had used them before me and others have used them since. They've been right under everyone's noses all along.

Only reason they never caught on is because they went against much of what the status quo had been preaching for so long. And they likely never would have caught on the way they did if it weren't for their being legitimized in a product with a contemptuously high price and EJAM (Elfin Japanese Audio Mystic) marketing.

I think instead of making Kimura out as some sort of savior, we should be asking ourselves what else we might be missing out on due to the same prejudices and insecurities which kept many people from even trying chip amps for themselves.

He had the confidence to believe in what he heard

Sure, but what's so special about that? We should all have that same confidence.

But we should have a plaque made with the names of the 3875 design team and present it to NS

Agreed. They're the folks who did the real work.

se

 

[MBK]

I think instead of making Kimura out as some sort of savior, we should be asking ourselves what else we might be missing out on due to the same prejudices and insecurities which kept many people from even trying chip amps for themselves.

se

Switching PSU's driving digital amps

I went a bit the same way as you Steve: built chip amps without awareness of any of the mystical background. Sounded OK. Then read Gaincard review (without mention of chip nature of the thing). Sounded like a gift from heaven, yet unobtainable due to price. Then found out Gaincard uses a chip. Felt confused bec. mine sounded OK but I thought, how come I own already the best of the best of the best SiryesSir, and it merely makes me feel a-OK-lah? [disclaimer, I used OPA548 per Siegfried Linkwitz's topology suggested for LM3886, surely the one and only LM would sound sooo much better than the OPA, if I only chose the right feedback resistor brand, which seems to make or break the LM].

A bit off hand, and aimed at the "politically correct passive component" discussion: I found only 3 things that always made a difference in my circuits - op amp decoupling quality, grounding layout, and oversized transformers for the job.

 

[Steve Eddy]

Switching PSU's driving digital amps

Hehehe. Yes, another good example of how the status quo pendulum eventually swings.

I went a bit the same way as you Steve: built chip amps without awareness of any of the mystical background. Sounded OK. Then read Gaincard review (without mention of chip nature of the thing). Sounded like a gift from heaven, yet unobtainable due to price. Then found out Gaincard uses a chip. Felt confused bec. mine sounded OK but I thought, how come I own already the best of the best of the best SiryesSir, and it merely makes me feel a-OK-lah? [disclaimer, I used OPA548 per Siegfried Linkwitz's topology suggested for LM3886, surely the one and only LM would sound sooo much better than the OPA, if I only chose the right feedback resistor brand, which seems to make or break the LM].

Hehehe. Yeah, I was surprised at how good they sounded given how the status quo looked upon such things as integrated circuits, opamps and global negative feedback.

Then I tried my first hybrid device (an LH-0101) which I felt bettered the monolithics I'd tried by a fair margin. I'm still using the Apex PA-02 and PA-16 hybrids.

Now I'm moving toward higher efficiency on the loudspeaker end and have been experimenting with simpler, low power discrete designs.

A bit off hand, and aimed at the "politically correct passive component" discussion: I found only 3 things that always made a difference in my circuits - op amp decoupling quality, grounding layout, and oversized transformers for the job.

HERETIC! 🙂

se

 

[patriz]

Mr. Sakura? You mean Mr. Kimura, as in Junji Kimura of 47 Labs? What did he contribute beside a contemptuously high pricetag and marketing doubletalk masquerading as eastern philosophy?

Shouldn't the credit go to the folks at National Semiconductor and other companies which have been producing inexpensive monolithic power opamps for decades now?

One thing I do admire Mr. Kimura for is having the balls to so blatantly thumb his nose at the prevailing status quo and go completely against most of its major tenets.

se

Of course you are right but perform a search on this site or AA and you will see that all this power opamp reinassance started after Kimura products were reviewed. You know life is strange,
Altec and WE product were made in Usa but japanese audiophile recognize them as great for domestic audio long before us in the west. It is not enough to make good products it is also important to implement and optimize for specific use and then let people know about of it.

