fedde said:
Here's a pic of a Gaincard board
http://www.diyaudio.com/forums/showthread.php?s=&postid=104819#post104819
Attachments
Don't be too critical of the Gaincard construction chaps. I have an inkling into what is going on in there and there isn't really a way to make it pretty.
Let's face it, you don't admire the inside of an amp when you are listening to it. Or at least, you shouldn't prefer a neat and tidy layout to a better sounding, but 'messier' one!
Let's face it, you don't admire the inside of an amp when you are listening to it. Or at least, you shouldn't prefer a neat and tidy layout to a better sounding, but 'messier' one!
carlosfm said:
That's ugly.
It seams like a PCB made in the kitchen.
Where are the PSU caps?
Yep, Peter definitely wins in the naked amplifier beaty contest!
The PSU caps ? On the other side perhaps ??? Have a look at the second photo:
http://www.diyaudio.com/forums/showthread.php?s=&postid=104820#post104820
Fedde
Nuuk said:Don't be too critical of the Gaincard construction chaps. I have an inkling into what is going on in there and there isn't really a way to make it pretty.
Let's face it, you don't admire the inside of an amp when you are listening to it. Or at least, you shouldn't prefer a neat and tidy layout to a better sounding, but 'messier' one!
That is true. Remember that this sort of board has to be assembled by hand, there is no solder mask and silkscreening as it supposedly degrades the sound (and is not necessary for hand assembled board). So this is what you get, a functional piece of electronics.
Konnichiwa,
INDEED.
Now I played with IC Amp's nearly (by now) two decades ago beyond (and before it came down) the big wall of Berlin. And with all respect, well implemented MOST IC based designs are better (objectively and subjectively) than discrete ones. In fact, I had nice discussion with WJ on the topic, when he noticed me supporting the same thesis in the mid 90's on RAHE.
Good thing too. Seeing as it is that "thick wires" usually entail severe compromises. Anyway, if your speakers voicecoil is made from 30m of 28 gauge wire (if you are lucky), what is the issue with a few inch (and in extension a few m in the speaker cable) of 24 or so gauge wire?
If it really mattered to have thick wire, the speaker manufacturers would start winding speaker voice coils with 10 gauge wire. Except, they don't....
Well, "crappy" depends upon your definition of "crappy". To some people "cheap" means crappy. To me personally "badly designed for the task at hand" means crappy. So to me many "High End Audiophile" approved parts are crappy (and I usually take major flack despite relativising remarks of "whatever worsk for you is fine") and many military/industrial grade moderatly expensive parts (not marketed as "audio specific") are "good".
Now, knowing almost to a T what is in the Gaincard I would MOSTLY support Kimura San's choices. In my view the biggest compromise in the GC is the PCB, NOT the "crappy" parts.
Ignoring for the moment that my suggestion of "Invaht da suckah" had some minor input into Daniel San's "gainclone" I agree that Daniel San's amplifier is an exceptional product combining extensive fundamental circuit research and even more extensive implementation research.
I hate calling it that [gainclone] as it is distinctively NOT a clone, but a combination of my old [by now > 20 years] understanding that "Op-Amp" structures [discrete or monolithic] tend to sound better inverted and a program of optimisation in parts, circuit and mechanics that I can only pull my hat to.
Now I would say that without Kimura San (as much as we may criticise his very prosaic design and implementation) I would have not even bothered to dig out my experience with chip amp's (untill Kimura San they where terminally uncool) and without the taking note of many erudite and educated listeners I guess Daniel San would not have taken note.
Now we may note that Sir Robert Stephensons "Rocket" was a boody legless dog of a locomotive (certainly next to a french TVG, Japanese Bullet Train of Chunnel Express). Yet without it I suspect there would be no railway. We may note that my mid 1980's chipamp experiments where exceedingly crude and that Kimura San's Gaincard leaves some to be desired in execution (or not as such may be). But we would not be talking about chipamps without Kimura San.
It's like George Washington won few battles and lost all major ones, but his side won the war. You cannot deny that GW was as military man, in his time a failure. Yet as someone to lead a revolutionary band of patriots, he was just right and NO CONVENTIONAL general could have done anywhere near as well as the Virginia Colinial Troups Coronel who, when a Major surrendered at the first sign of french troups....
Okay? If you have to diss someone, diss Don d'Agostino for ripping of Mark Levinson/John Curl designs and doing it badly....
Sayonara
PS - is the 50W Gaincard worth $ 4,500?
If I had the money and like an hour of my time was paid like a SAP Consultant - then, based strictly on sonic merit and time to build something better - the ****** yes.
As I still earn a lot less than a SAP consultant and CAN do better myself for a fraction - to me personally, right now - NO (that said - Kimura San, never send me anything for review - you may find it difficult to retrieve - I have after all a dining room, kitchen, bedroom and library system apart from the big rig).
grataku said:
INDEED.
