|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
Hi,I got the tweak bug and opened up my 4700s (50 watt) Power Humpty. I'm interested in trying some FREDs of HexFREDs in the circuit, so was particularly interested in finding out what the diodes are.
Inside the Humpty is a very dense and high quality transformer (not a toroidal type). There are two secondary windings, both rated at 23.5V 3.6A. There are 4 diodes which appear to be configured in a standard bridge.
The diodes are labelled FE5D 9808 and are packaged in the grey, round, ballish shaped package with the black cathode band at one end (I forgot what that package is called and am too lazy to look it up). Does anyone have any info and ratings on these diodes? I tried a preliminary cross-reference on the web but had had no luck.
With the discussions on the internet about Gainclones, I surmise that I could use some IR Hexfreds rated at 600v 8A and have plenty margin remaining, but I would like to get the low down on these stock diodes first.
Also, I've read on some posts that the IR Hexfreds are better sounding on higher voltage tube applications (they tend to sound soft at some parts of the frequency range at low voltages), and that the Harris freds are better for lower voltage SS applications. Does anyone have any opinions on this? If this is the case, the Harris ones might be the ticket.
Just to retierate: this is the 50 watt "S" version Humpty, not the 25 watt version.
Thanks in advance,
K.T.
Follow Ups:
Interesting. Thanks alot for posting this into. It is much appreciated.-Richard
A lot of people are interested in what is inside the gaincards and the power humpty. Perhaps you could tell us some info?
These are good diodes by general semiconductor,
the Hexfreds are a bit better, I bet you won't hear any
improvement...would you please tell us what is the power OPAMP
in the Gaincard itself ?Thanks!
Ozz,Thanks for the info on the diodes. I haven't opened up the Gaincard itself, just the Power Humpty, so I don't know which Op amps are in there.
I imagine I'll open it up at some point as I'm interested to see if the filtering capacitors can be improved (I'm also interested in seeing if the C37 lacquer will be of benefit in this amplifier).
I'll post the info once I get inside. Don't hold your breath, though. It might be a little while before I get around to it. I've got too many other things to take care of....you know how it goes.
Thanks again for the info,
K.T.
Hi!I think, no reasons to replace that diodes, they are not bad at all.
It's Glass Passivated Fast Efficient Rectifiers (35nS) by General Semiconductor.
Full info: http://www.chipinfo.ru/docs/GE/000162.pdfCould you tell, are there any capasitors (and what capasitors) in Power Humpty?
Are there only one rectifier bridge or two bridges?Thank you in advance
Alex.
Alex,Oh, by the way, the Power Humpty has only one bridge rectifier which both stereo channels draw from.
Thanks,
K.T.
Alex,Thanks for the information. As the diodes seem to be better than the standard 10 cent units and seem similar to the fast recovery diodes, I think I'll just keep them in there.
The only components inside the Power Humpty are: the transformer, the circuit board onto which the diodes are soldered, the main wiring, and some miscellaneous wires that powers the little LED on the front panel. That's it. I assume the capacitors are inside the Gaincard unit itself.
As far as the Humpty goes, there doesn't seem to be anything innovative going on in there. I have to say, though, that the transformer seems to be a very good quality one.
The core of the transformer isn't comprised of stacked metal plates, as in most transformers. Rather, it seems to be made of one continuous ribbon of metal which is wound together to form a sort of oval donut (the ribbon is wound together very tightly - at first glance, I initially thought that this core was a solid metal ring). This donut is fairly heavily lacquered (very smooth and glassy looking) and seems to be held together by some clamping bands. The coils are wound around this. The wire for the coils seems to be of a pretty heavy gauge. Again, there are two secondary windings. The coils are wrapped in a heavy craft paper.
The transformer is not a torroid, as some have surmised.
Even though the power supply is quite simple, perhaps the quality of the transformer contributes much to the performance of the amplifier (?).
Thanks,
K.T.
http://www.custommag.com/CMIHOME.HTG/rcore1.htm
Chris,No, the Gaincard transformer is much more massive and substantial. The shape of the core is similar, but the Gaincard core seems to be made out of a flat ribbon of metal wound up to make the donut shape. The cross-section is, therefore, square and not round. Also the GC core is much wider. If you took one of the pictured cores and laid it on its side, then laid another one on top of it, that would be more like the relative dimensions of the GC core.
Because the core is wider, the coils on the GC form a big, blockish mass. The windings are covered in craft paper. The transformers in the photos look flat. In the GC, the coils forms more of a block (equal in dimensions around the core, just a little longer along the length of the core).
