Thanks guys, I checked the wiki and didn't see anything about cost. I know the approximate costs of things like the toroid, chassis, and power caps, but since people have talked about this amp being on the more expensive side I wasn't sure whether the components themselves tend to be on the pricey side.
Austin
Austin
The parts kits I sold were in the 30-40 dollar a pair (stereo) range for Dale D60 resistors, Wima caps, Nichicon low esr electrolytics, bournes or Alps pots...
The parts list spread sheet is in the thread or wiki...
The parts list spread sheet is in the thread or wiki...
The single largest cost is potentially big enough heat sinks. Plan on at least 11 pounds per channel if not using fans. After that it is the transformer at $60-140 per channel depending on source. All the small parts should be <$150 per channel. You can do better if you are patient and look hard....Parts Express has Avel transformers on sale right now....
Bob G.
Bob G.
So now that I have the DC offset fixed and grounding scheme all squared away, I finally had some time to do some critical listening.
I find it hard to believe this is only a 50 - 70 wpc amp. Seems to have more punch at loud volumes than my UCD400 does. Must be that monster 1350VA transformer?
Anyway, hard to objectify the sound , but sounds very clear, fast, and snappy. Quite impressed. Best I have heard on my Maggies.
I have more output devices and was thinking of adding another pair of 2SA1943/2SC5200 to the 3 pair setup I have to increase output power another 20wpc (I have plenty of heatsink and VA, so I was going to keep bias up to existing 630mv), but don't know if I should really bother...
A pic of the Krell is below - the big heasink box on left -
The preamp on the mid-right is my 5687 tube Audio Note M7 (PCB off ebay) I built that sounds very nice...LED Bias, tons of mods to get her quiet (DC on filaments, CLCR power supply filter), but very happy with it now...
The amp on the top right is a UCD400AD and the silver amp on the lower shelf is another UCD400AD (my old reference amp in a Hexateq chassis). I was thinking of BI-amping the maggies with the two UCD amps - I might just sell them off now....
I find it hard to believe this is only a 50 - 70 wpc amp. Seems to have more punch at loud volumes than my UCD400 does. Must be that monster 1350VA transformer?
Anyway, hard to objectify the sound , but sounds very clear, fast, and snappy. Quite impressed. Best I have heard on my Maggies.
I have more output devices and was thinking of adding another pair of 2SA1943/2SC5200 to the 3 pair setup I have to increase output power another 20wpc (I have plenty of heatsink and VA, so I was going to keep bias up to existing 630mv), but don't know if I should really bother...
A pic of the Krell is below - the big heasink box on left -
The preamp on the mid-right is my 5687 tube Audio Note M7 (PCB off ebay) I built that sounds very nice...LED Bias, tons of mods to get her quiet (DC on filaments, CLCR power supply filter), but very happy with it now...
The amp on the top right is a UCD400AD and the silver amp on the lower shelf is another UCD400AD (my old reference amp in a Hexateq chassis). I was thinking of BI-amping the maggies with the two UCD amps - I might just sell them off now....
Attachments
Hum is gone. Very clean sounding now.
Didn't quite sound right when I had the Signal Ground tripped up - knew something wasn't quite right...too bad it took me a week to figure it out
Didn't quite sound right when I had the Signal Ground tripped up - knew something wasn't quite right...too bad it took me a week to figure it out
Nice job
...I think I might recognize that chassis... 😉
It does not surprise me that the Krell klone works well on the maggies. I have a pair of Apogee Duetta Signatures that are a little brutal on the driving amp. Thus far the Klones (and presumably real ones) are clearly the best sounding amps for my speakers, and I've built and tried a few different designs. (Aleph, Gainclone, Leach etc).
I have anecdotal evidence to suggest the Naim amps can sound pretty darn good driving hard loads, so I think after the present Krell project, a pair of nap135's will be next up.
Stuart
...I think I might recognize that chassis... 😉
It does not surprise me that the Krell klone works well on the maggies. I have a pair of Apogee Duetta Signatures that are a little brutal on the driving amp. Thus far the Klones (and presumably real ones) are clearly the best sounding amps for my speakers, and I've built and tried a few different designs. (Aleph, Gainclone, Leach etc).
I have anecdotal evidence to suggest the Naim amps can sound pretty darn good driving hard loads, so I think after the present Krell project, a pair of nap135's will be next up.
Stuart
Thanks Stuart. That chassis really makes the amp - Only have two of the four sinking and only 40C at 630mv...I have a bit of room (50C?) for more bias, and also have the other two sinks.
The Naims are not Class A are they? There's a thread on the Naim NAP140 PCB up on ebay somewhere here on DIYAudio...I asked someone on that thread if they plan on raising the bias over the specified 35ma...
The Aleph 60 is next, but I believe its going to be pretty hard to beat the Krell...
The Naims are not Class A are they? There's a thread on the Naim NAP140 PCB up on ebay somewhere here on DIYAudio...I asked someone on that thread if they plan on raising the bias over the specified 35ma...
