Hi Ragil,
Someone posted a schematic at the start of this thread somewhere, but I'm not convinced it is an accurate schematic. I do not know where you can find any others.
I remember reading an article in HFNRR 1985(?) - an interview with Dan D'Agostino. He revealled that the KSA50 was a low feedback design - flat 14dB gain margin out to >70kHz with about 50kHz input resistance. D'Ag claimed the design could be run in open loop but that they added some negative feedback to improve stability. If you check out the Krell website you'll find some more info. Like all great designs the Krell probably looks very simple on paper and like all great designs there is a lot more to it that the schematic doesn't show.
The fact that Krell uses a fully balanced, class A design supports the low feedback idea. Their basic approach is to make the circuit as linear as possible. Using double input LTPs will reduce distortion and improve PSRR if (and only if) the two halves behave identically. Getting this right is hard and IMO is not an easy thing to grapple with if you are building your first amp. But the idea of having a flat OL response and having low or no feedback helps to achieve a good sound fairly easily.
Someone posted a schematic at the start of this thread somewhere, but I'm not convinced it is an accurate schematic. I do not know where you can find any others.
I remember reading an article in HFNRR 1985(?) - an interview with Dan D'Agostino. He revealled that the KSA50 was a low feedback design - flat 14dB gain margin out to >70kHz with about 50kHz input resistance. D'Ag claimed the design could be run in open loop but that they added some negative feedback to improve stability. If you check out the Krell website you'll find some more info. Like all great designs the Krell probably looks very simple on paper and like all great designs there is a lot more to it that the schematic doesn't show.
The fact that Krell uses a fully balanced, class A design supports the low feedback idea. Their basic approach is to make the circuit as linear as possible. Using double input LTPs will reduce distortion and improve PSRR if (and only if) the two halves behave identically. Getting this right is hard and IMO is not an easy thing to grapple with if you are building your first amp. But the idea of having a flat OL response and having low or no feedback helps to achieve a good sound fairly easily.
Hey Nelson,
Krell being one of you primary competitors (I assume), I assume you've you have done considerably more research than I have. Can you confirm that the posted schematic is accurate?
My comment about the schematic not telling all is an indirect way of saying that the way the transistors are chosen (beyond just part numbers) and the way the performance is measured is not shown. In my experience this is crucial. I do not believe that someone can build a Krell based on the schematic and have it sound anything like as good as Krell build them without a lot of prior experience.
I presume this would be the same for someone building a Pass amp without the benefit of this forum. Is it not the knowledge of part choice (like Vgs threshold matching) and wiring and psu design and many other things that makes an X1000 sound as good as it costs? Which means it must sound very good indeed.
Krell being one of you primary competitors (I assume), I assume you've you have done considerably more research than I have. Can you confirm that the posted schematic is accurate?
My comment about the schematic not telling all is an indirect way of saying that the way the transistors are chosen (beyond just part numbers) and the way the performance is measured is not shown. In my experience this is crucial. I do not believe that someone can build a Krell based on the schematic and have it sound anything like as good as Krell build them without a lot of prior experience.
I presume this would be the same for someone building a Pass amp without the benefit of this forum. Is it not the knowledge of part choice (like Vgs threshold matching) and wiring and psu design and many other things that makes an X1000 sound as good as it costs? Which means it must sound very good indeed.
Schematic Krell
Hi Traderbam,
The secret is in the schematic and in the matching of the transistors; especially the inputtransistors.
I own a Gas Ampzilla. The Son of Ampzilla has roughly the same topology as the Krell KSA-100 and that is a very good sounding amplifier too.
Denis I don't consider the complementary input differential stage stupid. To get the best sound the four transistors have to be matched all four. And the electrolytic in the feedback path to get a 1x gain at DC preventing offset gives a much better sound most notably in the bass than a DC-servo with a IC. You can improve it by bypassing the electrolytic with a polypropylene capacitor. Interestingly some IC's have the same schematic as some poweramplifiers.
I can build a Krell, Ampzilla or Son of Ampzilla that sounds better than the original.
Hi Traderbam,
The secret is in the schematic and in the matching of the transistors; especially the inputtransistors.
I own a Gas Ampzilla. The Son of Ampzilla has roughly the same topology as the Krell KSA-100 and that is a very good sounding amplifier too.
Denis I don't consider the complementary input differential stage stupid. To get the best sound the four transistors have to be matched all four. And the electrolytic in the feedback path to get a 1x gain at DC preventing offset gives a much better sound most notably in the bass than a DC-servo with a IC. You can improve it by bypassing the electrolytic with a polypropylene capacitor. Interestingly some IC's have the same schematic as some poweramplifiers.
I can build a Krell, Ampzilla or Son of Ampzilla that sounds better than the original.
At the risk of shattering any illusions, anyone who
does a workmanlike job assembling these amplifiers
from schematics will get much the same results.
Matching input devices is nice, but depending on what
you are looking for you can get even better results by
substituting devices into a live circuit and evaluating them
in situ. This is a tedious process, but is truly the one that
milks the last bit of performance out of the circuit.