Also, the GC transformer has it's mounting plate at the short, round end of the core. This is then bolted to the front plate of the Power Humpty. In effect, the transformer is being suspended in air along the length of the Humpty by the front plate. The wires then come from the other end of the transformer and attach to the connectors on the back plate of the Humpty.
The GC transformer is much more hefty looking. If you hold the 4700 "s" Power Humpty, you'll see that it's quite heavy. Just about all of that weight is the transformer. It's quite impressive.
Hope this helps,
K.T.
1. If you check the 47 labs websitewww.sakurasystems.com
you'll see they list the power transformer as being a "C-core". Nobody claimed that it was a torroid, but a torroid is suggested as the best readily available substitute given R and C core transformers are expensive and hard to get in small quantities.
Nothing innovative? Deceptive. The innovation is justly the simplicity, and the extremely high quality, custom transformer, the very good diodes, and the ommission of the filter caps.
In my opinion - and this is just my own feeling after spending a lot of time researching the gaincard - is that trying to improve the real Gaincard by tweaking is something of a lost cause. You paid so much for it in part to allow Kimura to do all that for you. The gaincard is a fully optimised design. No stone was left unturned, as the cliche goes. Play with it and you are almost sure to fall off the curve.
Richard
PS. Only one diode bridge (4 diodes) means three power supply wires. I heard there are four wires (times two) which go to the amplifier. Can you confirm that the fourth wire is a duplicate ground, or earth ground?
PS. Only one diode bridge (4 diodes) means three power supply wires. I heard there are four wires (times two) which go to the amplifier. Can you confirm that the fourth wire is a duplicate ground, or earth ground?My guess would be that two of the wires are the + and - off the rectifier and the other two are the other ends of two secondary windings brought separately to the amplifier chassis and tied together there rather than in the power supply.
se
Richard,Yes, I agree that the outstanding feature of the Gaincard is not only the brilliantly simple design, but also the quality of the components. I can't agree with you more.
As others have indicated that the diode choice is in fact fully considered, I've decided to leave them as is.
As for the wiring scheme, each of the connectors does have four pins in it. I'll trace the wires to find out, but I'll be out of town for a few days and won't be able to do so until Wednesday. I'll post my findings then.
K.T.
Yes, I agree that the outstanding feature of the Gaincard is not only the brilliantly simple design...While I've nothing against the Gaincard, I've yet to understand what's so "brilliantly simple" about it. Or even just plain simple for that matter. Other than perhaps eschewing bypass caps, what exactly makes it particularly simple in any unusual way?
...but also the quality of the components.
Like these? :)
se
I'm happy to see that K.T. took it very maturely. In any event, if you study the circuit board carefully, you will find that it really is a very well thought out design, with minimum length sigal paths and intelligent positioning of components. The longer you look at it, the more you find yourself respecting it. The only people who laugh in derision at the carbon resistors etc. are those who have heard neither Gaincard nor Gainclone. Their loss.Build one. Try metal film resistors, try every other resistor, then try carbon film. Say: "oh."
Simon Yorke was once admonished by MF for using a plywood armboard on a very expensive turntable. He replied that plywood sounded best, and that he'd tried everything else. Would you then claim the table wasn't worth the money because it uses a plywood armboard? Au contraire!
Your point about the oft-quoted simplicity conveniently overlooking the internal circuitry of the op amp is well taken, but nonetheless it remains, within opamp circuits, an extremely simple and compact implementation.
The "brilliantly simple" is deserved because, like all great ideas, it is something that anyone could have thought of but nobody did.
Price vs. value, price vs. performance is something best left to those who propose to hear and see it for themselves.
cheers,
rjm
I'm happy to see that K.T. took it very maturely. In any event, if you study the circuit board carefully, you will find that it really is a very well thought out design, with minimum length sigal paths and intelligent positioning of components.Uh, that should describe ANY reasonably competent layout.
The longer you look at it, the more you find yourself respecting it.
I respect it in its being a competent layout. However I don't see anything particularly clever about it that warrants any more respect than any other competent layout. Unless one is totally incompetent, one whould have to intentionally try to come up with a less compact layout.
The only people who laugh in derision at the carbon resistors etc. are those who have heard neither Gaincard nor Gainclone. Their loss.
