Parts selection
No....... parts selection is anything but arbtirary! The parts on this list are there on the basis of listening test, use by designers in good sounding designs, and quaility of the manufacturer. Toshiba and Zetex make great transistors. Go to Nelson Pass's website and see what he is using. Go to Erno Borbely website and see what he is using. Go read the construction articles in The Audio Amatuer and audio electronics and see what they are using. Go to
the resources mentioned on this forum and see what they are using. There are so many resources on the Web that this is almost a no brainer......... I believe that pricing and where to buy most of these has even been mentioned in this forum.
H.H.
No....... parts selection is anything but arbtirary! The parts on this list are there on the basis of listening test, use by designers in good sounding designs, and quaility of the manufacturer. Toshiba and Zetex make great transistors. Go to Nelson Pass's website and see what he is using. Go to Erno Borbely website and see what he is using. Go read the construction articles in The Audio Amatuer and audio electronics and see what they are using. Go to
the resources mentioned on this forum and see what they are using. There are so many resources on the Web that this is almost a no brainer......... I believe that pricing and where to buy most of these has even been mentioned in this forum.
H.H.
My apologies. I must have left it at home.
So, if there are so many resources telling why a particular part is used, how about giving me a few links instead of telling me I'm sitting on my brain. Of course anyone can read through an article and see what someone is using, or look at a schematic. I want to know what the criteria is for selecting these parts. It has to go beyond the simple voltage/current/beta/current producton, etc. How do you know what to look for when designing?
You see, instead of coming off like some egotistical jackass and assuming that I know everything, I like to learn from other people and what they have to say. I'm obviously barking up the wrong tree here, as some people like to speak more about a topic than what they fully understand, and run for cover behind insults when their ignorance is showing. I guess the challenge to offer something more than "so and so used this" must be more than you can handle.
SteveG:
I share your frustration with the learning process.
Problem is, it does take time to accumlate knowledge and
experience. I pick up clues from as many sources as I can
discover, though I've found Doug Self and G. Randy Sloan's
amplifier design books to be especially helpful sources of information with concrete design examples. The Audio Amateur
and some other electronic hobby magazines were occasionally
helpful, though lately the Internet itself has been my biggest
source of news and information.
Most of my practical experience has been with Marshall Leach's
design, and the companies I worked for; that may have given
me a bias in favor of Motorola/ONSemi transistors. Nevertheless,
I simply don't have much information on other transistors, so
I go with what I know and hope to turn up new information,
however randomly.
After I lost my career in electronics in 1990, I also lost access to
engineering magazines, up to date data books, and just plain
hands-on-the-hardware experience. I had to pick up what I could
on my own through reading, building and modifying, and some
occasional communtity college classes.
I still want my career back, but it's looking grimmer every year that slips by. Best I can do is what I'm doing now.
Keep reading, building, and paying attention to the various threads here; that's the best general advice I can give.
I share your frustration with the learning process.
Problem is, it does take time to accumlate knowledge and
experience. I pick up clues from as many sources as I can
discover, though I've found Doug Self and G. Randy Sloan's
amplifier design books to be especially helpful sources of information with concrete design examples. The Audio Amateur
and some other electronic hobby magazines were occasionally
helpful, though lately the Internet itself has been my biggest
source of news and information.
Most of my practical experience has been with Marshall Leach's
design, and the companies I worked for; that may have given
me a bias in favor of Motorola/ONSemi transistors. Nevertheless,
I simply don't have much information on other transistors, so
I go with what I know and hope to turn up new information,
however randomly.
After I lost my career in electronics in 1990, I also lost access to
engineering magazines, up to date data books, and just plain
hands-on-the-hardware experience. I had to pick up what I could
on my own through reading, building and modifying, and some
occasional communtity college classes.
I still want my career back, but it's looking grimmer every year that slips by. Best I can do is what I'm doing now.
Keep reading, building, and paying attention to the various threads here; that's the best general advice I can give.
There are so many resources on the Web that this is almost a no brainer
Hmmmmm....... I will try again. Here's how to determine which transistors are good for audio.
Plan A.
1. Get a BSEE in Electrical Engineering or a B.S. in Physics.
2, Read everything that you can get your hands on about audio circuit design. This would include everything written in The Audio
Amatuer and Audio Electronics. This would specifically include articles by Nelson Pass, Erno Borbely, and Richard Marsh. Read all the articles by Ben Duncan in Hi- Fi News.
3. Spend 15 years modifying, repairing, and designing audio circuirts.
4. Look inside some of the best sounding amps and preamps to see what parts the best designers are using. They have also probably been through steps 1 to 3.
