Is anyone interested in an excellent quality very low distortion complementary symerty class A pre-amp running at amplifier voltages offering high dynamic range while remaining very simple.
It is seldom that one finds a pre-amp that can run off the same high voltage supply of your amplifier and I may have just the thing.
Kind regards
Nico
It is seldom that one finds a pre-amp that can run off the same high voltage supply of your amplifier and I may have just the thing.
Kind regards
Nico
I have never published any of my designs on diyAudio forum but will gladly do this if there are some interest shown by other enthusiasts and could turn it into a small project.
My second calling is that of being an acoustic jazz musician for more than 30 years, playing several instruments, thus the enjoyment of music to me is very much determined by the accuracy that an instrument's or voice character is retained when reproduced by the sound equipment that I design and use.
Although the measurements are excellent (for instance, our Hewlett Packard distortion analyser cannot measure the distortion with a 2V input signal (typical of a CD player)at any frequency up to 100kHz. Enjoying what you hear and being able to relate to the music is the real test.
It is my experience that it is not possible to design the ultimate equipment that will please every individual, simply because the vast majority of people has never actually heard the real instrument playing and therefor has no point of reference, thus it is perseption that set a trend not reality.
What I believe is one should strive for is to design simplistic audio circuits that approaches that of the ampliwire add ing as little character of its own.
Kindest regards
Nico
My second calling is that of being an acoustic jazz musician for more than 30 years, playing several instruments, thus the enjoyment of music to me is very much determined by the accuracy that an instrument's or voice character is retained when reproduced by the sound equipment that I design and use.
Although the measurements are excellent (for instance, our Hewlett Packard distortion analyser cannot measure the distortion with a 2V input signal (typical of a CD player)at any frequency up to 100kHz. Enjoying what you hear and being able to relate to the music is the real test.
It is my experience that it is not possible to design the ultimate equipment that will please every individual, simply because the vast majority of people has never actually heard the real instrument playing and therefor has no point of reference, thus it is perseption that set a trend not reality.
What I believe is one should strive for is to design simplistic audio circuits that approaches that of the ampliwire add ing as little character of its own.
Kindest regards
Nico
Attachments
Nico Ras said:
Hi Nico,
Nice design. I have a question on your output transistors. You show the BD139/140 series, which IIRC have a max Vce of 80V? That may be problematic with 100V total supply. But maybe this is just for sim purposes, maybe you used other transistors in real world?
Jan Didden
I have been using this design for the past three years, (always busy tweaking something or other) with 30V stabilised supply (LM317/337) followed by a capacitance multiplier.
The BD139/BD140 combination sounds nicer than some other transistors I have used. if one can define nicer.
But you are absolutely right, the transistors of choice for a 50V supply would probably be MJE15034/5, which I have used in the past.
If you asked me how they sounded, then I must admit, probably different whether different is better or worse is really difficult to say.
The reason for the power transistors are that of reduced thermal noise.
The BD139/BD140 combination sounds nicer than some other transistors I have used. if one can define nicer.
But you are absolutely right, the transistors of choice for a 50V supply would probably be MJE15034/5, which I have used in the past.
If you asked me how they sounded, then I must admit, probably different whether different is better or worse is really difficult to say.
The reason for the power transistors are that of reduced thermal noise.
Hi Steen,
You may not be surprised at all, but most people purchasing commercial equipment has no idea what it should sound like. That is why it is called commercial equipment. On the other hand, most youngsters these days listen to music generated electronically, what exactly should it sound like, who knows. There is no point of reference at all.
If all sound equipment was designed that way, it won't sell because it will not vibrate the windows and doors and the owner cannot impress his buddies with the fact that everything rattles when he cranks up the volume.
One of the easiest ways to create sloppy single note bass that sounds like a wet F&%T is to skimp on the power supply. It is almost always 100Hz.
You may not be surprised at all, but most people purchasing commercial equipment has no idea what it should sound like. That is why it is called commercial equipment. On the other hand, most youngsters these days listen to music generated electronically, what exactly should it sound like, who knows. There is no point of reference at all.
If all sound equipment was designed that way, it won't sell because it will not vibrate the windows and doors and the owner cannot impress his buddies with the fact that everything rattles when he cranks up the volume.
One of the easiest ways to create sloppy single note bass that sounds like a wet F&%T is to skimp on the power supply. It is almost always 100Hz.