Now I played with IC Amp's nearly (by now) two decades ago beyond (and before it came down) the big wall of Berlin. And with all respect, well implemented MOST IC based designs are better (objectively and subjectively) than discrete ones. In fact, I had nice discussion with WJ on the topic, when he noticed me supporting the same thesis in the mid 90's on RAHE.
grataku said:
Good thing too. Seeing as it is that "thick wires" usually entail severe compromises. Anyway, if your speakers voicecoil is made from 30m of 28 gauge wire (if you are lucky), what is the issue with a few inch (and in extension a few m in the speaker cable) of 24 or so gauge wire?
If it really mattered to have thick wire, the speaker manufacturers would start winding speaker voice coils with 10 gauge wire. Except, they don't....
grataku said:
Well, "crappy" depends upon your definition of "crappy". To some people "cheap" means crappy. To me personally "badly designed for the task at hand" means crappy. So to me many "High End Audiophile" approved parts are crappy (and I usually take major flack despite relativising remarks of "whatever worsk for you is fine") and many military/industrial grade moderatly expensive parts (not marketed as "audio specific") are "good".
Now, knowing almost to a T what is in the Gaincard I would MOSTLY support Kimura San's choices. In my view the biggest compromise in the GC is the PCB, NOT the "crappy" parts.
grataku said:
Ignoring for the moment that my suggestion of "Invaht da suckah" had some minor input into Daniel San's "gainclone" I agree that Daniel San's amplifier is an exceptional product combining extensive fundamental circuit research and even more extensive implementation research.
I hate calling it that [gainclone] as it is distinctively NOT a clone, but a combination of my old [by now > 20 years] understanding that "Op-Amp" structures [discrete or monolithic] tend to sound better inverted and a program of optimisation in parts, circuit and mechanics that I can only pull my hat to.
Now I would say that without Kimura San (as much as we may criticise his very prosaic design and implementation) I would have not even bothered to dig out my experience with chip amp's (untill Kimura San they where terminally uncool) and without the taking note of many erudite and educated listeners I guess Daniel San would not have taken note.
Now we may note that Sir Robert Stephensons "Rocket" was a boody legless dog of a locomotive (certainly next to a french TVG, Japanese Bullet Train of Chunnel Express). Yet without it I suspect there would be no railway. We may note that my mid 1980's chipamp experiments where exceedingly crude and that Kimura San's Gaincard leaves some to be desired in execution (or not as such may be). But we would not be talking about chipamps without Kimura San.
It's like George Washington won few battles and lost all major ones, but his side won the war. You cannot deny that GW was as military man, in his time a failure. Yet as someone to lead a revolutionary band of patriots, he was just right and NO CONVENTIONAL general could have done anywhere near as well as the Virginia Colinial Troups Coronel who, when a Major surrendered at the first sign of french troups....
Okay? If you have to diss someone, diss Don d'Agostino for ripping of Mark Levinson/John Curl designs and doing it badly....
Sayonara
PS - is the 50W Gaincard worth $ 4,500?
If I had the money and like an hour of my time was paid like a SAP Consultant - then, based strictly on sonic merit and time to build something better - the ****** yes.
As I still earn a lot less than a SAP consultant and CAN do better myself for a fraction - to me personally, right now - NO (that said - Kimura San, never send me anything for review - you may find it difficult to retrieve - I have after all a dining room, kitchen, bedroom and library system apart from the big rig).
Konnichiwa Daniel San,
Well, dunno 'bout Kimura San, but the biggest Amp I ever build was a Monster "Borg Cube" inspired by the Crown Macrotech 10,000 (your guess on the meaning of the number is accurate).
It had tree mains transformers, all with a primary fuse of 16 A slow blow (resettable) at 220V mains and used normally 380V 3-phase power (around 10KW nominal transformer rating and as Class B amp 70% efficiency, so around 7,000W max AES RMS output when connect to a suitable mains supply and bucketloads of momentary power).
The whole thing was a simple circuit, industrial, self protecting 300A (continous > 1,00A momentary) darlington modules, a huge pair of forced cooling heatsink tunnels and fans making a comfortable noise like a jet turbine (when on full power) close up.
The circuit was a 120V rails (nearly 80V RMS output into << 1 Ohm) circlotron with a video output transistor differential cascode (Hedge Circuit) running on +300V with some overall differential loop feedback and a differential input/predriver stage running open loop.
The worst punishment we ever meted out to those cubes was to hang 32 pcs 18" W bins of each bass amp. The worst the amp's ever did to speaker was to blow clean out on a single drum kick without limiter 16pcs of said 18" W Bins, equipped with EVM18B Pro Line (400W AES, 800W Program, 1,600W peak - per driver).
Howzdat for a warstory?
REAL AUDIO WEENIE AMP's....
BTW, I still think your gainclone rocks. Power is deceptive....
Sayonara
Peter Daniel said:
It shouldn't be surprising, as I started with real man amps, with lots of heatsinking, big transformers and big filter caps. Who knows if the creator of Gaincard ever built a true class A amp?