5. Listen, listen and listen some more.
6. Share the fruits of this lengthy labor for free on audio forums, for free because you love the hobby, and prepare to be insulted.
7. Compare notes with other people who have done steps 1 to 5.
Plan B
1. Whine and moan because it takes more than a week to learn to become an audio designer.
2. Insult graduates of plan A. Expect even more effort on your behalf after you insults, maybe even an apology!
3. Refuse to use www.altavista.com search engine with cryptic entries like "Erno Borbely", "Nelson Pass" and "DIY amplifiers".
4. Ignor links posted in other forums on diyAudio.com and on resources at diyaudio.com
5. Expect sympathy from other forum members who want to learn and build good sounding audio equipment because you were not spoon fed and didn't want to put any effort into learning.
H.H.
Hmmmmm....... I will try again. Here's how to determine which transistors are good for audio.
Plan A.
1. Get a BSEE in Electrical Engineering or a B.S. in Physics.
2, Read everything that you can get your hands on about audio circuit design. This would include everything written in The Audio
Amatuer and Audio Electronics. This would specifically include articles by Nelson Pass, Erno Borbely, and Richard Marsh. Read all the articles by Ben Duncan in Hi- Fi News.
3. Spend 15 years modifying, repairing, and designing audio circuirts.
4. Look inside some of the best sounding amps and preamps to see what parts the best designers are using. They have also probably been through steps 1 to 3.
5. Listen, listen and listen some more.
6. Share the fruits of this lengthy labor for free on audio forums, for free because you love the hobby, and prepare to be insulted.
7. Compare notes with other people who have done steps 1 to 5.
Plan B
1. Whine and moan because it takes more than a week to learn to become an audio designer.
2. Insult graduates of plan A. Expect even more effort on your behalf after you insults, maybe even an apology!
3. Refuse to use www.altavista.com search engine with cryptic entries like "Erno Borbely", "Nelson Pass" and "DIY amplifiers".
4. Ignor links posted in other forums on diyAudio.com and on resources at diyaudio.com
5. Expect sympathy from other forum members who want to learn and build good sounding audio equipment because you were not spoon fed and didn't want to put any effort into learning.
H.H.
unprocessable knowledge ...
Hello Harry and SteveG,
yes, agreed, but we rarely know why the designer chose a particular component. It his design and his sonic choice for this location and selection sometimes happened subconsciously, based on practical experience. Unfortunaltely you may have different prefences, Steve, and so you must build your own experience database. I am afraid, your question is an unanswerable one except by yourself or by an experineced buddy demonstrating sonic changes influenced by component choice.
One of my friends , a real SS guru, knows quite a row of transistors by their sonic footprints in differing circuit situations. Being almost a neophyte in respects of SS amp design, i often ask him what to choose and he can tell me which transistor in which circuit will suit my expectations or his expectations, but he cannot explain why, just report experineces. And i have to decide then whether i like it. He certainly selects devices suiting the needs of the circuit, voltage rating, SOA, max.current, temperature coefficient, beta, whatever. But he lately admitted to me that he sometimes uses stuff bad suited but sounding right and that he cannot tell why except saying "sonics".
But, Steve, i can understand you get lost in following Harry's advice, it is not helpful to get confronted with redundant advanced knowledge and the not having the "why" questions explained; this is frustrating.
I get lost, too.
To learn from a master requires you to stay in his close proximity.
But i disgress.
Harry, methinks your hint is not of much help except to those few who already have a considerable experience-knowledge base. It rather is maintaining the huge distance between pupil and master. Wouldn't it have helped more, if you posted e.g.
"2sk147 has that'n'that sonic footprint when used at X mA quiescent current in a longtailed pair. It also adds no SS-like coloration when used as lower system together with an ECC88 in cascode."
"2sk43 makes an almost sonically invisible gain=10 amp (wonderful MC-headamp) when used in single-ended topology with 3.5 mA and 2k as drain resistor but please use NiCD battery as supply." ?
Methinks the lists posted by Harry and Jocko are a very good start but near to useless for learners w/o additional "environmental" like in my example.
Steve went angry and i didn not even consider your post offensive on first glance, maybe the "no brainer remark" was unnecessary. But on a 2nd glance, i can understand why Steve went angry, he did ask and got answers in return unprocessable to him. I observed this very often in seminars, make people swallow knowledge they cannot process and they get anrgry if not violent, no, the do not leave, the get violent.