For the record, I didn't laugh in derision at the carbon film resistors. The smiley face was a reflection of my amusement at how the status quo pendulum swings in this hobby. Yesterday's bog standard parts become today's high quality parts. Everyone goes chasing after parts of the highest objective quality and then someone comes along and says "Hey, try these. They're not bad." And suddenly everyone's chasing after what they were previously running away from.
Personally I've never found any consistent correlation between the objective specs of a particular part or component and what I ultimately prefer to listen to. To that end I give everything a fair chance and just go with what works for me in the end regardless of what the prevailing status quo may have to say. And that's why you won't see me laughing derisively at someone else's choice of parts.
Build one. Try metal film resistors, try every other resistor, then try carbon film. Say: "oh."
Been there, done that, and got the t-shirt over 10 years ago. Wound up using Caddocks in the end.
Simon Yorke was once admonished by MF for using a plywood armboard on a very expensive turntable. He replied that plywood sounded best, and that he'd tried everything else. Would you then claim the table wasn't worth the money because it uses a plywood armboard? Au contraire!
Spare hair is fair.
In fact, hair can be rare.
Fred Astaire got no hair.
Nor a chair.
Nor a chocolate eclair.
And where is the hair on a pear?
Nowhere, mon frere! .Oops. Sorry. Your au contraire caused me to momentarily lapse into a George Carlin flashback. I knew I should have left those mushrooms on the cow pat. :)
Anyway, I think perhaps you misunderstood my intent with regard to pricing. All I'm saying is that if you use a piece of plywood, it should be factored into the retail price as a piece of plywood. Not as something vastly more expensive than a piece of plywood. So, if you're using 2¢ resistors, they should factor in as 2¢ resistors, not $5 resistors.
While I'm very much a capitalist and believe manufacturers have every right to put whatever price tag they want on their products, I also have the right to decide whether or not I think the price is worth it. And one of the primary deciding factors for me is whether the product's price bears what I think is a reasonable relationship to the cost of producing it.
And when I look at the Gaincard, I see a $1,500 product, tops. So to me, I don't consider the Gaincard's $3,300 price tag to bear a reasonable relationship to its cost. But that's just my personal outlook on things. I don't expect it to be anyone else's. I only offer it for consideration.
The "brilliantly simple" is deserved because, like all great ideas, it is something that anyone could have thought of but nobody did.
Not to brag, but I thought of it and did it over 10 years ago. I was looking to design a power amplifier to mate with a buffered attenuator "preamp" I was manufacturing at the time:
I wanted the amplifier to fit in the same size chassis as the preamp which was rather atypically small at the time at 12" x 12". So I started experimenting with monolithic power opamps and found that they didn't sound as bad as one would be led to believe.
However about the same time the recession hit and I got hosed by a major part supplier and decided to terminate the business so I never got 'round to producing a commercial version, but my personal amps have been based on simple power opamp circuits ever since except that I've been using hybrids rather than monolithics.
And I don't consider this bragging because I don't consider the idea to be particularly brilliant. In fact I consider it to be rather a no-brainer, just as the idea to strap a PMI BUF-03 on the output of an attenuator and create a buffered "passive" preamp (the gizmo above) even though that hadn't been done before to my knowledge.
Now, what I DO admire Kimura for is his balls. The balls to offer a commercial product that thumbed its nose at the status quo which had for some time declared that such things as integrated circuits, opamps, negative feedback, carbon film resistors, lack of bypass caps, etc. were all bad and should be avoided at all costs.
And I admire those didn't let prejudice get the best of them and gave the Gaincard a fair chance in spite of the direction the winds of the prevailing status quo were blowing.
So I'm not without some admiration. I just don't see the Gaincard as being particularly innovative (though I do think the OTA cabling system's pretty clever), I think it's quite overpriced, and I think their marketing is a bit misleading.
Price vs. value, price vs. performance is something best left to those who propose to hear and see it for themselves.
Nonsense. My criteria of a reasonable relationship to cost of production is no less valid than the criteria of sonic performance. It's an individual choice and therefore best left to each individual to determine for themselves. I don't see why I should keep my opinions to myself regarding price vs. value just because my criteria differs from yours or those of others.
se
Steve,So there weren't ANY caps in the Gaincard? There sure weren't any in the Humpty.
What happened to the fabled 1000uF caps?! :)
Once again, thanks for posting the picture - made my day.
Best,
K.T.
Hi Steve,Hmmm... Interesting. Did you already take your Gaincard apart?
