Something is right
Hello everyone.
I have built an NS10 pre-amplifier according to one of the first schematics that RussWhite put up in this thread (in post 20), using BC 549C, BC 546 and BC 556 with bog standard parts (on the circuit board I put up in post 374). This was just as a "quickie circuit" so that I could check the function of my LM3886 chipamps which I had put a fair bit more effort into ... Carlos FMs snubberized unregulated power supply (no snubbers on the amp boards yet), nice short signal paths and close components.
This is just to say that the combination, even with such cruddy parts, sounds very good to me. Certainly it is better than my old 100W/channel Denon PMA-770 which cost me double the money 20 years ago - and they were REAL dollars back then!
This has well and truly whet my apetite for some more DIY audio. The limiting feature is my 25 year old JBL L40 speakers with two "replacement" $3 tweeters!
Thank-you for putting this thread together.
Multiple thanks to Nelson Pass for allowing us to use, abuse and play with his older designs.
I have another post ... which is coming soon ... asking some questions about JFETs and this design.
Regards,
George.
www.stolenantiques.bravehost.com
Hello everyone.
I have built an NS10 pre-amplifier according to one of the first schematics that RussWhite put up in this thread (in post 20), using BC 549C, BC 546 and BC 556 with bog standard parts (on the circuit board I put up in post 374). This was just as a "quickie circuit" so that I could check the function of my LM3886 chipamps which I had put a fair bit more effort into ... Carlos FMs snubberized unregulated power supply (no snubbers on the amp boards yet), nice short signal paths and close components.
This is just to say that the combination, even with such cruddy parts, sounds very good to me. Certainly it is better than my old 100W/channel Denon PMA-770 which cost me double the money 20 years ago - and they were REAL dollars back then!
This has well and truly whet my apetite for some more DIY audio. The limiting feature is my 25 year old JBL L40 speakers with two "replacement" $3 tweeters!
Thank-you for putting this thread together.
Multiple thanks to Nelson Pass for allowing us to use, abuse and play with his older designs.
I have another post ... which is coming soon ... asking some questions about JFETs and this design.
Regards,
George.
www.stolenantiques.bravehost.com
JFETS and Russ's take on NS10
Education on specification sheets please:
Looking at Lumunauw's transistor options early in the thread, he talked about using 2SK30 or 2SK170 jfets. These have a V gds breakdown of 40 and 50 volts.
I am ordering some other parts from a shop (some AD815s for another chip-amp style of pre-amp) which stocks PN4392 Jfets. The specification sheets for these talk in terms of V ds ... and come in two flavours Phillips which has this value being 40 V(as is its V gdo and V gso) and the Fairchild version which has these values each at 30 Volts.
1. How do the latter values (Vds etc) compare with the V gds. Are they different terms for the same thing? Do you ADD Vds to Vgd to get Vgds???
2. Would the 30V versions be satisfactory with rails of +/- 25 Volts? (If it just a matter of slightly shorter working life - years rather than decades, then I could live with that.
If the whole thing will not work, well I couldn't live with THAT!
I also must say that I have become quite lost by lots of the back and forth on this thread.
Who has actually built what?
What power supplies are currently being used and how successfully?
Thank-you,
George.
www.stolenantiques.bravehost.com
Education on specification sheets please:
Looking at Lumunauw's transistor options early in the thread, he talked about using 2SK30 or 2SK170 jfets. These have a V gds breakdown of 40 and 50 volts.
I am ordering some other parts from a shop (some AD815s for another chip-amp style of pre-amp) which stocks PN4392 Jfets. The specification sheets for these talk in terms of V ds ... and come in two flavours Phillips which has this value being 40 V(as is its V gdo and V gso) and the Fairchild version which has these values each at 30 Volts.
1. How do the latter values (Vds etc) compare with the V gds. Are they different terms for the same thing? Do you ADD Vds to Vgd to get Vgds???
2. Would the 30V versions be satisfactory with rails of +/- 25 Volts? (If it just a matter of slightly shorter working life - years rather than decades, then I could live with that.
If the whole thing will not work, well I couldn't live with THAT! I also must say that I have become quite lost by lots of the back and forth on this thread.
Who has actually built what?
What power supplies are currently being used and how successfully?
Thank-you,
George.
www.stolenantiques.bravehost.com
Just commenting on GeorgeBoles who said "Multiple thanks to Nelson Pass for allowing us to use, abuse and play with his older designs. " Yes! I agree. Thanks Nelson for all you have published. Seems nelson has understood that in this age of copying, finding "everything" on the web etc. the best way is to constantly invent new circuit, being ahead of the competition. More companies and people should learn from Nelson. but, then, not everyone is as ingenious as Nelson.