On some fields i have some competence, on others i do not even know what i do not yet know, so i read heaps of books. If a book forces me to swallow unprocessable knowledge, giving obscure, not easily trackable references, incomplete explanations, equation sequences where i get lost in tracking the algebraic simplifications because toooo many lines in between were skipped, i trash this book (even if the author is Barkhausen, Self, Giacoletto, Jones, Tietze-Schenk or some other demi-god) and am proud of myself doing this as i eliminated something preventing me from progressing. Same topic as above, think about it.
So, please folks, let's be friendly and helpful and cooperative again and launch another try
Hello Harry and SteveG,
yes, agreed, but we rarely know why the designer chose a particular component. It his design and his sonic choice for this location and selection sometimes happened subconsciously, based on practical experience. Unfortunaltely you may have different prefences, Steve, and so you must build your own experience database. I am afraid, your question is an unanswerable one except by yourself or by an experineced buddy demonstrating sonic changes influenced by component choice.
One of my friends , a real SS guru, knows quite a row of transistors by their sonic footprints in differing circuit situations. Being almost a neophyte in respects of SS amp design, i often ask him what to choose and he can tell me which transistor in which circuit will suit my expectations or his expectations, but he cannot explain why, just report experineces. And i have to decide then whether i like it. He certainly selects devices suiting the needs of the circuit, voltage rating, SOA, max.current, temperature coefficient, beta, whatever. But he lately admitted to me that he sometimes uses stuff bad suited but sounding right and that he cannot tell why except saying "sonics".
But, Steve, i can understand you get lost in following Harry's advice, it is not helpful to get confronted with redundant advanced knowledge and the not having the "why" questions explained; this is frustrating.
I get lost, too.
To learn from a master requires you to stay in his close proximity.
But i disgress.
Harry, methinks your hint is not of much help except to those few who already have a considerable experience-knowledge base. It rather is maintaining the huge distance between pupil and master. Wouldn't it have helped more, if you posted e.g.
"2sk147 has that'n'that sonic footprint when used at X mA quiescent current in a longtailed pair. It also adds no SS-like coloration when used as lower system together with an ECC88 in cascode."
"2sk43 makes an almost sonically invisible gain=10 amp (wonderful MC-headamp) when used in single-ended topology with 3.5 mA and 2k as drain resistor but please use NiCD battery as supply." ?
Methinks the lists posted by Harry and Jocko are a very good start but near to useless for learners w/o additional "environmental" like in my example.
Steve went angry and i didn not even consider your post offensive on first glance, maybe the "no brainer remark" was unnecessary. But on a 2nd glance, i can understand why Steve went angry, he did ask and got answers in return unprocessable to him. I observed this very often in seminars, make people swallow knowledge they cannot process and they get anrgry if not violent, no, the do not leave, the get violent.
On some fields i have some competence, on others i do not even know what i do not yet know, so i read heaps of books. If a book forces me to swallow unprocessable knowledge, giving obscure, not easily trackable references, incomplete explanations, equation sequences where i get lost in tracking the algebraic simplifications because toooo many lines in between were skipped, i trash this book (even if the author is Barkhausen, Self, Giacoletto, Jones, Tietze-Schenk or some other demi-god) and am proud of myself doing this as i eliminated something preventing me from progressing. Same topic as above, think about it.
So, please folks, let's be friendly and helpful and cooperative again and launch another try
Damon,
Thanks for the sane reply. I think you understand what I am after. Unfortunately, some people read everything through tinted glasses, and instead of volunteering helpful information to admitted newcomers to this hobby/profession, they use their extensive knowledge to get off on being a DIY guru. It's a sort of mental masturbation, I think.
I was fortunate in that I have been able to get an entry level position in electrical design, and I have learned a lot and have access to a lot of info and great minds. My description of parts selection being arbitrary comes from the fact that all the engineers that I work with tend to have their favorite parts, and they use them because that is what they are familiar with. I am trying to get to the core of the issue and understand why certain parts are better in certain circuits, and how the great ones like Pass and Borbely choose their components. I think that is what sets the real trend-setters apart from the average follower. Sure there is a lot of info on the net, and I constantly sort through it and read everything I can get my hands on. I find much of it to be re-hashed opinions with little new information. It is great to study other people's designs, but I want more!
HH,
Not everyone is able to go get that EE degree... sometimes, as Damon has found out, life gives you a kick in the teeth, and you have to dig your way out, sometimes over a period of years. That was me after high school- I won't go into the details, but things were not right for me to go to school. I had to find a job right away to support myself.
I will admit it: I use this board to gain knowledge from those with more experience, at the risk of bothering some who are easily frustrated with "newbie" questions. I will admit that I don't usually give back as much as I take in.I thought that was what this was all about! I appreciate all the open-minded thought here. Obviously knowledge and wisdom are two different things, as you seem to have much knowledge, and are totally lacking in wisdom. No one with any character would react the way you have in the previous posts to someone who is seeking an honest answer. Your attitude is an insult to those who visit this board.