The case on the right looks a lot like the Gaincard. Also the red and blue rubber sleeves towards the top of your photo look a lot like the speaker binding wires on the Gaincard...and the circuit board looks quite a bit like the "hand-etched" circuit board that the Power Humpty diodes were mounted on. It looks pretty crude in there, doesn't it?
Well, whatever they do, it sure makes for a good sounding amp (IMO).
Any info on the chip, or the resistors? Do you have a photo of the other side of the board?
Very intersting. Thanks for posting this.
Best,
K.T.
Hmmm... Interesting. Did you already take your Gaincard apart?No, I don't own one.
The photos were provided by another inmate, cfraser who was given the photos by someone who had taken theirs apart. The photos were already published here on AA in General Asylum back in February. Yoshi Segoshi of Sakura Systems responded claiming that the photos were not of a 47 Labs Gaincard and when asked to provide photos of the internals of a real Gaincard, he refused.
However I showed the photos to someone else I know who has cracked open a Gaincard and he verified that the photos showed what he had seen in the Gaincard he took apart. That along with your saying that the circuit board in the Humpty is of the same "hand etched" type pretty much tells me that Segoshi was being less than truthful in an attempt at damage control (the photos sparked some amount of ridicule).
Here's the thread from General Asylum:
http://www.audioasylum.com/audio/general/messages/172637.html
The case on the right looks a lot like the Gaincard. Also the red and blue rubber sleeves towards the top of your photo look a lot like the speaker binding wires on the Gaincard...and the circuit board looks quite a bit like the "hand-etched" circuit board that the Power Humpty diodes were mounted on. It looks pretty crude in there, doesn't it?
Yup. Not something one expects to see in a product with an asking price over $3,000 (the photos were of the 25 watt Gaincard by the way). Personally I'd always thought that the Gaincard was rather grossly overpriced. When I saw the photos, I just had to shake my head.
Well, whatever they do, it sure makes for a good sounding amp (IMO).
I agree with you there. I played around with monolithic power opamps back in the early 90s and was surprised then at how good they can sound. I then tried some hybrid devices and found them to be better still.
Any info on the chip, or the resistors?
The chip is a National LM-3875. Don't know anything about the resistors other than the appear to be garden variety carbon films.
Do you have a photo of the other side of the board?
Sure. Here you go:
se
Steve,Thanks for the additional photo.
Hmmm...I don't know how to feel now. On one hand, the Gaincard is very cool - I think it has a great form factor, and it sounds really good. Looking inside, though, I can't help but feel a little ripped off. I say "a little" because I still think the Gaincard is cool in form, and it sounds great, but it just looks so cheesy on the inside.
When you hear that it has only nine parts in the signal path, that sounds really wonderful. In fact, that simplicity has real appeal to it. Then when you see it, it hits you and you realize just how simple it is. In fact, it's so simple that if I didn't have one, and knew what I know now, I'd go ahead and build one from scratch myself! In fact, I might just do that anyway.
I supposed if the fit and finish on the inside were as clean and neat at the case design I would feel fine. I just don't like it when my construction sensibilities are better than a manufacturer's.
Anyhow, the point is moot. I already have the Gaincard and enjoy it greatly, and now I have enough information to make a decent clone if I wanted. I guess that's all I can ask for at this point.
One thing though: I'm still very impressed with the power transformer. I would really like to find one of this quality.
Also, I'm itching to experiment with new resistors and capacitors. I've alway liked the Shinkoh tantalum film resistors but don't think they make them this small. The smallest tantalums that I've seen are 1/2 watts (Shinkoh and Audio Note) and I know these are quite a bit bigger than the ones pictured. Also, I wonder if these would be too detailed for a crisp and clean sounding design like the GC. Perhaps the Riken Ohm carbons films (which are smaller) or the Kiwame would be a better choice. Also, I wonder if Black Gates or Cerefine would make a good cap replacement.
About your experiments with the hybrid devices: do you have any links or leads to good designs using these?
Thanks,
K.T.
Hmmm...I don't know how to feel now. On one hand, the Gaincard is very cool - I think it has a great form factor, and it sounds really good. Looking inside, though, I can't help but feel a little ripped off. I say "a little" because I still think the Gaincard is cool in form, and it sounds great, but it just looks so cheesy on the inside.Understandable.
When you hear that it has only nine parts in the signal path, that sounds really wonderful. In fact, that simplicity has real appeal to it. Then when you see it, it hits you and you realize just how simple it is. In fact, it's so simple that if I didn't have one, and knew what I know now, I'd go ahead and build one from scratch myself! In fact, I might just do that anyway.