Me again Steen, I get carried away when at times, us fanatics try to design stuff that could have sounded pretty reasonable if it did not pass through 100s of op-amps, kms of cable, spit covered microphones, overdriven PeeVee amps, and a hung over recording engineers that compresses the audio to the point that all frequencies are at the same loudness.
Now what, who are we trying to please!
We actually design audio equipment essentially for our own pleasure, never to impress anybody other than ourselves.
But if someone else shares the same enthusiasm, it genreates even more enthusiasm, friendships are created and great ideas and experiences are shared.
It all becomes worth while if you can tell that the artist is playing on a Gibson, Fender or Hoffnar. But in most cases the average listener would not know if it is a bass drum, guitar or the bottom notes on a piano.
Now what, who are we trying to please!
We actually design audio equipment essentially for our own pleasure, never to impress anybody other than ourselves.
But if someone else shares the same enthusiasm, it genreates even more enthusiasm, friendships are created and great ideas and experiences are shared.
It all becomes worth while if you can tell that the artist is playing on a Gibson, Fender or Hoffnar. But in most cases the average listener would not know if it is a bass drum, guitar or the bottom notes on a piano.
I am driving into my main amp, input is about 22K. My prvious design used 10 x small signal transistors in parralel in each of the differential input stages. The geometric sum of the non coherent noise causes a 2.5dB noise reduction. This was great in the days of Linn, Itok, Karma, etc, etc.
I found that power transistors achieves almost the same result, having larger junctions.
To be absolutely honest, I built the original using BD139/140 some three years ago (and still use it every day) I used MJE15035/6, for a fellow enthusiast, since I had several hand matched pairs left from the time I designed my main amp.
I remember when we first compared the two pre-amps, they sounded different, but again it may all be in the mind. We discussed and philosified about it but never concluded anything, neither sounded better than the other. Nor could we really say it was really different.
It could have been because the two pre-amps were running of different rails. The BDs where running at 38V while the MJEs were at 50V. Again, which one sounded better, heck I don't know they both sounded damn good!
If you ask is it the best I have ever heard. Of course I designed it, have you ever met anyone who's own brewed system is not the best?
Kindest Regards
I found that power transistors achieves almost the same result, having larger junctions.
To be absolutely honest, I built the original using BD139/140 some three years ago (and still use it every day) I used MJE15035/6, for a fellow enthusiast, since I had several hand matched pairs left from the time I designed my main amp.
I remember when we first compared the two pre-amps, they sounded different, but again it may all be in the mind. We discussed and philosified about it but never concluded anything, neither sounded better than the other. Nor could we really say it was really different.
It could have been because the two pre-amps were running of different rails. The BDs where running at 38V while the MJEs were at 50V. Again, which one sounded better, heck I don't know they both sounded damn good!
If you ask is it the best I have ever heard. Of course I designed it, have you ever met anyone who's own brewed system is not the best?
Kindest Regards
Only ourself's, as you note. Not many people are interested in serious hifi, these days. All you see is people carrying home 5 channel sourround junk from the nearest supermarket
I like it too
I also like the enthusiasm and getting carried away. Two good signs of being dedicated.Nico, did you P2P wire your preamp and, as you mentioned running it from the poweramp supply, did you clean the supply up,further?
Steen
Hi Steen,
My amp is class A and the power supply is pretty hefty at +-50V with 94 000 uF (20 x 4700 uF) on each rail.
I did add another 4700uF on each rail of the pre-amp PCB not that this would make much difference, but it makes the board at least look more interesting.
There is a very faint power supply buzz (100 Hz) that I can hear on my headphones if pushed hard onto my ears, but not audible on the speakers.
I suppose that a series regulator on the pre-amp with capacitance multiplier could improve matters and chop off any remaining ripple present on the supply.
I will update the circuit and add a regulator to the board which is no sweat at all maybe even bring it down a few volt so that it is not necessary.
I need to make a confession that I am cheating a bit. I actually have a Linkwitz Railey equalisation network in the feed-back path of my pre-amp that corrects my speaker response and it does work great. It is already on the PCB and I was going to remove this to satisfy our purist friends out there.
Anyway if you want to run off a higher power supply voltage I would use MJE15030/1 or MJE15034/5 that I am using as driver stage in the power-amp.
It would be totally subjective to tell you which one sounds better, because I don't think you will ever notice.
