Re: Tree huggers unite!
yeah, i know a lot of people think ICs are a big compromise. i would tend to agree except i have not built enough yet to say so empirically. i am still willing to look at IC circuits, if only for comparison purposes.
haha... quite true that.
there is definitely a potential for a "too many cooks spoil the soup" fiasco here. however i'm willing to give it a shot and act as a moderator of sorts. keep in mind that in the end, i'm not going to randomly throw a bunch of different ideas together (yes that would sound like a Pioneer indeed); rather i'm looking for everyone to present their best ideas for how they think a given part of the preamp should be designed. then using my best judgement, i will pick out the most promising concepts, prototype them, then fire them up and see how they compare... almost like a design contest of sorts i suppose. who knows, in the end we may end up building something that's basically a Pass design, or a Jam design, or a HarryHaller design... it's not what we get but how we got there that i'm most interested in. well ok, i'm also interested in what we get as well, since that's what's gonna go in my system... but you get my point.
yeah, i know a lot of people think ICs are a big compromise. i would tend to agree except i have not built enough yet to say so empirically. i am still willing to look at IC circuits, if only for comparison purposes.
haha... quite true that.
there is definitely a potential for a "too many cooks spoil the soup" fiasco here. however i'm willing to give it a shot and act as a moderator of sorts. keep in mind that in the end, i'm not going to randomly throw a bunch of different ideas together (yes that would sound like a Pioneer indeed); rather i'm looking for everyone to present their best ideas for how they think a given part of the preamp should be designed. then using my best judgement, i will pick out the most promising concepts, prototype them, then fire them up and see how they compare... almost like a design contest of sorts i suppose. who knows, in the end we may end up building something that's basically a Pass design, or a Jam design, or a HarryHaller design... it's not what we get but how we got there that i'm most interested in. well ok, i'm also interested in what we get as well, since that's what's gonna go in my system... but you get my point.
Re: ok so maybe this is the dumbest idea ever, but...
maybe, but hey, i'm still willing to give it a try. at the very least i'll be able to collect some design ideas for a web page on the matter. and some people will appreciate the information this thread will provide, me included. if you're not interested, that's fine, participation is optional of course.
Jocko Homo said:
maybe, but hey, i'm still willing to give it a try. at the very least i'll be able to collect some design ideas for a web page on the matter. and some people will appreciate the information this thread will provide, me included. if you're not interested, that's fine, participation is optional of course.
ok, let's start with the attenuator.
my requirements were for something that would allow 1dB adjustment resolution over a 70dB range. and of course, it must be as sonically transparent as possible.
i am most interested in AudioFreak's reverse engineering of the Pass X volume control, which has the potential to be a great combination of control flexibility and transparency. i'm a little concerned about patent infringement though (though Mr. Pass himself told us to look into the design), and the complexity of implementation.
until we get word from audiofreak... another option is the Wolfson WM8816 IC, which has no internal active circuitry. its primary drawback is the need to be placed in the feedback loop of an opamp. this will probably leave a bad taste in the mouths of negative feedback opponents. however, as an off-the-shelf solution, it is very flexible, easy to implement, and at least allows us to use our own opamp (discrete perhaps?). i am ruling out the Crystal and BB/TI parts because they have internal opamps of questionable quality. the Maxim part is also passive, but does not have the control range i desire.
i am also willing to implement a custom design using fixed-series resistor and switched shunt resistors. i can get a very compact layout using SMT resistors (tantalum film anyone?) but the switching element is the problem. i can try to minimize the resistor/switch count by optimizing values and switching combinations of resistors in and out but i will still need fairly small switches to keep the over layout reasonable, so that may rule out mechanical relays. i could use something like a AD SSM2404 CMOS switch but the on-resistance will be an issue at the highest attenuation settings. also, most likely we will want to buffer the input to this sort of attenuator to keep a fixed load on the source, as well as keep the impedance of the switch nice and low for lowest noise figures... this was all covered in another thread: http://www.diyaudio.com/forums/showthread.php?threadid=3267
any further thoughts?
my requirements were for something that would allow 1dB adjustment resolution over a 70dB range. and of course, it must be as sonically transparent as possible.
i am most interested in AudioFreak's reverse engineering of the Pass X volume control, which has the potential to be a great combination of control flexibility and transparency. i'm a little concerned about patent infringement though (though Mr. Pass himself told us to look into the design), and the complexity of implementation.
until we get word from audiofreak... another option is the Wolfson WM8816 IC, which has no internal active circuitry. its primary drawback is the need to be placed in the feedback loop of an opamp. this will probably leave a bad taste in the mouths of negative feedback opponents. however, as an off-the-shelf solution, it is very flexible, easy to implement, and at least allows us to use our own opamp (discrete perhaps?). i am ruling out the Crystal and BB/TI parts because they have internal opamps of questionable quality. the Maxim part is also passive, but does not have the control range i desire.
i am also willing to implement a custom design using fixed-series resistor and switched shunt resistors. i can get a very compact layout using SMT resistors (tantalum film anyone?) but the switching element is the problem. i can try to minimize the resistor/switch count by optimizing values and switching combinations of resistors in and out but i will still need fairly small switches to keep the over layout reasonable, so that may rule out mechanical relays. i could use something like a AD SSM2404 CMOS switch but the on-resistance will be an issue at the highest attenuation settings. also, most likely we will want to buffer the input to this sort of attenuator to keep a fixed load on the source, as well as keep the impedance of the switch nice and low for lowest noise figures... this was all covered in another thread: http://www.diyaudio.com/forums/showthread.php?threadid=3267
any further thoughts?
Pass Labs is generally very friendly when it comes to DIYers so i doubt that they will have a problem with me doing this .... I will of course send them an email before i post the schematic here .... I'm currently tied up with EE studies but after my exams i will have a month or so that i can do this kind of thing in .... If you want to wait for the X volume control from me it'll probably be late July.
Forget about SMT's
The are at least two problems with SMT:
1. They are so small that they easily get modulated by signal, even at very low power levels (otherwise known as distortion)
2. They are not DIY friendly (as mentioned by others)
What some people try to achieve with SMT resistors in audio is a reduction of the signal path lenght. I doubt that this will be very helpful in a design such as the proposed one.
As an example:
Look at DACT attenuators. The problem of signal path can probably be fixed with ordinary resistors using LPAD instead of series connections ...
So, PLEASE use ordinary components ane make our lives a lot easier without having to resort to toothpicks and magnifying glasses to solder SMT's.
Petter
The are at least two problems with SMT:
1. They are so small that they easily get modulated by signal, even at very low power levels (otherwise known as distortion)
2. They are not DIY friendly (as mentioned by others)
What some people try to achieve with SMT resistors in audio is a reduction of the signal path lenght. I doubt that this will be very helpful in a design such as the proposed one.
As an example:
Look at DACT attenuators. The problem of signal path can probably be fixed with ordinary resistors using LPAD instead of series connections ...
So, PLEASE use ordinary components ane make our lives a lot easier without having to resort to toothpicks and magnifying glasses to solder SMT's.
Petter
Possibly because I am a Project Manager
I like the idea, and it would be fun to watch even if I never built one.
Some thoughts - its hard to build something without a really good set of requirements laid out and agreed to right out front - and the methods to track them as we agree to the changes. I know that there are some really good tools available for this process in freeware since this sort of thing is done so often in software development - so borrow a page from open source software. I wonder if Sourceforge.net would mind if we used their tools to develop a preamp? To quote thier site: SourceForge.net provides free services to Open Source developers, including project hosting, version control, bug and issue tracking, project management, backups and archives, and communication and collaboration resources. I guess we could say that there will be a digital control portion of the beast.
And finally, while I understand the humor requirement that we all have, there is a line where it becomes non-productive. So at some point, if you have nothing good to say, say nothing at all.
I like the idea, and it would be fun to watch even if I never built one.
Some thoughts - its hard to build something without a really good set of requirements laid out and agreed to right out front - and the methods to track them as we agree to the changes. I know that there are some really good tools available for this process in freeware since this sort of thing is done so often in software development - so borrow a page from open source software. I wonder if Sourceforge.net would mind if we used their tools to develop a preamp? To quote thier site: SourceForge.net provides free services to Open Source developers, including project hosting, version control, bug and issue tracking, project management, backups and archives, and communication and collaboration resources. I guess we could say that there will be a digital control portion of the beast.
And finally, while I understand the humor requirement that we all have, there is a line where it becomes non-productive. So at some point, if you have nothing good to say, say nothing at all.
Definitely because you are product manager
This tool is for software design.... I believe this is a hardware project for the most part.
"The basic idea behind open source is very simple: When programmers can read, redistribute, and modify the source code for a piece of software, the software evolves. People improve it, people adapt it, people fix bugs. And this can happen at a speed that, if one is used to the slow pace of conventional software development, seems astonishing."
I nominate you for software management of source code for digital volume control and to keep track of any Spice model results and configurations. Your first asignment is to form a commitee to study selection of programming language for volume control code. The second asignment is to select a Spice platform or simulation. I believe the requirements are that it have accurate distortion modeling and cost no money. Being from Washington DC, perhaps you could provide imformation on budget and schedule overuns. Perhaps we could structure this a long the lines of a goverment project.
I was going to ask to be in charge of the humor requirements, but I see that I could never compete with the subtle humor in your last post.
H.H.
Interim chairman of the Analog Design Humor Oversite Commitee
to known as A.D.H.O.C. in future references.
This tool is for software design.... I believe this is a hardware project for the most part.
"The basic idea behind open source is very simple: When programmers can read, redistribute, and modify the source code for a piece of software, the software evolves. People improve it, people adapt it, people fix bugs. And this can happen at a speed that, if one is used to the slow pace of conventional software development, seems astonishing."
I nominate you for software management of source code for digital volume control and to keep track of any Spice model results and configurations. Your first asignment is to form a commitee to study selection of programming language for volume control code. The second asignment is to select a Spice platform or simulation. I believe the requirements are that it have accurate distortion modeling and cost no money. Being from Washington DC, perhaps you could provide imformation on budget and schedule overuns. Perhaps we could structure this a long the lines of a goverment project.
I was going to ask to be in charge of the humor requirements, but I see that I could never compete with the subtle humor in your last post.
H.H.
Interim chairman of the Analog Design Humor Oversite Commitee
to known as A.D.H.O.C. in future references.
heh, sourceforge.net. Funny joke. I have used a CVS repository at work, managing vhdl source for projects, and it is quite a hassle, unless you are doing large projects.
Anyway, I would possibly be interested in participating in a pre-amp project. I was actually considering my own pre-amp project next year for my senior design project in school.
It will be interesting to see if something actually comes out of this.
--
Brian
gte619j@prism.gatech.edu
Anyway, I would possibly be interested in participating in a pre-amp project. I was actually considering my own pre-amp project next year for my senior design project in school.
It will be interesting to see if something actually comes out of this.
--
Brian
gte619j@prism.gatech.edu
Re: Re: The diyAudio.com preamp project!
hi,
there is no requirement for balanced inputs/outputs. it does not mean that we shouldn't make them, but we can assume this is an unbalanced project for now. i think balanced outs are nice, but balanced ins are definitely a luxury. do many of you have source components w/balanced outs? i certainly don't.
line level only. this is a preamp designed to integrate into a very modern system with the latest formats (multichannel SACD, DTS, etc.), for better or worse.
maximum output level is flexible, but i would say that at least 10V RMS with minimal distortion is desirable. even higher voltages would be great for added headroom but are not a requirement.
LTP or CFB... ??? sorry haven't come across those acronyms yet.
marc
AudioFreak said:
hi,
there is no requirement for balanced inputs/outputs. it does not mean that we shouldn't make them, but we can assume this is an unbalanced project for now. i think balanced outs are nice, but balanced ins are definitely a luxury. do many of you have source components w/balanced outs? i certainly don't.
line level only. this is a preamp designed to integrate into a very modern system with the latest formats (multichannel SACD, DTS, etc.), for better or worse.
maximum output level is flexible, but i would say that at least 10V RMS with minimal distortion is desirable. even higher voltages would be great for added headroom but are not a requirement.
LTP or CFB... ??? sorry haven't come across those acronyms yet.
marc
no SMT?
hmm petter, you seem pretty bent against SMT. your reasons sound, well, reasonable but i can't corroborate it myself. does anyone have any similar observations, either pro or con, about SMT technology?
i agree that if we are dealing with a lot of currents, e.g. a class A design, perhaps some components that need to dissipate a lot of power will need to be traditional form factor. i guess i was thinking more about the benefits of SMT in more complex circuits, e.g. opamps with lots of feedback where it is advantageous to keep paths (feedback, input terminals) extremely short. i believe if we were to use an IC opamp, SMT woudl definitely be the way to go. but since everyone hates ICs anyway... with a very simple topology (e.g. single-ended Pass style) the benefits of SMT are probably moot, and the drawbacks more signifcant. thoughts?
as for the attenuator, i think we will hold off on that once, and see if audiofreak can deliver the goods on the Pass X control.
hmm petter, you seem pretty bent against SMT. your reasons sound, well, reasonable but i can't corroborate it myself. does anyone have any similar observations, either pro or con, about SMT technology?
i agree that if we are dealing with a lot of currents, e.g. a class A design, perhaps some components that need to dissipate a lot of power will need to be traditional form factor. i guess i was thinking more about the benefits of SMT in more complex circuits, e.g. opamps with lots of feedback where it is advantageous to keep paths (feedback, input terminals) extremely short. i believe if we were to use an IC opamp, SMT woudl definitely be the way to go. but since everyone hates ICs anyway...
with a very simple topology (e.g. single-ended Pass style) the benefits of SMT are probably moot, and the drawbacks more signifcant. thoughts?as for the attenuator, i think we will hold off on that once, and see if audiofreak can deliver the goods on the Pass X control.
btw, we are using an Atmel AVR controller
i chose it over PIC cuz it seemed a little easier/cheaper/more flexible. that could be just me, but another one of my EE friends thought so too.
the current part is a AT90S8515, but we can always bump that up to a ATMega if things get *really* dicey...
source code, of course, will be easy to share. we'll use the web site for that. if people REALLY want i will set up a CVS repository for it. ugh.
cheers,
marc
i chose it over PIC cuz it seemed a little easier/cheaper/more flexible. that could be just me, but another one of my EE friends thought so too.
the current part is a AT90S8515, but we can always bump that up to a ATMega if things get *really* dicey...
source code, of course, will be easy to share. we'll use the web site for that. if people REALLY want i will set up a CVS repository for it. ugh.
cheers,
marc
Because I'm a software project manager
Hardware designers should be open to the ideas from the software design community (and vice-versa), so don't be too quick to naysay. When we stop being open to learning we might as well camp out on the couch and watch Three's Company reruns. [No need for a picture, Harry, I remember what Crissy looks like]
Personally, I think what you guys should be designing is a preamp test bed, where you can try different circuitry ideas. A rapid prototype approach. There are too many different ideas and philosophies here to build the perfect preamp.
p.s. Better get used to VCS (as we call it). It's your friend.
Hardware designers should be open to the ideas from the software design community (and vice-versa), so don't be too quick to naysay. When we stop being open to learning we might as well camp out on the couch and watch Three's Company reruns. [No need for a picture, Harry, I remember what Crissy looks like
]Personally, I think what you guys should be designing is a preamp test bed, where you can try different circuitry ideas. A rapid prototype approach. There are too many different ideas and philosophies here to build the perfect preamp.
p.s. Better get used to VCS (as we call it). It's your friend.
Software Management
I have managed software. Only three people writing software without a manager is usually a disaster in the making. I also had to do annoying hardware workarounds so we wouldn't have to touch the precious software. My hands shake when I hear the word software. I believe that software management is the one of the greatest oxymorons of all time. And product management....
I sat through 6 weekly teleconference meetings on a matter that could have and should have been solved with one meeting had anyone had the guts to make a decision. I will now sit back and enjoy the rest of this adventure without further detraction.
"There are too many different ideas and philosophies here to build the perfect preamp." Hmmmmm..........
H.H.
I have managed software. Only three people writing software without a manager is usually a disaster in the making. I also had to do annoying hardware workarounds so we wouldn't have to touch the precious software. My hands shake when I hear the word software. I believe that software management is the one of the greatest oxymorons of all time. And product management....
I sat through 6 weekly teleconference meetings on a matter that could have and should have been solved with one meeting had anyone had the guts to make a decision. I will now sit back and enjoy the rest of this adventure without further detraction.
"There are too many different ideas and philosophies here to build the perfect preamp." Hmmmmm..........
H.H.
Hi,
Have built a preamp with a remote. The preamp itself is a kit from a dutch company (P11 audio&techniek). What is more interesting for you is probably the controller and the software.
I am using microchips 16f84 controller. Maybe not as flexible, but that is 'solved' by using I2C ports. With this bus it is easy to hook up anything you need. Input relays are from SDS, type DS2E.
Have a look at my site, maybe there is something usefull there.
Guido
Have built a preamp with a remote. The preamp itself is a kit from a dutch company (P11 audio&techniek). What is more interesting for you is probably the controller and the software.
I am using microchips 16f84 controller. Maybe not as flexible, but that is 'solved' by using I2C ports. With this bus it is easy to hook up anything you need. Input relays are from SDS, type DS2E.
Have a look at my site, maybe there is something usefull there.
Guido
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.