The Aleph 60 is next, but I believe its going to be pretty hard to beat the Krell...
Yup,
...Naims are not class A.
AFAIK The best examples are fully regulated, quasi complementary, low bias, class AB designs.
The Naim clone kits from ebay (I have a few) are the standard Naim circuit implemented without regulation. One version includes output current limiting, the other does not.
I have a number of articles about how they play music so well, but all the reviews are ages old and use Linn speakers...
I've never heard of one of these being 'overbiased', I don't know if the thermal compensation is calibrated properly for class A operation.
Plus it really isn't all that Naim like as a class A amp...thats what krells etc are for...
just my 2c
Stuart
...Naims are not class A.
AFAIK The best examples are fully regulated, quasi complementary, low bias, class AB designs.
The Naim clone kits from ebay (I have a few) are the standard Naim circuit implemented without regulation. One version includes output current limiting, the other does not.
I have a number of articles about how they play music so well, but all the reviews are ages old and use Linn speakers...
I've never heard of one of these being 'overbiased', I don't know if the thermal compensation is calibrated properly for class A operation.
Plus it really isn't all that Naim like as a class A amp...thats what krells etc are for...
just my 2c
Stuart
sto_ag said:
http://www.diyaudio.com/forums/showthread.php?postid=1397526#post1397526
Bob took over distribution for this and the KSA-100 clone.
Thanks again Bob for continuing to carry the torch, and Mark for getting it lit...
Let me see if I can help
Hi,
so how you want to go about KSA 50 ?
do you need ready PCB, or need help in making PCB yourself.
Do you have Gerber or schematics ?
-suds
Hi,
so how you want to go about KSA 50 ?
do you need ready PCB, or need help in making PCB yourself.
Do you have Gerber or schematics ?
-suds
some components can replace??
my pcb have designed,the pcb the same as DELTA Audio's.
now i have problem that some components can't find.
so i want consult :
1. the Bias Q "2sc3955" can use MJE15030 replace??
2. the MJE15032/MJE15033, can use the MJE15030 MJE15031 replace?
because i find the ksa100's sch use the MJE15030 for BIAS Q,and use the MJE15030 MJE15031 for dirver.
my pcb have designed,the pcb the same as DELTA Audio's.
now i have problem that some components can't find.
so i want consult :
1. the Bias Q "2sc3955" can use MJE15030 replace??
2. the MJE15032/MJE15033, can use the MJE15030 MJE15031 replace?
because i find the ksa100's sch use the MJE15030 for BIAS Q,and use the MJE15030 MJE15031 for dirver.
Attachments
Zeng,
the original had MJE15030/31 devices, see the schematic that Bra posted at the beginning of this thread.
So : 1 and 2 = yes.
But using a 50W MJE15030 transistor as a Vbe multiplier device is really stupid.
The "Bias Q" does not need to handle much current, and it also does not need a Vce of several hundred volts.
It is common to take a bigger device for the "Bias Q" than necessary to have it stay at a fairly constant temperature.
A 2SC3955 is particularly pretty because the package is fully insulated, there are several other TO126 devices that can be used for the "Bias Q" instead of a MJE15030 power (driver) transistor.
the original had MJE15030/31 devices, see the schematic that Bra posted at the beginning of this thread.
So : 1 and 2 = yes.
But using a 50W MJE15030 transistor as a Vbe multiplier device is really stupid.
The "Bias Q" does not need to handle much current, and it also does not need a Vce of several hundred volts.
It is common to take a bigger device for the "Bias Q" than necessary to have it stay at a fairly constant temperature.
A 2SC3955 is particularly pretty because the package is fully insulated, there are several other TO126 devices that can be used for the "Bias Q" instead of a MJE15030 power (driver) transistor.
some question of pair of NPN&PNP
Thanks Jacco vermeulen very much!
1. Can the ksa50 work in class A or class AB?
2. If it work in class A, how much Amperes of current per pair of MJE21193/94 generally?
3. How watts the transformer per channal?
4. And if the ksa50 work in class AB, how much Amperes per pair driver?
5. Is it necessary that the parameter of NPN and PNP mush be symmetrical?
such as the dirver MJE15032 15033, 2SC2240 2SA970 , if they not symmetrical,output offset is zero??
6. what is the good method of pair NPN and PNP? (it means :select the symmetrical parameter of NPN&PNP)
5 and 6 = necessary?
thanks.
Thanks Jacco vermeulen very much!
1. Can the ksa50 work in class A or class AB?
2. If it work in class A, how much Amperes of current per pair of MJE21193/94 generally?
3. How watts the transformer per channal?
4. And if the ksa50 work in class AB, how much Amperes per pair driver?
5. Is it necessary that the parameter of NPN and PNP mush be symmetrical?
such as the dirver MJE15032 15033, 2SC2240 2SA970 , if they not symmetrical,output offset is zero??
6. what is the good method of pair NPN and PNP? (it means :select the symmetrical parameter of NPN&PNP)
5 and 6 = necessary?
thanks.