V3 Boards
Pierre
You wrote:
This reminds me of a famous turntable manufacturer (with whose products I had a very speedy divorce) and others who used to claim in their adverts that their products were "handcrafted, specially made from ancient metals that had been secretly smuggled under the seat of a camel through the Gobi desert to an old yogi high up in the mountain where it had been made with the utmost care" before being brought to you at a highly inflated special low price of $%&@.%$ !! Only to sound terrible at home.
Maybe one should just stick to the rules of supply and demand?
Jozua
Pierre
You wrote:
This reminds me of a famous turntable manufacturer (with whose products I had a very speedy divorce) and others who used to claim in their adverts that their products were "handcrafted, specially made from ancient metals that had been secretly smuggled under the seat of a camel through the Gobi desert to an old yogi high up in the mountain where it had been made with the utmost care" before being brought to you at a highly inflated special low price of $%&@.%$ !! Only to sound terrible at home.
Maybe one should just stick to the rules of supply and demand?
Jozua
Extrapolating from the statistics so far, 10 stereo sets with no "disclaimer" will probably lead to less than three complete amps being built.
Given that I've never built a KSA for myself and never will, it's pushing the goodwill to go through all the effort if the assembled result will not end up in less than at least five homes.
If I was paid for the work I couldn't be bothered if only one board was made and ends up dangling in the guy's car rear-view mirror, but this is purely for the love of the cause and therefore I'd like to see at least a few in active use.
Given that I've never built a KSA for myself and never will, it's pushing the goodwill to go through all the effort if the assembled result will not end up in less than at least five homes.
If I was paid for the work I couldn't be bothered if only one board was made and ends up dangling in the guy's car rear-view mirror, but this is purely for the love of the cause and therefore I'd like to see at least a few in active use.
Well, I'm bussy putting the lat screws into my DX amp, but I'm totaly willing to do a comparison...then again you are not so far that I can't listen to yours...if it wins and can live off +-35V rails, its just a matter of swapping out PCBs...
2 x 300VA 0-26-0-26VAC transformers
2 x 2.3m² Heatinks
2 x 300VA 0-26-0-26VAC transformers
2 x 2.3m² Heatinks
PWatts said:
It seems like only yesterday, but mine (jan's/jens' boards) has been up and running for 1 year, 11 months. Only 1 small problem with the on/off switch that was easily fixed.
One more KSA50 clone finished.
🙂
http://www.diyaudio.com/forums/showthread.php?postid=1303666#post1303666
🙂
http://www.diyaudio.com/forums/showthread.php?postid=1303666#post1303666
I will be very interested in hearing what Jozua thinks of the DX amps in comparison to his Krells!
Dx vs Commercial 50.2
Harry
My first 50 watt clone was build with high quality components. The only area where I skimped was the 25-0-25 volt PSU transformers I had lying around. The clone could possibly sound better with a stronger PSU.
The clone although nice does simply does not have the soundstage depth and dynamic slam and overall naturalness of the commercial 50.2. For me the commercial version wins hands down. It is a day-night difference.
I am still toying at making a second attempt with a set of P-Mouse boards that are 80% complete and using another spare 35-0-35 PSU that I have. This time sticking much closer to the original design. The question for me now is just how much money do I still want to throw at this project when the results are uncertain.
My 50 watt clone uses better parts compared to the original but still does not sound as good. The question is: WHY ? !!
I often wonder what could be the X-factor that makes the commerical unit sound so good? Especially also from a long term service perspective when parts start failing. With hindsight it now seems to me that you need to tread very carefully when you wish to recreate an amplifier with the exact sonic print than the original.
Jozua
Harry
My first 50 watt clone was build with high quality components. The only area where I skimped was the 25-0-25 volt PSU transformers I had lying around. The clone could possibly sound better with a stronger PSU.
The clone although nice does simply does not have the soundstage depth and dynamic slam and overall naturalness of the commercial 50.2. For me the commercial version wins hands down. It is a day-night difference.
I am still toying at making a second attempt with a set of P-Mouse boards that are 80% complete and using another spare 35-0-35 PSU that I have. This time sticking much closer to the original design. The question for me now is just how much money do I still want to throw at this project when the results are uncertain.