Thanks for the sane reply. I think you understand what I am after. Unfortunately, some people read everything through tinted glasses, and instead of volunteering helpful information to admitted newcomers to this hobby/profession, they use their extensive knowledge to get off on being a DIY guru. It's a sort of mental masturbation, I think.
I was fortunate in that I have been able to get an entry level position in electrical design, and I have learned a lot and have access to a lot of info and great minds. My description of parts selection being arbitrary comes from the fact that all the engineers that I work with tend to have their favorite parts, and they use them because that is what they are familiar with. I am trying to get to the core of the issue and understand why certain parts are better in certain circuits, and how the great ones like Pass and Borbely choose their components. I think that is what sets the real trend-setters apart from the average follower. Sure there is a lot of info on the net, and I constantly sort through it and read everything I can get my hands on. I find much of it to be re-hashed opinions with little new information. It is great to study other people's designs, but I want more!
HH,
Not everyone is able to go get that EE degree... sometimes, as Damon has found out, life gives you a kick in the teeth, and you have to dig your way out, sometimes over a period of years. That was me after high school- I won't go into the details, but things were not right for me to go to school. I had to find a job right away to support myself.
I will admit it: I use this board to gain knowledge from those with more experience, at the risk of bothering some who are easily frustrated with "newbie" questions. I will admit that I don't usually give back as much as I take in.I thought that was what this was all about! I appreciate all the open-minded thought here. Obviously knowledge and wisdom are two different things, as you seem to have much knowledge, and are totally lacking in wisdom. No one with any character would react the way you have in the previous posts to someone who is seeking an honest answer. Your attitude is an insult to those who visit this board.
Bernhard,
Thank you for summing it up and trying to make peace. I am done with my rant. You have answered my questions in a clear and understandable way- I appreciate that. So, to summarize, it's all about experience! I can accept that answer.
I don't expect anyone to write a personalized book for me explaining everything they know about transistors. I was sincerely looking for an answer as to the how and why.
As far as seeing red because of what I did not understand- I don't think that is so much it. I can accept that I don't understand some things yet. I just don't see the point in posting a message to say "go read a book or something". How is that helpful?
Well, anyway, I apologize for going off the deep end if I misunderstood the intent. Please accept my sincere apology if I misunderstood, HH.
<end rant>
Thank you for summing it up and trying to make peace. I am done with my rant. You have answered my questions in a clear and understandable way- I appreciate that. So, to summarize, it's all about experience! I can accept that answer.
I don't expect anyone to write a personalized book for me explaining everything they know about transistors. I was sincerely looking for an answer as to the how and why.
As far as seeing red because of what I did not understand- I don't think that is so much it. I can accept that I don't understand some things yet. I just don't see the point in posting a message to say "go read a book or something". How is that helpful?
Well, anyway, I apologize for going off the deep end if I misunderstood the intent. Please accept my sincere apology if I misunderstood, HH.
<end rant>
Steve:
Yes a lot of designers choose the components they are familiar with. Best buy and circuit city are full of examples designed by SOME of them. People like Erno and Nelson have a passion about them that makes them dig a little deeper and ask your question "Why". They go the extra mile in building smaller circuits and running a bunch of different ones in order to obtain data for comparison and evaluate by listening. This is tedious and time consuming and bores me to death, but they enjoy it. Hence the passion. Four or five years ago Audio Amateur got me into this hobby but I didn't have any place like this to go for answers. So I had to find out for myself (god I bought a lot of books). Occasionally I would call Nelson or Erno for an answer and was surprised to find them willing to take the time to help. Not wanting to become a pest I kept those calls to a minimum. This is a great board with a lot of info and it is very easy to go off the deep end. The only thing that sets this forum apart is civility towards each other. Always bear in mind not everyone does background on a post before they let fly with their reply. I have seen some members here for a month or two wind up in some conversations that would make one think they did this for a living when I know they could'nt possibly have picked up that much. When we start asking deep questions or using correct terminology a certain level becomes expected of you. Sometimes we must be careful how much we ask of those who know, and slow it down a little bit. DIY isn't going anywhere, and I'm still learning too just like you.