Well, dunno 'bout Kimura San, but the biggest Amp I ever build was a Monster "Borg Cube" inspired by the Crown Macrotech 10,000 (your guess on the meaning of the number is accurate).
It had tree mains transformers, all with a primary fuse of 16 A slow blow (resettable) at 220V mains and used normally 380V 3-phase power (around 10KW nominal transformer rating and as Class B amp 70% efficiency, so around 7,000W max AES RMS output when connect to a suitable mains supply and bucketloads of momentary power).
The whole thing was a simple circuit, industrial, self protecting 300A (continous > 1,00A momentary) darlington modules, a huge pair of forced cooling heatsink tunnels and fans making a comfortable noise like a jet turbine (when on full power) close up.
The circuit was a 120V rails (nearly 80V RMS output into << 1 Ohm) circlotron with a video output transistor differential cascode (Hedge Circuit) running on +300V with some overall differential loop feedback and a differential input/predriver stage running open loop.
The worst punishment we ever meted out to those cubes was to hang 32 pcs 18" W bins of each bass amp. The worst the amp's ever did to speaker was to blow clean out on a single drum kick without limiter 16pcs of said 18" W Bins, equipped with EVM18B Pro Line (400W AES, 800W Program, 1,600W peak - per driver).
Howzdat for a warstory?
REAL AUDIO WEENIE AMP's....
BTW, I still think your gainclone rocks. Power is deceptive....
Sayonara
Kuei Yang Wang said:
while I agree with you on the rest of your post (especially the point on good industrial / military parts), I have trouble understanding why and how the above is true. There is no reasons, known to me anyway, that an inverting topology is electrocially superior to the non-inverting design, not to mention the highly subjective "sounding better" part.
Afterall, most amps are implemented as non-inverting amps.
Konnichiwa,
Hmmm. Have you ever taken any funloving discrete or monolithic Op-Amp using a long tailed pair input and have you measured both the spectrum and absolute values of distortion, using identical cloed loop gains, loads inad voltages? If you did you would have the spectrum less complex (less higher harmonics) and the distortion level overall 4-6db lower at least (the 4-6db are for "best" circuits).
This "common mode" effect has actually been extensively documented by a number of extremist "objectivist" writers, especially Douglas "Burn the heretics who hear audible differences" Self.
So. I remember a time when most rifles where implemented as muzzle loaders.... And before that some people thought crossbows the latest fashion....
Simply because a given approach is popularily accepted does not mean it's "right"....
But hell, how about instead of talking smart you took 4 pcs LM3875 and build both the stripped down "inverted" and a "noninverted" gainclone, using the exact same (external) PSU for both and taking care to use the same parts in the same position for the different circuits. Then you may measure, listen, have other listen, issue a challenge for goldenears to hear the difference or simply have luch with Kelly Rowland or Samantha Mumba.
At least you will after that have a certain angle on empirical work. It seems these days everyone thinks all the empirical work has already been done and we can reach conclusions by simply using orthodox formulas and approaches....
We (humankind) would have never developed nuclear devices, flight heavier than air (powered or not), ICBM's, microwaves, mobile phones, CD's and so may other modern conveniences....
I do not know if the orthodoxy would have been the better choice. I for one would as heretic enjoy the close attentions of the DominiCanis carrying out the Holy Inquisition if it had, so I am for progress (of sorts - rather Mc Carthy than Torquemada....).
Sayonara
millwood said:
Hmmm. Have you ever taken any funloving discrete or monolithic Op-Amp using a long tailed pair input and have you measured both the spectrum and absolute values of distortion, using identical cloed loop gains, loads inad voltages? If you did you would have the spectrum less complex (less higher harmonics) and the distortion level overall 4-6db lower at least (the 4-6db are for "best" circuits).
This "common mode" effect has actually been extensively documented by a number of extremist "objectivist" writers, especially Douglas "Burn the heretics who hear audible differences" Self.
millwood said:
So. I remember a time when most rifles where implemented as muzzle loaders.... And before that some people thought crossbows the latest fashion....
Simply because a given approach is popularily accepted does not mean it's "right"....
But hell, how about instead of talking smart you took 4 pcs LM3875 and build both the stripped down "inverted" and a "noninverted" gainclone, using the exact same (external) PSU for both and taking care to use the same parts in the same position for the different circuits. Then you may measure, listen, have other listen, issue a challenge for goldenears to hear the difference or simply have luch with Kelly Rowland or Samantha Mumba.
At least you will after that have a certain angle on empirical work. It seems these days everyone thinks all the empirical work has already been done and we can reach conclusions by simply using orthodox formulas and approaches....
We (humankind) would have never developed nuclear devices, flight heavier than air (powered or not), ICBM's, microwaves, mobile phones, CD's and so may other modern conveniences....
I do not know if the orthodoxy would have been the better choice. I for one would as heretic enjoy the close attentions of the DominiCanis carrying out the Holy Inquisition if it had, so I am for progress (of sorts - rather Mc Carthy than Torquemada....).
Sayonara
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.