RK
RK
It still puzzles me how similar this is to the NAIM preamp gain stage, NAIM uses since the late 70ies (j. Vereker never denied that he did not develop it from several influences, as far as I heard or read) .
So many similarities: all single side classA, clever outputstage, single sided with active CCS load (no simple resistor), etc.
The main difference seems to me - besides very different component values - is the input stage biasing. Here the Pass circuit is more intelligent and self dependent, while the NAIM circuit is very much depending on the quality of the power supply, maybe that is also one reason while NAIM preamps profit so much from very good power supplies.
Hen and/or egg? Or just logical development?
So many similarities: all single side classA, clever outputstage, single sided with active CCS load (no simple resistor), etc.
The main difference seems to me - besides very different component values - is the input stage biasing. Here the Pass circuit is more intelligent and self dependent, while the NAIM circuit is very much depending on the quality of the power supply, maybe that is also one reason while NAIM preamps profit so much from very good power supplies.
Hen and/or egg? Or just logical development?
For some of you that have not followed all the threads about the NS10.
The original circuit was in fact a single rail (24V) stage for this reason Q1 also needed biasing that was supplied by a nice and clever Opamp setup.
You can see a schematic of the original on this thread on a post by R-K some pages back (pdf file), or alternativly on the Threshold NS10 Lineamp PCB thread.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=76557
The original circuit was in fact a single rail (24V) stage for this reason Q1 also needed biasing that was supplied by a nice and clever Opamp setup.
You can see a schematic of the original on this thread on a post by R-K some pages back (pdf file), or alternativly on the Threshold NS10 Lineamp PCB thread.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=76557
I have been so busy lately I have really had to shelve a lot of new projects for just a bit, but I have done a bit of research and testing here.
So, having worked up a prototype of both single and dual rail versions, I can honestly say that now that I have heard both (although only breifly so far) that the dual rail version seems to me slightly better. It of course is a variation on the original, and not the orginal NS10, but how is that bad? It is in fact "My Take on Threshold NS10" and not really a pure clone. 
Now I am working on a new layout which I think will be better than this first single rail prototype.
Thanks to everyone for this very informative and enjoyable thread. I am having a ton of fun doing these.
Cheers!
Russ
So, having worked up a prototype of both single and dual rail versions, I can honestly say that now that I have heard both (although only breifly so far) that the dual rail version seems to me slightly better. It of course is a variation on the original, and not the orginal NS10, but how is that bad?
It is in fact "My Take on Threshold NS10" and not really a pure clone. Now I am working on a new layout which I think will be better than this first single rail prototype.
Thanks to everyone for this very informative and enjoyable thread. I am having a ton of fun doing these.
Cheers!
Russ
Here is an updated version of the layout.
I have added R20 to test loading a bit the Opamp, which of course is optional. This ideas come from Walter Jung and John Curl.
R19 has change position to clean traces arround the Opamp.
All traces are now 1mm which would be easier to etch.
I will follow up with a corrected schematic
I have added R20 to test loading a bit the Opamp, which of course is optional. This ideas come from Walter Jung and John Curl.
R19 has change position to clean traces arround the Opamp.
All traces are now 1mm which would be easier to etch.
I will follow up with a corrected schematic
Attachments
choky said:
hm,regarding "finding "everything" on the web" ...can you elaborate more what I found on net- about AH5012 and AH5020 forsome
?
Erno Borbely used these AH-parts as matched dual (quadruple) p-channel Junction FETs, in the 80ties. Then-counterpart was the 2N5565 (or its brother). See his old Audio Amateur articles. I believe I have a tube of AH5012 somewhere.
Well, OK, not exactly everything is on the web, especially not hardware as such! ;-)
JFETs
Dear All,
I am pretty sure that I managed to obfuscate and confuse in my previous post about using JFETs in this pre-amplifier, so I have tried to be clearer.
I have built an NS10 clone using the Lumunauw's values as described at the beginning of this thread.
Lumanauw mentions using 2sk30 jfets as an option.
http://www.datasheetarchive.com/search.php?search=2SK30ATM&sType=part
These are a little hard to get easily.