Yes a lot of designers choose the components they are familiar with. Best buy and circuit city are full of examples designed by SOME of them. People like Erno and Nelson have a passion about them that makes them dig a little deeper and ask your question "Why". They go the extra mile in building smaller circuits and running a bunch of different ones in order to obtain data for comparison and evaluate by listening. This is tedious and time consuming and bores me to death, but they enjoy it. Hence the passion. Four or five years ago Audio Amateur got me into this hobby but I didn't have any place like this to go for answers. So I had to find out for myself (god I bought a lot of books). Occasionally I would call Nelson or Erno for an answer and was surprised to find them willing to take the time to help. Not wanting to become a pest I kept those calls to a minimum. This is a great board with a lot of info and it is very easy to go off the deep end. The only thing that sets this forum apart is civility towards each other. Always bear in mind not everyone does background on a post before they let fly with their reply. I have seen some members here for a month or two wind up in some conversations that would make one think they did this for a living when I know they could'nt possibly have picked up that much. When we start asking deep questions or using correct terminology a certain level becomes expected of you. Sometimes we must be careful how much we ask of those who know, and slow it down a little bit. DIY isn't going anywhere, and I'm still learning too just like you.
Parts Selection
I have been accused of expecting people to read between the lines before so I will try to be more explicit this time.
1. I don't know all the answers. Never did, never will.
2. The transistors I have listed are mainly ones familiar to me and the core of designers I work with. Trying to pick a transistor by specification alone will not insure a good design. I listen. I talk to other people who design. I listen to other people's designs.
3. The Toshiba transistors sound good. They make low noise and high transconductance jfets that no one elses parts match in my experience. I also have been happy with the Zetex bipolar and mosfet parts. I believe this two companies have excellent semicondutor fabrication plants and good designs. I have heard many amps that use them that sound good.
4. Read the articles by Nelson Pass and Erno Borbely go into how some of these parts measure in particular mosfets and jfets.
5. I don't think this stuff is quite as cut and dried as people think. If it was, there wouldn't be so many bad sounding circuits out there. The good circuits I have heard are from realitively simple topologies using conservative amounts of feedback and the best
semicondutors and passive parts that the designer could aford at the price level for his product. Power supplies are as important as the "signal path", maybe more so in some designs
6.The main point is that I wanted to make is that there is lot of imformation out there and sorting through it is based in a large part on personal experience. Go build the DIY circuits. Form clubs and share info with other hobbiest. Audition different parts. Listen!
7. I only get annoyed because I have done a tremendous amount of work to get to where I am and what I know and am stiil learning. Maybe as much as I did getting my BSEE. (which was a bitch for me and not easy!) Most of the research was before the ease of the web by the way. I don't throw this out there to get my ego reinforced but because of my passion for the art. When you get far far enough into it it becomes much more than a hobby and is very satisfying and frustrating.
I once read in a book where someone asked how to become an art critic. The answer " Go look at a million paintings." My advice is to go listen to as many circuits, parts, semiconductors as you can.
It is the jorney and not the desination that is important.
H.H.
I have been accused of expecting people to read between the lines before so I will try to be more explicit this time.
1. I don't know all the answers. Never did, never will.
2. The transistors I have listed are mainly ones familiar to me and the core of designers I work with. Trying to pick a transistor by specification alone will not insure a good design. I listen. I talk to other people who design. I listen to other people's designs.
3. The Toshiba transistors sound good. They make low noise and high transconductance jfets that no one elses parts match in my experience. I also have been happy with the Zetex bipolar and mosfet parts. I believe this two companies have excellent semicondutor fabrication plants and good designs. I have heard many amps that use them that sound good.
4. Read the articles by Nelson Pass and Erno Borbely go into how some of these parts measure in particular mosfets and jfets.
5. I don't think this stuff is quite as cut and dried as people think. If it was, there wouldn't be so many bad sounding circuits out there. The good circuits I have heard are from realitively simple topologies using conservative amounts of feedback and the best
semicondutors and passive parts that the designer could aford at the price level for his product. Power supplies are as important as the "signal path", maybe more so in some designs
6.The main point is that I wanted to make is that there is lot of imformation out there and sorting through it is based in a large part on personal experience. Go build the DIY circuits. Form clubs and share info with other hobbiest. Audition different parts. Listen!
7. I only get annoyed because I have done a tremendous amount of work to get to where I am and what I know and am stiil learning. Maybe as much as I did getting my BSEE. (which was a bitch for me and not easy!) Most of the research was before the ease of the web by the way. I don't throw this out there to get my ego reinforced but because of my passion for the art. When you get far far enough into it it becomes much more than a hobby and is very satisfying and frustrating.
I once read in a book where someone asked how to become an art critic. The answer " Go look at a million paintings." My advice is to go listen to as many circuits, parts, semiconductors as you can.
It is the jorney and not the desination that is important.