 

[MBK]

Of course you are right but perform a search on this site or AA and you will see that all this power opamp reinassance started after Kimura products were reviewed. You know life is strange,
Altec and WE product were made in Usa but japanese audiophile recognize them as great for domestic audio long before us in the west. It is not enough to make good products it is also important to implement and optimize for specific use and then let people know about of it.

From memory the sequence of events didn't start with 47 Labs promoting chip use. They promoted 9 mm signal path and such. Reviews started gushing about the sound and attributed it to this signal path thing. Someone then "revealed" that 47 Labs used this chip. The public and reviewers then had to either redefine their opinion about chips, or look duped and by extension ridiculous. To me this looks like adaptation of reality to opinion, not the other way round. Had 47 Labs sold the same thing at $200, they would have gotten neither fame, nor good reviews ("nice mid fi"), and the public would still look down on chips.

"Perception is reality", and a high price tag must mean high quality. They did the right thing. NS should sell a new and improved chip with gold plated pins at $100 and the reviews would get better by the day.

Since it still just costs $6 or so, we have to look for other ways to explain the high quality of each individual implementation: Sakura going through 1000's of production chips to find the optimal hand picked one when the moon is at its 4th quadrant in the year of goat cheese, magic feedback resistors that make the sound blossom or collapse like Eco's rose. Et caetera.

In the early days of science people called this "auxiliary hypothesis": when some prestigious man had a wrong or ridiculous idea, the public wouldn't dare criticize it openly but invent an auxiliary hypothesis that gave the claimed idea or fact a hint of possibility (in a parallel universe) to save face to the man of honor.

We will never know whether 47 Labs thusly opened the eyes to the public about the truly excellent sound quality of a mere chip, or whether the high price tag coaxed reviewers into gushing reviews, hereby forcing them and the public to live forever with the auxiliary hypothesis that under the right star and PSU caps made from unobtainium, it actually does do wonders.

Disclaimer: his statement represents my polemic opinion, not fact. 😉

 

[MBK]

Then I tried my first hybrid device (an LH-0101) which I felt bettered the monolithics I'd tried by a fair margin. I'm still using the Apex PA-02 and PA-16 hybrids.

Now I'm moving toward higher efficiency on the loudspeaker end and have been experimenting with simpler, low power discrete designs.

se

Yeah, a high eff speaker would make things easier... my fullrange gets 89 dB eff., I'd prefer 96... And of course the fullrange trades integration and coherence for dispersion problems.

Hybrid devices: I don't know those... you're making me curious...

I had fantasized elsewhere to use parallel BUF134 as amps, OPA637 driven, after I saw the fantastic numbers 😉 PMA obtained in distortion, just for kicks: THE HALC-CLONE , Hahahahahahahaaaaaaaaaaaaarrrrr!!!!

I hereby claim copyright to all variations of the term "HALC-CLONE ", MBK 2003, u heard it here first.

[I have a sick feeling that Halcro likely uses smthg like dat indeed hahaha]

 

[paulb]

Ah, how things change. Back in August 2001, National's chip amps were not well regarded at all on this forum. Here's a sample:

I'm not sure why you think they're "pathetic"(there's a lot of unsubstantiated opinion thrown around these forums).

The 3886 has foldback current limiting.If you read the spec sheet it says the power rating is for an INFINITE heat sink.Since the limiting is also temperature sensitive any real heatsink will cause the protection to activate unless you run the thing at about half power.Foldback current limiting driving an inductive load makes "farting" and "popping" sounds that are hell on your tweeters as well as sounding bad."Pathetic" may not be a strong enough term to describe this part.If you want to buy four of these fine quality parts and parallel and bridge them with current balancing resistors and DC servos for each module go right ahead.In my mind it is both a waste of time and money.

I was involved with the design of an infamous subwoofer amp that used two of these things bridged.We had problems with them poping and farting in the design phase.The company is now broke.