Although the BD139/40 Vce is only 80V, they have not croaked in the three years I have used the pre-amp, and I cannot see that either of them will ever have the 100V across as previously suggested by someone on the thread.
Unless you are driving it with a very large signal and either output swings to rail. I guess you will allso be searching for your speaker cones after this incident.
Anyway, there is absolutely no harm in using higher voltage devices and is probably better and safer in the long run.
Kind regards
My amp is class A and the power supply is pretty hefty at +-50V with 94 000 uF (20 x 4700 uF) on each rail.
I did add another 4700uF on each rail of the pre-amp PCB not that this would make much difference, but it makes the board at least look more interesting.
There is a very faint power supply buzz (100 Hz) that I can hear on my headphones if pushed hard onto my ears, but not audible on the speakers.
I suppose that a series regulator on the pre-amp with capacitance multiplier could improve matters and chop off any remaining ripple present on the supply.
I will update the circuit and add a regulator to the board which is no sweat at all maybe even bring it down a few volt so that it is not necessary.
I need to make a confession that I am cheating a bit. I actually have a Linkwitz Railey equalisation network in the feed-back path of my pre-amp that corrects my speaker response and it does work great. It is already on the PCB and I was going to remove this to satisfy our purist friends out there.
Anyway if you want to run off a higher power supply voltage I would use MJE15030/1 or MJE15034/5 that I am using as driver stage in the power-amp.
It would be totally subjective to tell you which one sounds better, because I don't think you will ever notice.
Although the BD139/40 Vce is only 80V, they have not croaked in the three years I have used the pre-amp, and I cannot see that either of them will ever have the 100V across as previously suggested by someone on the thread.
Unless you are driving it with a very large signal and either output swings to rail. I guess you will allso be searching for your speaker cones after this incident.
Anyway, there is absolutely no harm in using higher voltage devices and is probably better and safer in the long run.
Kind regards
If you are a purist and drive a CD through only a potentiometer to the power amp, then there is no reason for a pre-amp.
However, if you want to connect your Linn LP 12, then you can easiliy achieve RIAA equalisation in the feedback and have a very good sounding low noise low distortion to your amp.
This is the basis for whatever the need is, not particulary to boost a CD player output to 40V.
There are many uses for a good pre-amp and I am keen to find out what others may want to do with it.
If you run measurements you will find that the higher the supply the lower THD and IMD because you extend the third order intercept and linear range little gain compression and absolutely no clipping.
However, if you want to connect your Linn LP 12, then you can easiliy achieve RIAA equalisation in the feedback and have a very good sounding low noise low distortion to your amp.
This is the basis for whatever the need is, not particulary to boost a CD player output to 40V.
There are many uses for a good pre-amp and I am keen to find out what others may want to do with it.
If you run measurements you will find that the higher the supply the lower THD and IMD because you extend the third order intercept and linear range little gain compression and absolutely no clipping.
Nico,
Yes, the simple class-A three transistor discret op-amp (diff.amp) in symetrical style, sounds very good and have also a nice stability behavior.
GAS (Great American Sound) use near identical circuit as a line amp in "Thaila" preamp from 1977, but it have +/- 24 volt as voltage feed.
John Curl`s JC-2 line amp from 1978, have a similar topology, besides the JFET input N and P channel transistor pair.
Also other preamps as Accuphase C200, Onkyo P303 and Harman Kardon 2500, use same bipolar topology.
BTW,
Hitachi have a BD139/140 style
(in same TO126 case) with higher Vceo.
NPN, 2SD669-A and PNP, 2SB649A
160V - 1,5A- 20W- hfe 60-200 - 140mHz
(2SD669 and 2SB649 with out A, have Vceo 120V)
Try also a old workhorse as NPN, 2N3440 and PNP, 2N5415 in TO39 case, 200V - 1A - 10W
--Bo
Yes, the simple class-A three transistor discret op-amp (diff.amp) in symetrical style, sounds very good and have also a nice stability behavior.
GAS (Great American Sound) use near identical circuit as a line amp in "Thaila" preamp from 1977, but it have +/- 24 volt as voltage feed.
John Curl`s JC-2 line amp from 1978, have a similar topology, besides the JFET input N and P channel transistor pair.
Also other preamps as Accuphase C200, Onkyo P303 and Harman Kardon 2500, use same bipolar topology.