H.H.
Steve,
I can add a few thoughts about choosing transistors. I would suggest the most important thing is to understand the circuit surrounding the device and really get it clear in your head what the important characterisitcs are that the device should exhibit. Not device is ideal so you have to get clear in your mind which compromises are the least hurtful. This is the hardest thing.
One way you can do this is on paper if you are good at reasoning and have rules of thumb from experience. Another is to use a simulator and create a circuit with ideal transistors (variable current sources) and then degrade them in realistic ways and see which degradation the circuit is most sensitive to. Another is to build parts of the circuit and try different devices and try to correlate measurements with device differences published in the datasheets. All approaches require determination and perseverance.
The most likely parameters that will matter to sonic performance in semiconductors are the dc transfer characteristic (eg: Ic vs Ib vs Vbe OR Id vs Vgs vs Vds), capacitances (Cbe vs Ic, Ccb vs Vcb OR Cgs vs Id, Cdg vs Vdg) and Ft. The circuitry surrounding the device will guide you as to which are most sensitive and what ranges are acceptable. For example, Cdg is not so important if you are using a cascode topology. I don't think device noise matters much in power amps; it matters a lot in phono stages of pre-amps.
Once you know what balance of characteristics you are seeking you then have the difficult job of searching thorough datasheets. Even on the web the mfrs only provide very noddy search facilities and this can be frustrating.
Clearly an easier method is to copy the devices that others have used as Harry suggested. This presupposes that they had the same design goals and circuit requirements as you and that they chose well. Don't take this for granted.
Personally, I don't judge devices by their covers - plastic, metal, TO3, TO220 whatever. I cannot definitively recommend parts because it depends what your circuit needs. I often use AD SSM2210 matched npns for LTPs but your circuit may prefer a couple of discrete npns to eliminate the collector to collector capacitance caused by the shared die. I have found MJ15022/23 to have low distortion in push-pull configuration, but they have fairly high capacitances and are not so suitable to high feedback topologies. I can say that in push-pull designs the choice of output devices is likely to dominate the sound. At the end of the day there are lots of choices and you must decide.
Finally, please don't become superstitious. Electrons laugh in the faces of the superstitious. It's all engineering and there are real reasons for everything you hear and the "art" is not so black for those who understand the engineering.
BAM
I can add a few thoughts about choosing transistors. I would suggest the most important thing is to understand the circuit surrounding the device and really get it clear in your head what the important characterisitcs are that the device should exhibit. Not device is ideal so you have to get clear in your mind which compromises are the least hurtful. This is the hardest thing.
One way you can do this is on paper if you are good at reasoning and have rules of thumb from experience. Another is to use a simulator and create a circuit with ideal transistors (variable current sources) and then degrade them in realistic ways and see which degradation the circuit is most sensitive to. Another is to build parts of the circuit and try different devices and try to correlate measurements with device differences published in the datasheets. All approaches require determination and perseverance.
The most likely parameters that will matter to sonic performance in semiconductors are the dc transfer characteristic (eg: Ic vs Ib vs Vbe OR Id vs Vgs vs Vds), capacitances (Cbe vs Ic, Ccb vs Vcb OR Cgs vs Id, Cdg vs Vdg) and Ft. The circuitry surrounding the device will guide you as to which are most sensitive and what ranges are acceptable. For example, Cdg is not so important if you are using a cascode topology. I don't think device noise matters much in power amps; it matters a lot in phono stages of pre-amps.
Once you know what balance of characteristics you are seeking you then have the difficult job of searching thorough datasheets. Even on the web the mfrs only provide very noddy search facilities and this can be frustrating.
Clearly an easier method is to copy the devices that others have used as Harry suggested. This presupposes that they had the same design goals and circuit requirements as you and that they chose well. Don't take this for granted.
Personally, I don't judge devices by their covers - plastic, metal, TO3, TO220 whatever. I cannot definitively recommend parts because it depends what your circuit needs. I often use AD SSM2210 matched npns for LTPs but your circuit may prefer a couple of discrete npns to eliminate the collector to collector capacitance caused by the shared die. I have found MJ15022/23 to have low distortion in push-pull configuration, but they have fairly high capacitances and are not so suitable to high feedback topologies. I can say that in push-pull designs the choice of output devices is likely to dominate the sound. At the end of the day there are lots of choices and you must decide.
Finally, please don't become superstitious. Electrons laugh in the faces of the superstitious. It's all engineering and there are real reasons for everything you hear and the "art" is not so black for those who understand the engineering.
BAM
More pedigreed audio tansistors
I anm starting to feel like a gossip columnist for transistors....