Actually they don't claim it has only nine parts in the signal path. Just nine parts in the basic circuit. And I've been rather critical of that nine parts claim. I think it's highly misleading when you consider that one of those parts is the power opamp itself which ultimately comprises literally dozens of parts and a circuit topology at least as complex if not moreso than many discrete amplifier designs.
So it's really nowhere near as simple as they'd have people believe. The complexity is rather out of sight and out of mind. If you want REAL simplicity, then you need to look at something like a SET or say Nelson's Son of Zen.
Anyhow, the point is moot. I already have the Gaincard and enjoy it greatly, and now I have enough information to make a decent clone if I wanted. I guess that's all I can ask for at this point.
Absolutely. Ultimate enjoyment is all that matters in my opinion.
One thing though: I'm still very impressed with the power transformer. I would really like to find one of this quality.
I'm still trying to figure it out. What you describe sounds like the core is made like that of a toroid (i.e. a coiled ribbon) but has windings more akin to that of an EI core. Although I'm giving up on AC entirely, I'm still rather curious about the transformer you're desribing.
Also, I'm itching to experiment with new resistors and capacitors. I've alway liked the Shinkoh tantalum film resistors but don't think they make them this small. The smallest tantalums that I've seen are 1/2 watts (Shinkoh and Audio Note) and I know these are quite a bit bigger than the ones pictured. Also, I wonder if these would be too detailed for a crisp and clean sounding design like the GC. Perhaps the Riken Ohm carbons films (which are smaller) or the Kiwame would be a better choice. Also, I wonder if Black Gates or Cerefine would make a good cap replacement.
That's an area I stay well away from as I believe you're the only one who can provide a truly meaningful answer for yourself.
About your experiments with the hybrid devices: do you have any links or leads to good designs using these?
You know I haven't come across anyone, commercial or DIY that's using hybrids. As to why that is I can only guess. One is that hardly anyone is aware of them and another is that they're very expensive relatively speaking. The ones I'm currently using are the Apex PA-16s which go for a little over $90 a pop.
The circuit I'm using in my current amps is about as simple as it gets. Just a Jensen JT-11P-1 input transformer, a 10k load resistor across its secondary, a pair of resistors for the feedback network's voltage divider and a pair of current sense resistors (part of the PA-16's design). That's it.
se
Steve,Thanks for the info on the Apex chip. I'm intrigued.
Nice construction on the passive preamp. Now that you've worked with the hybrids, what's your current opinion on those BUF-03s? I remember they were all the rage when Stereophile published that Corey Greenberg DIY article in '92. I got most of the parts but never did get around to building it.
Best,
K.T.
Nice construction on the passive preampThanks. :)
Now that you've worked with the hybrids, what's your current opinion on those BUF-03s?
Well the two are really rather apples and oranges. The BUF-03 (or any of the various buffer chips) is best suited for impedance buffering, offering a high input impedance and a low output impedance which is ideal for converting the relatively high and variable output impedance of an attenuator into something more suitable for driving cable.
The buffer chips have no voltage gain and limited current gain so you can't quite compare them to a power opamp which gives you voltage gain and a considerable amount of current gain and is intended for driving low impedance loads.
I haven't had anything in my system which uses a BUF-03 for about 6 years and haven't tried any of the newer buffer chips (the PMI BUF-03 is no longer in production as far as I'm aware) so I'm afraid I can't offer much in the way of comparison there. Sorry.
I remember they were all the rage when Stereophile published that Corey Greenberg DIY article in '92.
Hehehe. Yeah. It was the AL-1 that Corey based his DIY version on. This was back when we were all on The Audiophile Network BBS (it was basically the pre-Internet version of the Asylum). Corey saw the press release for the AL-1 in the file area on TAN and asked if I'd help him do a DIY article based on it for Stereophile.
So I gave him the schematics, parts sources, etc. and the rest, they say, his history. :)
Corey's version was actually quite good, improving on the AL-1 with the addition of a tweak or two that Walt Jung contributed. So when I see the kits out there for Corey's version as well as kits that are improvements on Corey's version, I can't help letting my modesty slide a little and get a bit of a prideful smile on my face knowing where it started. :)
I got most of the parts but never did get around to building it.
Hehehe. Know how that goes. But the remnants of all those ill-fated projects can sure come in handy when you're in a pinch. :)
se
-
This post is made possible by the generous support of people like you and our sponsors: