well, i doubt this will help at all...
i tried taking pictures of my textbook with a digital camera. doesn't work all that well. anyway, you can take a look:
http://www.dorkus.org/junk/book.html
you're probably better off just reading the Pass document hp suggests.
incidentally, this is from Sedra/Smith, Microelectronic Circuits, 4th edition, Oxford University Press... the book is ok, it is thorough but the presentation is not all that great and towards the end it focuses more on implementations relevant to IC and VLSI, rather than absolute performance.
i tried taking pictures of my textbook with a digital camera. doesn't work all that well. anyway, you can take a look:
http://www.dorkus.org/junk/book.html
you're probably better off just reading the Pass document hp suggests.
incidentally, this is from Sedra/Smith, Microelectronic Circuits, 4th edition, Oxford University Press... the book is ok, it is thorough but the presentation is not all that great and towards the end it focuses more on implementations relevant to IC and VLSI, rather than absolute performance.
I thank you (my wife won't because...) I'll read it right now! Any other references while I'm collecting reading material?
HPotter said:
Thanks dorkus, those pics turned out nicely.
HH,
I assume you are willing to try folded cascode in which case i will answer Aye.
HH,
I assume you are willing to try folded cascode in which case i will answer Aye.
Runnin' Borbely Balanced out
Most of Borbely's line stages and buffers can indeed run bal in/out and or convert bal/se. He uses 2 channels to do it with the kind of goofy feedback/input lash-up. It seems to work well enough and I guess it's good marketing if you're selling kits, you get to sell twice as much. Last year I wired up some of his buffer boards as balanced in/out, took 4 boards for 2 channels.
I can not remember the details as it was before I had fairly major surgery last year and the whole year is kinda fuzzy. 😀
Most of Borbely's line stages and buffers can indeed run bal in/out and or convert bal/se. He uses 2 channels to do it with the kind of goofy feedback/input lash-up. It seems to work well enough and I guess it's good marketing if you're selling kits, you get to sell twice as much. Last year I wired up some of his buffer boards as balanced in/out, took 4 boards for 2 channels.
I can not remember the details as it was before I had fairly major surgery last year and the whole year is kinda fuzzy. 😀
Marc,
Those pages are kinda wavey I might be getting sea sick.😛 Thanks for the info.
Harry,
A big Aye from me but I reserve the right to sub mosfets for the differential pair. Please post another picture of yourself I have a new set of horns for you........heh, heh.
Jam
Those pages are kinda wavey I might be getting sea sick.😛 Thanks for the info.
Harry,
A big Aye from me but I reserve the right to sub mosfets for the differential pair. Please post another picture of yourself I have a new set of horns for you........heh, heh.
Jam
High Point Blues!
Arr..... Scottie is in the house trying to complicate matters. We will have to send Harry and Jocko up there to fix him.
Jam
Arr..... Scottie is in the house trying to complicate matters. We will have to send Harry and Jocko up there to fix him.
Jam
Re: well, i doubt this will help at all...
Ahhhh,
Thanks. My Sedra and Smith is only 2nd edition (no jokes about how old I am 🙂 ), no mention of folded or complementary cascode.
I think it's a good book. I also like "Handbook of Analog Circuit Design" by Dennis L. Feucht of Tektronix fame.
mlloyd1
Ahhhh,
Thanks. My Sedra and Smith is only 2nd edition (no jokes about how old I am 🙂 ), no mention of folded or complementary cascode.
I think it's a good book. I also like "Handbook of Analog Circuit Design" by Dennis L. Feucht of Tektronix fame.
mlloyd1
dorkus said:
HH,
Folded cascode also reduces the available output voltage swing when compared to standard cascodes so increasing the voltage is required if you are to maintain a given voltage swing
Folded cascode also reduces the available output voltage swing when compared to standard cascodes so increasing the voltage is required if you are to maintain a given voltage swing
Can I stick tubes in my preamp?
Of course I can. Wouldn't need no folded cascodes.. and Harry Jocko and Jam would surely talk trash about me and really hurt my feelings.
Harry Jocko Jam you go girls! Time for big nap!
Of course I can. Wouldn't need no folded cascodes.. and Harry Jocko and Jam would surely talk trash about me and really hurt my feelings.
Harry Jocko Jam you go girls! Time for big nap!
Hello,
I just spend one hour to read this topic...very interesting.
The idea of a six (or maybe more) channels preamp is great !!
You want to make a project with several actors and it can be a good sinergy.
I say it CAN because the way you are facing problems is a DIYers way !!
You are now discussing of the gain stage. Well, it must be a discussion.
But as some of you pointed it can be discussed to the end of time.
The project is to design a preamplifier which will be very complex :
- Big switching circuit
- Microcontroller
- LCD or VFD display
- several boxes
and so on
If I can give my opinion this is a great challenge but there are so many questions you did not answer before choosing the topology of the gain stage.
I think the first thing to do is to agree on some physical and fonctionnal things.
The second points is to know who is going to design each part of the preamp.
So here are the points you must decide and agree on :
- Fonctionalities : number of inputs, number of boxes, kind of volume control...
- Mechanical aspect : how many boards, kind of front pannel (display, knob...)
- Modularity : fix design or upgreadable design, board to board connectors, physical connections between boards
- Target price
For each of this point you need to know :
- Who is able to make this section ?
- Which software will be used ?
- How many people are ready to spend time on this project ?
- How is the leader for each section : mechanical, software, schematics, board design....
I know some of you will say (or at least think) : what this frenchie is doing on our project ????
I can understand.
But I decided to send this "agressive" thread for two reasons :
- I am very interestd in this project I planned to realise for myself but it is such a huge job to design the electronic, mechanical, software, test, debug....
- I worked several years to Micromega R&D before designing telecom multiplexer so I think I can bring you my experience of this kind of project.
Finaly, I would like to participate to this project if you agree.
I have a lot of idea I can submit but I want to do it step by step and I am not interested if discret topolgy is better than IC's one...not yet.
Excuse me for my english which is not my natural language.
Bruno
I just spend one hour to read this topic...very interesting.
The idea of a six (or maybe more) channels preamp is great !!
You want to make a project with several actors and it can be a good sinergy.
I say it CAN because the way you are facing problems is a DIYers way !!
You are now discussing of the gain stage. Well, it must be a discussion.
But as some of you pointed it can be discussed to the end of time.
The project is to design a preamplifier which will be very complex :
- Big switching circuit
- Microcontroller
- LCD or VFD display
- several boxes
and so on
If I can give my opinion this is a great challenge but there are so many questions you did not answer before choosing the topology of the gain stage.
I think the first thing to do is to agree on some physical and fonctionnal things.
The second points is to know who is going to design each part of the preamp.
So here are the points you must decide and agree on :
- Fonctionalities : number of inputs, number of boxes, kind of volume control...
- Mechanical aspect : how many boards, kind of front pannel (display, knob...)
- Modularity : fix design or upgreadable design, board to board connectors, physical connections between boards
- Target price
For each of this point you need to know :
- Who is able to make this section ?
- Which software will be used ?
- How many people are ready to spend time on this project ?
- How is the leader for each section : mechanical, software, schematics, board design....
I know some of you will say (or at least think) : what this frenchie is doing on our project ????
I can understand.
But I decided to send this "agressive" thread for two reasons :
- I am very interestd in this project I planned to realise for myself but it is such a huge job to design the electronic, mechanical, software, test, debug....
- I worked several years to Micromega R&D before designing telecom multiplexer so I think I can bring you my experience of this kind of project.
Finaly, I would like to participate to this project if you agree.
I have a lot of idea I can submit but I want to do it step by step and I am not interested if discret topolgy is better than IC's one...not yet.
Excuse me for my english which is not my natural language.
Bruno
Re: Convergence?
Harry, just for clarity, did you mean that you dont want to use folded cascode or that you are happy enough for it to be there?