Transistors (bipolar small signal and drivers)
ZTX384
ZTX214
ZTX450
ZTX214
2SC1627A
2SC2547E
2SC1844
2SC3381
2SC2911
2SC3601
2SC2591
2SC2336
2SC4793
2SA817A
2SA1349
2SA1209
2SA1407
2SA1111
2SA1837
2SA1085E
Fets
2SK246BL
2SJ103BL
2SK216
2SJ79
Regulators
LT1086CT
LT1033CT
Spice models
http://power.teipat.gr/download/OrCad Libraries/Library for Capture/PSPICE/
http://power.teipat.gr/download/OrCad Libraries/Library for Capture/PSPICE/jbipolar.lib
H.H.
I anm starting to feel like a gossip columnist for transistors....
Transistors (bipolar small signal and drivers)
ZTX384
ZTX214
ZTX450
ZTX214
2SC1627A
2SC2547E
2SC1844
2SC3381
2SC2911
2SC3601
2SC2591
2SC2336
2SC4793
2SA817A
2SA1349
2SA1209
2SA1407
2SA1111
2SA1837
2SA1085E
Fets
2SK246BL
2SJ103BL
2SK216
2SJ79
Regulators
LT1086CT
LT1033CT
Spice models
http://power.teipat.gr/download/OrCad Libraries/Library for Capture/PSPICE/
http://power.teipat.gr/download/OrCad Libraries/Library for Capture/PSPICE/jbipolar.lib
H.H.
Rumors of my demise.......
I get the flu, and look what happens.........
Now I know how Samuel Clements must have felt.
SteveG:
I don't know that I can explain to your satisfaction why a certain device has properties that I think are good, especially the physics behind it. The original point wasn't what transistors I think will sound good, but which ones I KNOW will NOT sound good.
Basically a 2N****.
I have to disagree with Hairy Holler somewhat. A degree in engineering will do very little to help you build good sounding audio gear.
Engineering school is basically boot camp for engineers. Plain and simple. They throw as much crap at you in 4 years to see who has what it takes to get even more (but different crap) thrown at you for the next 40 years. If anything, it is a hinderance to building audio stuff. You have to clean your head out of all the debris they fill it with. Maybe that is why 2 of the best known designers to people here are both physicists.
There is no substitute for experience. You will learn more by doing, even if it is just perusing other stuff than you will in boot camp. A famous audio designer friend (who I don't think will allow me to quote him) likes to say: "Steal the best, invent the rest." Engineers do not design in a vacuum; they are constantly taking bits and pieces of things they have accumulated over the years and turning them into something different.
And as for this nonsense on how life kicks you in the teeth, or wherever, and how hard:
I don't think anyone wants to get in a contest with me about that. You will lose.
As Joe Bob Briggs, the Drive-in Movie Critic of Grapevine, TX likes to say:
"I don't want to have to 'splain this to you again." [joke]
Jocko
I get the flu, and look what happens.........
Now I know how Samuel Clements must have felt.
SteveG:
I don't know that I can explain to your satisfaction why a certain device has properties that I think are good, especially the physics behind it. The original point wasn't what transistors I think will sound good, but which ones I KNOW will NOT sound good.
Basically a 2N****.
I have to disagree with Hairy Holler somewhat. A degree in engineering will do very little to help you build good sounding audio gear.
Engineering school is basically boot camp for engineers. Plain and simple. They throw as much crap at you in 4 years to see who has what it takes to get even more (but different crap) thrown at you for the next 40 years. If anything, it is a hinderance to building audio stuff. You have to clean your head out of all the debris they fill it with. Maybe that is why 2 of the best known designers to people here are both physicists.
There is no substitute for experience. You will learn more by doing, even if it is just perusing other stuff than you will in boot camp. A famous audio designer friend (who I don't think will allow me to quote him) likes to say: "Steal the best, invent the rest." Engineers do not design in a vacuum; they are constantly taking bits and pieces of things they have accumulated over the years and turning them into something different.
And as for this nonsense on how life kicks you in the teeth, or wherever, and how hard:
I don't think anyone wants to get in a contest with me about that. You will lose.
As Joe Bob Briggs, the Drive-in Movie Critic of Grapevine, TX likes to say:
"I don't want to have to 'splain this to you again." [joke]
Jocko
Jocko
I am glad to see Jocko is still with us. Okay.....you can omit part 1. from plan A. I will confess that both Jocko and I have BSEEs. It doesn't hurt anything as long as you listen and are willing to not let theory dominate things. Still....knowing how to design and measure things sure doesn't hurt when it comes to building circuits that don't misbehave at R.F. or blow up. I have been trying to provide a gentle hint (with a slegehammer sometimes) at some of the work done by some talented individuals that is available online. Steal the rest? Well, only the best.