The simpler the circuit, the better this works.
The KSA50 is a nice clean simple Class A amplifier, and that's
why I regard it as the best of the Krells. Don't bother to
flame me if you disagree....
does a workmanlike job assembling these amplifiers
from schematics will get much the same results.
Matching input devices is nice, but depending on what
you are looking for you can get even better results by
substituting devices into a live circuit and evaluating them
in situ. This is a tedious process, but is truly the one that
milks the last bit of performance out of the circuit.
The simpler the circuit, the better this works.
The KSA50 is a nice clean simple Class A amplifier, and that's
why I regard it as the best of the Krells. Don't bother to
flame me if you disagree....
Yes Nelson..
Im with you...some people think that amplifiers work in a magical way...but they work in a cientific way.
If anibody make a amplifier with the same schematic,the same parts...and the some grounding tecnics...it will end up with a similar sounding amp...be it made by you.by me or by Krell!
Audio is engenearing...not mistic or black magic!
Cheers
Jorge Santos
Im with you...some people think that amplifiers work in a magical way...but they work in a cientific way.
If anibody make a amplifier with the same schematic,the same parts...and the some grounding tecnics...it will end up with a similar sounding amp...be it made by you.by me or by Krell!
Audio is engenearing...not mistic or black magic!
Cheers
Jorge Santos
traderbam said:
Thanks Traderbam,
But that's KSA-100 ... I want to see the KSA-50 schematic, that sound better than KSA-100 ...
I agree with Nelson:
Nelson Pass said:
My interest as DIYer's is how to maximize the minimalist ...... I never clone the schematic as is, I always try to change the transistor and the value of the resistor or capacitor as I calculate myself or take suggestion from this forum ...
Thanks all ...
... & Regards:
ragil.hastomo
Jorge said: "Audio is engenearing...not mistic or black magic!"
Well that is true. However, I have encountered two classic problems among audiophiles and engineers alike. One is indeed to think that "the reason for something sounding better than than the rest is some form of magic or black art and is therefore unknowable through engineering" and the other is to think that "if standard engineering principles are applied then it is possible to make a duplicate that will sound as good because there is no hocus-pocus involved"
Well I can report from personl experience that neither of these views is correct. The former is born out of ignorance - what I like to call the "corn circle" belief system: I don't understand it therefore it is magic (or aliens). The second view is born out of ignorance and lack of listening experience. In fact it is usually espoused by engineers who listen with their eyes. The problem here is that too limited an engineering theory is applied to assess the design and therefore the conclusion is that the design can be duplicated from the schematic - or, more commonly, a denial that sonic differences exist or are controllable. I could call this the "it should sound this way so it does and if you don't think it does then you are deaf" phenomena.
I have learned from this that I should open my ears and then open my mind, in that order.
BAM
"Don't bother to flame me if you disagree...."
Spoil-sport.
Well that is true. However, I have encountered two classic problems among audiophiles and engineers alike. One is indeed to think that "the reason for something sounding better than than the rest is some form of magic or black art and is therefore unknowable through engineering" and the other is to think that "if standard engineering principles are applied then it is possible to make a duplicate that will sound as good because there is no hocus-pocus involved"
Well I can report from personl experience that neither of these views is correct. The former is born out of ignorance - what I like to call the "corn circle" belief system: I don't understand it therefore it is magic (or aliens). The second view is born out of ignorance and lack of listening experience. In fact it is usually espoused by engineers who listen with their eyes. The problem here is that too limited an engineering theory is applied to assess the design and therefore the conclusion is that the design can be duplicated from the schematic - or, more commonly, a denial that sonic differences exist or are controllable. I could call this the "it should sound this way so it does and if you don't think it does then you are deaf" phenomena.
I have learned from this that I should open my ears and then open my mind, in that order.
BAM
"Don't bother to flame me if you disagree...."
Spoil-sport.
Re another idea
It is a Spanish RAM PA amplifier and I see that nothing in this world is as it appears to be.
Every kind of electrotech designs is very often built upon others "own" design and maybe patents so... why is the Krell schematics secrets it must be a part of the consumers right to know what his product consists of especially if it is broken and the only way is to ship it across the big see.
It is a Spanish RAM PA amplifier and I see that nothing in this world is as it appears to be.
Every kind of electrotech designs is very often built upon others "own" design and maybe patents so... why is the Krell schematics secrets it must be a part of the consumers right to know what his product consists of especially if it is broken and the only way is to ship it across the big see.
Tube_Dude said:
To some extent this is correct. But when you want to go further and improve the basic engineering and extract maximum potential from a design the materials that were used for building the amp and the way they interact with each other play tremendous role. I can take two exact circuit boards with exactly the same PS and put them in two different enclosures and I will have two different sounding amps. My recent experiments proved that.
And don't forget that when you place those amps on two different supporting platforms, they would also sound different. Even the feet change the sound.
To me it really is more like black magic and I will go as far as saying that it is as important as engineering.