They dont really complicate the circuit much (relative to standard cascode) and provide some other nice advantages too (higher input common mode range and a cleaner signal path just to name 2). Only 2 major disadvantages of folded cascode is that it is inverting and that it has a more limitted output voltage swing. The first is not really a disadvantage more so something that just has to be noted; the second can be countered for by increasing the operating voltage.
HarryHaller said:
Harry, just for clarity, did you mean that you dont want to use folded cascode or that you are happy enough for it to be there?
They dont really complicate the circuit much (relative to standard cascode) and provide some other nice advantages too (higher input common mode range and a cleaner signal path just to name 2). Only 2 major disadvantages of folded cascode is that it is inverting and that it has a more limitted output voltage swing. The first is not really a disadvantage more so something that just has to be noted; the second can be countered for by increasing the operating voltage.
Bruno's the Man!
All the talk of to fold not to fold, Diff/Comp/SE
balanced has been hashed over for years on the net and it's really boring. No offense to those who thrieve on it, but Bruno's 100% right about getting a mechanical package for all the house keeping (volume/switching/remote etc..) done and 'paid for'. That's the tough part, then you can stick it in any jeweled case or plywood box and place your dream circuit of choice. There is a real possibility than no more than 3 would agree to a specific circuit, much less general topology, because that's the way it's always been in audio. Follow the cult of cache, whoever's hip at the moment. I for one would put a lot of time and effort to the in/out/remote/attenuator/display and will put time into that. That's my interest in this project. But agreeing on those parameters look iffy. Bruno your english is great, far better than my french! 😀
All the talk of to fold not to fold, Diff/Comp/SE
balanced has been hashed over for years on the net and it's really boring. No offense to those who thrieve on it, but Bruno's 100% right about getting a mechanical package for all the house keeping (volume/switching/remote etc..) done and 'paid for'. That's the tough part, then you can stick it in any jeweled case or plywood box and place your dream circuit of choice. There is a real possibility than no more than 3 would agree to a specific circuit, much less general topology, because that's the way it's always been in audio. Follow the cult of cache, whoever's hip at the moment. I for one would put a lot of time and effort to the in/out/remote/attenuator/display and will put time into that. That's my interest in this project. But agreeing on those parameters look iffy. Bruno your english is great, far better than my french! 😀
Maybe it's time to fix the topology and move on to other areas. Thhis would help to complete the design and issues such as component selection and pcb layout can be taken care of, or at least started.
Jam
Jam
I think that the topology is most important now. You cannot decide on a size of a board without knowing your circuit and components used. And don't forget that a preamp is not a set of switches, inputs, outputs, volume controllers etc.. This things are only additions to the active circuit and power supply which are the real preamp you listen to. Let's decide on final circuit so somebody could build it and we can discuss additions in a meantime.😉
Folded cascode
Harry, Mr. Freak:
We need to move this discussion to another thread, so as not to get in the way of "War and Peace". Especially since this one is shaping up as a US vs. Australia grudge match.
HPotter is on the right track, but if you guys think that you can throw the power supply in at the last moment because they are all the same , then I suggest you have a lot of learning left to do.
Remember, the idea is to build 6 channels of this stuff, with micros, and readouts, and an elaborate attenuator rig. We are looking at a big box here.
Jocko
Harry, Mr. Freak:
We need to move this discussion to another thread, so as not to get in the way of "War and Peace". Especially since this one is shaping up as a US vs. Australia grudge match.
HPotter is on the right track, but if you guys think that you can throw the power supply in at the last moment because they are all the same , then I suggest you have a lot of learning left to do.
Remember, the idea is to build 6 channels of this stuff, with micros, and readouts, and an elaborate attenuator rig. We are looking at a big box here.
Jocko
I could be wrong but it was my impression that one of the goals of this preamp was the option to swap gain stages. I understand wanting to find the ultimate gain stage, but who is the judge?
If the group could set a standard for board foot print that would allow for different topologies, set interconnect standards. Power supplies should be local regulation with raw DC supplied to the board, this helps with isolation.
It may be that one of the high count posters has the best gain stage rattling around in his brain, but maybe not. To my mind this is the route to the Ultimate Preamp.
If the group could set a standard for board foot print that would allow for different topologies, set interconnect standards. Power supplies should be local regulation with raw DC supplied to the board, this helps with isolation.
It may be that one of the high count posters has the best gain stage rattling around in his brain, but maybe not. To my mind this is the route to the Ultimate Preamp.
i don't think we're wasting time
i agree w/hp. not to say that bruno et al don't have a good point. we will indeed need to build this thing at some point, and that will require hardware. but the thing is that control circuitry, connectors, switching schemes, etc. are more of a practical matter. the actual gain stage is where the real art is, and there is a reason why it is taking so long to get anywhere with it. maybe we will never agree on anything but i think people are learning a lot in the process, so i am willing to dwell on it a little longer. it is too early to move on to the "glue" parts of the preamp yet. we should probably move off the cascode debate by next week and start trying circuits but we need to figure out the gain stage first, that is the heart of the preamp. again, i am not worried about control logic. software is software and i can work out the hardware details, digital is easy. i've already stated some of the possible hardware configurations (number of chassis etc.) so we are not totally lost in that matter.
but yes, let's try to make headway on the gain circuit so we can move on. i'm not sure we've decided yet which ones try. i nominate for prototyping the borberly complementary dc-coupled circuit (to appease me) and the SE version (to appease everyone else). it will take me at least a week or two to get all the parts and start putting it together so in the mean time we can move on to the next goody - the power supply!
cheers,
marc
p.s. can we start talking about "how" instead of "what" - i.e. how many gain stages we're using, how they're arranged with respect to the attenuator, etc. we need to figure out what attenuator we're using too. that is actually the biggest obstacle at the moment in terms of implementation. i may just break down and use the Wolfson part, but that requires using an opamp, which means i will have to use the borberly complementary topology or something. i can hear the screams of pain from everyone...
i agree w/hp. not to say that bruno et al don't have a good point. we will indeed need to build this thing at some point, and that will require hardware. but the thing is that control circuitry, connectors, switching schemes, etc. are more of a practical matter. the actual gain stage is where the real art is, and there is a reason why it is taking so long to get anywhere with it. maybe we will never agree on anything but i think people are learning a lot in the process, so i am willing to dwell on it a little longer. it is too early to move on to the "glue" parts of the preamp yet. we should probably move off the cascode debate by next week and start trying circuits but we need to figure out the gain stage first, that is the heart of the preamp. again, i am not worried about control logic. software is software and i can work out the hardware details, digital is easy. i've already stated some of the possible hardware configurations (number of chassis etc.) so we are not totally lost in that matter.
but yes, let's try to make headway on the gain circuit so we can move on. i'm not sure we've decided yet which ones try. i nominate for prototyping the borberly complementary dc-coupled circuit (to appease me) and the SE version (to appease everyone else). it will take me at least a week or two to get all the parts and start putting it together so in the mean time we can move on to the next goody - the power supply!
cheers,
marc
p.s. can we start talking about "how" instead of "what" - i.e. how many gain stages we're using, how they're arranged with respect to the attenuator, etc. we need to figure out what attenuator we're using too. that is actually the biggest obstacle at the moment in terms of implementation. i may just break down and use the Wolfson part, but that requires using an opamp, which means i will have to use the borberly complementary topology or something. i can hear the screams of pain from everyone...
Marc,
A few questions that need to be answered
a) Is the regulator on a seperate board? I hope not.
b) The answer to the first question will help us answer the second question size of the board? I hope that each channel has a seperate board.
c) By opting for two designs you are slowing down the process. I can sympathize with your romantic notions of a direct coupled design but I can assure you that Harry and I have been down that path many times before and we are speaking from years of experience. This option should be left for later. If the basic design is flexable enough we could swap out boards later and even have a folded cascode design, which would please AudioFreak no end. I have built all the topologies brought up and as much as it pains me to say this 😀 Harry and Jocko are right (don't let this get to your heads guys).
Also for your information all Nelson Pass's current preamp designs are capacitor coupled.
Jam
A few questions that need to be answered
a) Is the regulator on a seperate board? I hope not.
b) The answer to the first question will help us answer the second question size of the board? I hope that each channel has a seperate board.
c) By opting for two designs you are slowing down the process. I can sympathize with your romantic notions of a direct coupled design but I can assure you that Harry and I have been down that path many times before and we are speaking from years of experience. This option should be left for later. If the basic design is flexable enough we could swap out boards later and even have a folded cascode design, which would please AudioFreak no end. I have built all the topologies brought up and as much as it pains me to say this 😀 Harry and Jocko are right (don't let this get to your heads guys).
Also for your information all Nelson Pass's current preamp designs are capacitor coupled.
Jam
you guys don't read!!! just kidding, sort of... i know this thread has gotten very protracted so much of what was said got lost in the noise. i did have the answers to a lot of your questions very early on, as i have a pretty good idea of the "high-level" design of the preamp and hinted at it with my requirements, which i thought were pretty specific.
no, there is on-board regulation, as close to the gain circuit(s) as possible. most likely passive regulation. note that this may be a secondary regulation stage, the main (external) supply will most likely have regulation of some sort (i'm thinking choke regulation). the board can also have a good amount of storage capacitance locally, i was originally thinking a few thousand uF would be sufficient but others may want more.
yes they are separate! each channel is completely independent, i've said this at least 4 times already. (sorry if i sound irritated, nothing against you jam 🙂) the boards can be in separate chassis or can be stacked up or whatever in a single chassis. i am hoping the board will be no larger than 8" x 8" footprint and under 3" total height.
i trust you about the capacitor coupling, but i'd still like to try the other just for prototyping purposes. i'm not building the preamp yet! i'm just trying circuits out, the actual preamp will be built at a much later date only after i've ascertained the sound quality of the individual stages. i do NOT want to spend anymore time tweaking this thing after it's built, all the tweaking must be done now (except for the inevitable last-minute adjustments of course). as i originally said, this must be a finished, polished product, no more swapping and playing after i'm done because i just want to listen to my music! even if i assume i'm using the SE design i'd still like to listen to the other variants. hell i'm even going to build a buffered IC version just to see how it sounds. while i'm prototyping it we'll be busy arguing over power supply, attenuator, etc. anyway. 😛
ok, was wondering about that. that's fine, i can deal with a cap, i just wanted to make sure we weren't being lazy about the DC issue. 😉
marc
no, there is on-board regulation, as close to the gain circuit(s) as possible. most likely passive regulation. note that this may be a secondary regulation stage, the main (external) supply will most likely have regulation of some sort (i'm thinking choke regulation). the board can also have a good amount of storage capacitance locally, i was originally thinking a few thousand uF would be sufficient but others may want more.
yes they are separate! each channel is completely independent, i've said this at least 4 times already. (sorry if i sound irritated, nothing against you jam 🙂) the boards can be in separate chassis or can be stacked up or whatever in a single chassis. i am hoping the board will be no larger than 8" x 8" footprint and under 3" total height.
i trust you about the capacitor coupling, but i'd still like to try the other just for prototyping purposes. i'm not building the preamp yet! i'm just trying circuits out, the actual preamp will be built at a much later date only after i've ascertained the sound quality of the individual stages. i do NOT want to spend anymore time tweaking this thing after it's built, all the tweaking must be done now (except for the inevitable last-minute adjustments of course). as i originally said, this must be a finished, polished product, no more swapping and playing after i'm done because i just want to listen to my music! even if i assume i'm using the SE design i'd still like to listen to the other variants. hell i'm even going to build a buffered IC version just to see how it sounds. while i'm prototyping it we'll be busy arguing over power supply, attenuator, etc. anyway. 😛
ok, was wondering about that. that's fine, i can deal with a cap, i just wanted to make sure we weren't being lazy about the DC issue. 😉
marc
- Status
- Not open for further replies.