My 50 watt clone uses better parts compared to the original but still does not sound as good. The question is: WHY ? !!
I often wonder what could be the X-factor that makes the commerical unit sound so good? Especially also from a long term service perspective when parts start failing. With hindsight it now seems to me that you need to tread very carefully when you wish to recreate an amplifier with the exact sonic print than the original.
Jozua
That clone (which I built) used better passive components. The rest is lacking compared to the original in critical areas:
1) Transformer voltage more than 10V too low
2) Transformer power rating too low
3) Poor transformer load regulation
4) MJE15032/33 for the predrivers compared to the original's superior 2SA968/2SC2238
5) MPSA42&MPSA92 instead of the original's 2SA968/2SC2238 (Krell used them for both).
These factors contribute far more than the quality of any passive component over the standard Dale and Roederstein components, and the weak transformers are probably the cause, or at least major contributing factor, of the lack of slam and dynamics.
DX vs Commercial
Pierre
If these parts are so critical please refresh my memory as to why the group agreed to the substitutions?
Can the correct parts be used on the existing PCB's and are they obtainable ? If so it might be worth it to revisit this project.
JH
Pierre
If these parts are so critical please refresh my memory as to why the group agreed to the substitutions?
Can the correct parts be used on the existing PCB's and are they obtainable ? If so it might be worth it to revisit this project.
JH
Those decisions were debated by Jan and Al quite early in the thread before the first boards were even made, and changed even further for the PM boards. The trouble was that nobody had a 100% verified schematic of the original, so it was open for discussion. I didn't read and tried to validate all the decisions of the whole thread from the beginning especially since early builders had very favourable comments. They didn't have a real one to compare it side-by-side to so it wasn't possible to have a benchmark reference as you have now.
Although there were good arguments for the choices and substitutions and they're not necessarily bad (though the choice for the MJE's as predrivers is a little strange), they are likely to yield a different sound. Maybe Jacco can provide a detailed insightful opinion for all the transistor options and alternatives? IIRC the higher voltage ratings of some of the transistors were a deciding factor for those who wanted to built prodigously high-power versions.
On the Delta boards the predrivers are a straight swop if going for the 2S ones, but the MPSA42/92 have a different pinout and therefore difficult to change. The same applies to the PM boards, but in addition the input stage will also need to be modified if the original's 2SC2240/2SA970 LTP transistors are to be used since it catered for the larger variety of MPS, BC and 2N series with EBC pinouts.
As I've mentioned earlier, if there's sufficient interest I can route and have a small board batch made for a 100% original design.
Although there were good arguments for the choices and substitutions and they're not necessarily bad (though the choice for the MJE's as predrivers is a little strange), they are likely to yield a different sound. Maybe Jacco can provide a detailed insightful opinion for all the transistor options and alternatives? IIRC the higher voltage ratings of some of the transistors were a deciding factor for those who wanted to built prodigously high-power versions.
On the Delta boards the predrivers are a straight swop if going for the 2S ones, but the MPSA42/92 have a different pinout and therefore difficult to change. The same applies to the PM boards, but in addition the input stage will also need to be modified if the original's 2SC2240/2SA970 LTP transistors are to be used since it catered for the larger variety of MPS, BC and 2N series with EBC pinouts.
As I've mentioned earlier, if there's sufficient interest I can route and have a small board batch made for a 100% original design.
Although I have the PCB (made by myself) according to
http://www.delta-audio.com/Krell-Clone.htm, I 100% interested in
this project !!!
http://www.delta-audio.com/Krell-Clone.htm, I 100% interested in
this project !!!
Hi,
the MP and MJ substitutions for the 2s devices does seem a retrograde step.
I bought the Pink boards but would never has dreamed of fitting those MP/J devices. The specifications are too poor for those locations.
I have had my 100Klone sitting doing nothing for a month while contemplating the next step.
I have taken a sudden dislike for the complexities involved in the four stage design.
It may build up the 50Klone PCB ( a three stage design) and replace the 100Klone PCB, but with the same PSU and output stage, to hear what it sounds like. If this removes the stability problems and sounds nice, job done.
the MP and MJ substitutions for the 2s devices does seem a retrograde step.
I bought the Pink boards but would never has dreamed of fitting those MP/J devices. The specifications are too poor for those locations.
I have had my 100Klone sitting doing nothing for a month while contemplating the next step.
I have taken a sudden dislike for the complexities involved in the four stage design.
It may build up the 50Klone PCB ( a three stage design) and replace the 100Klone PCB, but with the same PSU and output stage, to hear what it sounds like. If this removes the stability problems and sounds nice, job done.
The consensus was to go for easily available devices, so the Japanese transistors lost out. Group decision, and I think the right one, as it gave more people the chance to get involved in building amps with a much tested layout, and more experienced constructors could cope with swapping pins to fit what they wanted. I have no regrets.
Can anybody source these predrivers ?
Pinkmouse
Your comment makes a lot of sense. I dont think it is a case of pointing fingers at anyone.
However what are the chances that somebody can source the original predrivers? Especially for those who wish to go the absolute purist route?
Since I bought the commercial version of the 50.2 a couple of months ago I have come to realise that the amp meets all of my audio needs. Had I bought the amp 19 years ago i would never have spent money on buying a 100.1 then a KSA200s and then a 100.2.
I would love to have a modern 50.2 clone that has the sonic attributes of the commercial version.
Jozua
Pinkmouse
Your comment makes a lot of sense. I dont think it is a case of pointing fingers at anyone.
However what are the chances that somebody can source the original predrivers? Especially for those who wish to go the absolute purist route?
Since I bought the commercial version of the 50.2 a couple of months ago I have come to realise that the amp meets all of my audio needs. Had I bought the amp 19 years ago i would never have spent money on buying a 100.1 then a KSA200s and then a 100.2.
I would love to have a modern 50.2 clone that has the sonic attributes of the commercial version.
Jozua
Get your PSU up to something decent, and I'm sure you will. During my preproduction testing, the PSU voltage and biasing made a much greater difference than component substitution. It's still a good amp with lower rails, but get them up to 40V or so and it starts to sing.
Andrew,
What you said,
"I have had my 100Klone sitting doing nothing for a month while contemplating the next step.
I have taken a sudden dislike for the complexities involved in the four stage design.
It may build up the 50Klone PCB ( a three stage design) and replace the 100Klone PCB, but with the same PSU and output stage, to hear what it sounds like. If this removes the stability problems and sounds nice, job done."
It is the same case of me. The only different is that I had my KSA50 driver bd partly soldered. (because I found the spacing for the components, small heat sinks are not so fit)
I am looking for what people's advises !
What you said,
"I have had my 100Klone sitting doing nothing for a month while contemplating the next step.
I have taken a sudden dislike for the complexities involved in the four stage design.
It may build up the 50Klone PCB ( a three stage design) and replace the 100Klone PCB, but with the same PSU and output stage, to hear what it sounds like. If this removes the stability problems and sounds nice, job done."
It is the same case of me. The only different is that I had my KSA50 driver bd partly soldered. (because I found the spacing for the components, small heat sinks are not so fit)
I am looking for what people's advises !
I still think that someone needs to trace out the circuit on the original 50watt amp and compare it to the klones that we have built.
What is the power supply voltages on the original? I was lead to believe that 25Volts ac was optimum with a 500VA transformer per channel.
As to the comparison that I was refering to in my earlier post I was hoping that if Nordic lived close to Jozua they could compare the DX amp to the Krells amps.
What is the power supply voltages on the original? I was lead to believe that 25Volts ac was optimum with a 500VA transformer per channel.
As to the comparison that I was refering to in my earlier post I was hoping that if Nordic lived close to Jozua they could compare the DX amp to the Krells amps.
I've to restore a few small things on Jozua's original so when the boards come out I'll trace a full schematic. I think I've stipulated most of the differences already a few posts ago though.
The voltage is considerably higher than 35VDC at about 49V, so 35-0-35V 500VA would be needed. This coincides with Al's comments that it really starts to sing >40V.
The voltage is considerably higher than 35VDC at about 49V, so 35-0-35V 500VA would be needed. This coincides with Al's comments that it really starts to sing >40V.