I can get pn4392 easily.
http://pdf1.alldatasheet.com/datasheet-pdf/view/18727/PHILIPS/PN4392.html
The datasheets talk about different parameters regarding voltages. The 2SK30 have a V gds (breakdown) of 50 volts. The PN4392 have a V gss(breakdown) of 30volts from Fairchild or 40 volts from Phillips.
1. Is V gds(breakdown) the same thing as V gss(breakdown)? I suspect that they are not.
2. Would the PN4392 be suitable for an NS10 built with rails of +/- 25 Volts? Would it have a very shortened life - two or three years would be OK, one or two weeks would be no fun
Regards,
George.
Dear All,
I am pretty sure that I managed to obfuscate and confuse in my previous post about using JFETs in this pre-amplifier, so I have tried to be clearer.
I have built an NS10 clone using the Lumunauw's values as described at the beginning of this thread.
Lumanauw mentions using 2sk30 jfets as an option.
http://www.datasheetarchive.com/search.php?search=2SK30ATM&sType=part
These are a little hard to get easily.
I can get pn4392 easily.
http://pdf1.alldatasheet.com/datasheet-pdf/view/18727/PHILIPS/PN4392.html
The datasheets talk about different parameters regarding voltages. The 2SK30 have a V gds (breakdown) of 50 volts. The PN4392 have a V gss(breakdown) of 30volts from Fairchild or 40 volts from Phillips.
1. Is V gds(breakdown) the same thing as V gss(breakdown)? I suspect that they are not.
2. Would the PN4392 be suitable for an NS10 built with rails of +/- 25 Volts? Would it have a very shortened life - two or three years would be OK
, one or two weeks would be no fun Regards,
George.
I think Erno uses Japanese dual JFETs these days. I don't know about 2N or whatever as both N and P.
http://www.linearsystems.com/
http://www.interfet.com/
Maybe this could be turned into something spectacular http://www.interfet.com/pdf/DS_IFPA300_1.pdf
I used the Siliconix equivalent of this in a Levinson JC2 clone 25 years ago:
http://www.interfet.com/pdf/DS_P1086_87.pdf . You have to select the devices (I bought some hundreds).
But, sorry, I cannot find any dual P-channels JFETs.
Good luck!
RK
http://www.linearsystems.com/
http://www.interfet.com/
Maybe this could be turned into something spectacular http://www.interfet.com/pdf/DS_IFPA300_1.pdf
I used the Siliconix equivalent of this in a Levinson JC2 clone 25 years ago:
http://www.interfet.com/pdf/DS_P1086_87.pdf . You have to select the devices (I bought some hundreds).
But, sorry, I cannot find any dual P-channels JFETs.
Good luck!
RK
R-K Rønningstad said:I think Erno uses Japanese dual JFETs these days. I don't know about 2N or whatever as both N and P.
http://www.linearsystems.com/
http://www.interfet.com/
Maybe this could be turned into something spectacular http://www.interfet.com/pdf/DS_IFPA300_1.pdf
I used the Siliconix equivalent of this in a Levinson JC2 clone 25 years ago:
http://www.interfet.com/pdf/DS_P1086_87.pdf . You have to select the devices (I bought some hundreds).
But, sorry, I cannot find any dual P-channels JFETs.
Good luck!
RK
Evening Everyone.
1. I have the NS10 schematic from R-K and, given that I have already built a functional NS10-like clone based on the schematic in Post #30, how would I modify that line stage to make an RIAA stage using the dual rail design?
a. Do I use the same values as in the R-K schematic for the filter section and just plug them in between the junction of R1 and R2 and the emitter of Q1? Or is it more complicated? Change the value of R4?
b. How much gain would I put into the RIAA stage (R1/R2)?
2. Has anyone built a really nice quiet power supply suitable for the dual rail version yet? I have become very confused about the power supply situation in this thread ... but the intellectual and engineering power in this thread has far exceed my meagre capabilities.
With thanks,
George.
1. I have the NS10 schematic from R-K and, given that I have already built a functional NS10-like clone based on the schematic in Post #30, how would I modify that line stage to make an RIAA stage using the dual rail design?
a. Do I use the same values as in the R-K schematic for the filter section and just plug them in between the junction of R1 and R2 and the emitter of Q1? Or is it more complicated? Change the value of R4?
b. How much gain would I put into the RIAA stage (R1/R2)?
2. Has anyone built a really nice quiet power supply suitable for the dual rail version yet? I have become very confused about the power supply situation in this thread ... but the intellectual and engineering power in this thread has far exceed my meagre capabilities.
With thanks,
George.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.