H.H.
I am glad to see Jocko is still with us. Okay.....you can omit part 1. from plan A. I will confess that both Jocko and I have BSEEs. It doesn't hurt anything as long as you listen and are willing to not let theory dominate things. Still....knowing how to design and measure things sure doesn't hurt when it comes to building circuits that don't misbehave at R.F. or blow up. I have been trying to provide a gentle hint (with a slegehammer sometimes) at some of the work done by some talented individuals that is available online. Steal the rest? Well, only the best.
H.H.
"You have to clean your head out of all the debris they fill it with. Maybe that is why 2 of the best known designers to people here are both physicists. "
I suppose if you are saying that physicists are more open-minded because they graduate having learned very little that is applicable to audio design then I suppose I could agree.
These sort of generalisations are not helpful nor informative. There are people who can think and those who are challenged in all disciplines and besides it also depends on the content of your degree and the quality of the teachers.
BAM
I suppose if you are saying that physicists are more open-minded because they graduate having learned very little that is applicable to audio design then I suppose I could agree.
These sort of generalisations are not helpful nor informative. There are people who can think and those who are challenged in all disciplines and besides it also depends on the content of your degree and the quality of the teachers.
BAM
Hut, two, three, four.......
I don't think physicists are necessarily more open-minded. Just a lot smarter. They have to actually understand the entire realm of physics, not just a small niche that goofballs like me and Hairy Holler have apparently figured out.
Speaking of physicists.......read Bob Pease's articles in the online version of Electronics Design magazine. (No, I don't know the URL....)
Buy and read his book "Troubleshooting Analog Circuits". It belongs in everyones library. Borrow it if you can't buy it.
More useful stuff than you will get in 4 years at any US engineering bootcamp.
I don't think physicists are necessarily more open-minded. Just a lot smarter. They have to actually understand the entire realm of physics, not just a small niche that goofballs like me and Hairy Holler have apparently figured out.
Speaking of physicists.......read Bob Pease's articles in the online version of Electronics Design magazine. (No, I don't know the URL....)
Buy and read his book "Troubleshooting Analog Circuits". It belongs in everyones library. Borrow it if you can't buy it.
More useful stuff than you will get in 4 years at any US engineering bootcamp.
Just to throw my 2 cents in here, I'm a little disposed to object to the statement that you won't learn anything from any US engineering "bootcamp". Believe it or not, there are actually schools still left which aren't bootcamps for EEs....I know mostly because I go to one, and those who say you don't need any theory to design good amps will never have the satisfaction of doing anything great that's really their own. You CANNOT (almost by definition) get very far just by copying/modifying others designs and to do anything else you need to know what's going on.
Try complete and utter disdain. Bob's work has had an influence on my development as an engineer and is probably a big part of the reason why I usually end up on the objectivist side of arguments. Everything I have read of his that I understand is totally right on the money. I reread his book every year or so and devour his columns every month.
I tend to give his opinions the benefit of the doubt plus probably a little bit more. Heck, sometimes I just push my I-Believe button like everyone else and go on.
I spent a day with him at a design conference last year and really loved it. I love the way he derives circuits in his columns too. It is such fun to watch a gifted analog designer iterate into a good solution to a tricky problem.
Bob is quite an iconoclast. There are few things he likes better than bursting somebody's bubble, especially when it's full of hot air. If you want a some real heat, just get him started the quality programs corporate leadership likes to inflict on the troops.
Peace,
Phil
I tend to give his opinions the benefit of the doubt plus probably a little bit more. Heck, sometimes I just push my I-Believe button like everyone else and go on.
I spent a day with him at a design conference last year and really loved it. I love the way he derives circuits in his columns too. It is such fun to watch a gifted analog designer iterate into a good solution to a tricky problem.
Bob is quite an iconoclast. There are few things he likes better than bursting somebody's bubble, especially when it's full of hot air. If you want a some real heat, just get him started the quality programs corporate leadership likes to inflict on the troops.
Peace,
Phil
Nelson is right......Bob has no patience for the weirdness in subjective audio. But he is brilliant, and his insights are a lesson to all. Get the book, read it, then come back and tell us what, and how much, you learned.
I'm glad someone is enrolled in a good university. Although some of the worst engineers I have worked with were from GA Tech and MIT. True story, no joke.
Jocko
I'm glad someone is enrolled in a good university. Although some of the worst engineers I have worked with were from GA Tech and MIT. True story, no joke.
Jocko
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.