BTW,
Hitachi have a BD139/140 style
(in same TO126 case) with higher Vceo.
NPN, 2SD669-A and PNP, 2SB649A
160V - 1,5A- 20W- hfe 60-200 - 140mHz
(2SD669 and 2SB649 with out A, have Vceo 120V)
Try also a old workhorse as NPN, 2N3440 and PNP, 2N5415 in TO39 case, 200V - 1A - 10W
--Bo
Hello Bo,
Yes I used MJE15034/5 combination this pre-amps made for a few non-electronic enthusiasts. (Vce at 350V) and I will utilise these shortly in my pre-amp.
My original circuit using this topology, was in about 1977 just after completing my studies. This pre-amp had RIAA equalisation and used an array of MPSA06/MPSA56 in the long tail pair.
I had to replace it about three years ago when it became very noisy clicking, popping and hissing and I eventually decided to shoot the old work-horse.
The raw PCB which at the time was made using an art pen, sunlight and brown smelly Ferro Chloride solution had deteriorated over the 26 years to green gunge and looked like it was eaten by rats.
Simulation software did make empirical design a lot quicker but it still took several months before finally placing it on my shelf.
I am lucky to co-own an electronic company having sheet metal, machine shop, modern multi-layer PCB manufacturing, transformer winding facility and three high speed pick-and place lines, allowing a very professional result.
Our R&D department also has the advantage of receiving early samples of new components from manufacturers for evaluation and I am also a keen collector of parts that goes missing from the lab sample stock.
The +-200 watt class A power amp I have been working on and of for the best part of two years will be complete before Xmas using an array of what I consider being fairly decent transistors from On-Semi MJL4218/4302 (Vce 350V; 230W; 15A; High SOA).
The prototypes have proven very pleasing and as you rightly mention, a similar topology probably exists elsewere, but this is not very important.
Interestingly, during this exercise I concluded that "high-end" unaffordable amp manufacturing budget spread is the most expensive parts being purpose made heatsinks, the power supply then machined case and screening, and finaly the rest of the amplifier electronics only being a fraction of the total cost.
I must admit, I cribbed a few of Nelson's amplifier aestehitcs because I think they are quite beautiful and must admit a meter here and there will hurt no-one. Unlike bar-graphs, meters have hypnotic effects because they are alive.
Kindest regards
Nico
P.S. I like the photo of you proudly standing with your spekers.
Yes I used MJE15034/5 combination this pre-amps made for a few non-electronic enthusiasts. (Vce at 350V) and I will utilise these shortly in my pre-amp.
My original circuit using this topology, was in about 1977 just after completing my studies. This pre-amp had RIAA equalisation and used an array of MPSA06/MPSA56 in the long tail pair.
I had to replace it about three years ago when it became very noisy clicking, popping and hissing and I eventually decided to shoot the old work-horse.
The raw PCB which at the time was made using an art pen, sunlight and brown smelly Ferro Chloride solution had deteriorated over the 26 years to green gunge and looked like it was eaten by rats.
Simulation software did make empirical design a lot quicker but it still took several months before finally placing it on my shelf.
I am lucky to co-own an electronic company having sheet metal, machine shop, modern multi-layer PCB manufacturing, transformer winding facility and three high speed pick-and place lines, allowing a very professional result.
Our R&D department also has the advantage of receiving early samples of new components from manufacturers for evaluation and I am also a keen collector of parts that goes missing from the lab sample stock.
The +-200 watt class A power amp I have been working on and of for the best part of two years will be complete before Xmas using an array of what I consider being fairly decent transistors from On-Semi MJL4218/4302 (Vce 350V; 230W; 15A; High SOA).
The prototypes have proven very pleasing and as you rightly mention, a similar topology probably exists elsewere, but this is not very important.
Interestingly, during this exercise I concluded that "high-end" unaffordable amp manufacturing budget spread is the most expensive parts being purpose made heatsinks, the power supply then machined case and screening, and finaly the rest of the amplifier electronics only being a fraction of the total cost.
I must admit, I cribbed a few of Nelson's amplifier aestehitcs because I think they are quite beautiful and must admit a meter here and there will hurt no-one. Unlike bar-graphs, meters have hypnotic effects because they are alive.
Kindest regards
Nico
P.S. I like the photo of you proudly standing with your spekers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.