Enclosure
Enclosure! Thats the "secrets"about the Krell and Pass amplifiers, they use very good enclosures, even the oldies like the ksa100 and Nelsons Treshold series uses very good and robust enclosures.
If you going to build an amplifier try make get the best but... do not buy the enclosure just build it.
The high tech machine shop can make a very good high end enclosure for your amp.
They use high tech cutting machines and the cost of material is very cheap use stainless steel, this material need no after treatment just simple polish helps to hold the costs down.
Aluminium is a little more expensive as it need anodizing.
Enclosure! Thats the "secrets"about the Krell and Pass amplifiers, they use very good enclosures, even the oldies like the ksa100 and Nelsons Treshold series uses very good and robust enclosures.
If you going to build an amplifier try make get the best but... do not buy the enclosure just build it.
The high tech machine shop can make a very good high end enclosure for your amp.
They use high tech cutting machines and the cost of material is very cheap use stainless steel, this material need no after treatment just simple polish helps to hold the costs down.
Aluminium is a little more expensive as it need anodizing.
Enclosure idea
I got an idea, is it anybody out there who can make pro drawings with computer CAD software if so this forum can construct an enclosure that hold very high standard same as the high end industry just make it with international measures.
The art of making an enclosure is not a thing just for Krell,Pass and Co.
I repeat it is cheapier than you think.
An example make the faceplate thick say 10 mm and the rest af enclosure parts 5mm.
I got an idea, is it anybody out there who can make pro drawings with computer CAD software if so this forum can construct an enclosure that hold very high standard same as the high end industry just make it with international measures.
The art of making an enclosure is not a thing just for Krell,Pass and Co.
I repeat it is cheapier than you think.
An example make the faceplate thick say 10 mm and the rest af enclosure parts 5mm.
materials matter
When designing an enclosure don't forget that...
Mechanical rigidity is a good thing for electronic circuits.
Sheilding circuits from external RF noise is a good thing.
Ferrous metals (iron/steel) conduct magnetic fields.
Non-ferrous, conductive metals (Al, Cu) reflect magnetic fields.
Non-ferrous, non-conductive materials (wood, plastic) are invisible to magnetic fields.
When designing an enclosure don't forget that...
Mechanical rigidity is a good thing for electronic circuits.
Sheilding circuits from external RF noise is a good thing.
Ferrous metals (iron/steel) conduct magnetic fields.
Non-ferrous, conductive metals (Al, Cu) reflect magnetic fields.
Non-ferrous, non-conductive materials (wood, plastic) are invisible to magnetic fields.
Enclosure material
I think that you can find nonmagnetic stainless steel but Al is also very handy to use as you can anodizie it and get a very attractive surface.
And when you cut the enclosure parts it is also very easy to make any hole with high precision and this is very important for the stability and it also make it easier to assemble.
So if any can make an example and saving it as dfx, it may be useful for any amateur to begin with and even make changes to the basic drawing in a cheapier software as the AutoSketch.
I think that you can find nonmagnetic stainless steel but Al is also very handy to use as you can anodizie it and get a very attractive surface.
And when you cut the enclosure parts it is also very easy to make any hole with high precision and this is very important for the stability and it also make it easier to assemble.
So if any can make an example and saving it as dfx, it may be useful for any amateur to begin with and even make changes to the basic drawing in a cheapier software as the AutoSketch.
anodizing
It would be better to use nonmagnetic stainless steel as the shielding of the enclosure would be much better as the anodizing Al enclosure parts is isolated very hard to make any good contact between the different parts ofcourse it would be better to make the enclosure of one or two pieces but the enclosure wouldn't be so versatile to use, the idea is to make one or two standard drawings that can be used either as an preamp/buffer and power amp enclosure.
The assembling structure must be very rigid and straightforward.
It would be better to use nonmagnetic stainless steel as the shielding of the enclosure would be much better as the anodizing Al enclosure parts is isolated very hard to make any good contact between the different parts ofcourse it would be better to make the enclosure of one or two pieces but the enclosure wouldn't be so versatile to use, the idea is to make one or two standard drawings that can be used either as an preamp/buffer and power amp enclosure.
The assembling structure must be very rigid and straightforward.
Re: Krell scematics.
Hi Peter,
I'm looking (for more than 4 years)for the schematics of the PAM-1 and PAM-3, Would you be so kind, as to post the schematic of the PAM-3 (and if you have the 1 also)?
Thanks in advance.
Audiofanatic
webercarbmann said:Hi again
The scematics i have are not werry good. I will have to make a better version using protel. By the way, i also have a scematics for a PAM 3 preamp. I will try to post them in a near future. Please be patient. Would some of you be interested in some KMA 100 PCB`s ??
Peter
Hi Peter,
I'm looking (for more than 4 years)for the schematics of the PAM-1 and PAM-3, Would you be so kind, as to post the schematic of the PAM-3 (and if you have the 1 also)?
Thanks in advance.
Audiofanatic
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.