Hi Evan,
You may consider me really enthusiastic, at least to the extent of helping underwrite your development costs.
Got a BOM?
RandyB
You may consider me really enthusiastic, at least to the extent of helping underwrite your development costs.
Got a BOM?
RandyB
Z33 said:
Thanks, Randy. I started working on a BOM the other day, but then got kind of distracted by work (of all things, right?). It shouldn't take too long to put it together, though. I will try to make one this weekend. I should also make clear, in case you didn't already realize, that you would need to provide a +/- DC input. However, since the board uses adjustable regs, there is quite a bit of leeway in choosing these values. I plan to use +/- 15V input and +/- 12 V for the op amp, so I will spec the BOM as such.
Evan,
I've got +-15VDC available, from the same PS I'm using for active XO, so that's not a problem. Work does encroach into serious playtime, but how else to afford the toys? I'm also out of town for a week or so around the end of the month, so not much will get done then. I don't have a lot of experience with SMT, but I've been soldering for 35 years and still have a steady hand. And access to a binocular microscope. 😉
RandyB
I've got +-15VDC available, from the same PS I'm using for active XO, so that's not a problem. Work does encroach into serious playtime, but how else to afford the toys? I'm also out of town for a week or so around the end of the month, so not much will get done then. I don't have a lot of experience with SMT, but I've been soldering for 35 years and still have a steady hand. And access to a binocular microscope. 😉
RandyB
ezkcdude said:
I think if you are going to put a lot of effort into this project you really ought to have good reasons for choices you make. What’s the purpose of parallel ground planes on the periphery of the board? What’s the purpose of ferrites on the input to the regulators? What’s the purpose of series resistors on the outputs of the CS8416? What’s the purpose of the parallel power traces? What’s the purpose of all the vias connecting the top and bottom planes? Note: These are rhetorical questions. I don’t want your answers but you should have sound technical reasons for each decision other than, “Someone at diyAudio said I should.”
If it were my project I would congratulate myself for a good first try and start again with better components, better placement, better routing, and a full ground plane on the component side where it does the most good. If I thought there was good reason for series resistors I put them where they would do the most good: The same with ferrites. I would definitely add an input transformer and accommodations for every what-if contingency I could think of, such as, pads for another DAC chip for dual mono operation with differential outputs. Once you build the board it’s too late. Yes, you can buy three new boards for $51 but what about all the components? Are you going to unsolder them from the old board and transfer them to the new?
Ulas, every data sheet I could find has those series resistors and ferrites. If someone (other than myself) wants to leave them out, they could put jumpers or 0R resistors instead. As for ground plane, not everyone believes it should be on top. In addition, I initially thought it would be very difficult, given the mini-board size, to fit all the traces and ground on the component side. So I made the decision I did.
It's easy to say don't listen to other people, try it for yourself, but everyone has to start somewhere. Would you have me do it exactly your way? You might as well make the board then, and send me one. Until that time, I'm pretty happy with the choices I made --- yes, after a lot of reading...data sheets, books, and forums. If it sounds bad, well, then I learned something, and I will try again. As a scientist, I know I learn the most from my mistakes. I don't, as a matter of personal policy, try to do everything perfect the first time. Why? Because I have found, more often than not, I end up not finishing anything that way. I had three goals for this project 1) fits ExpressPCB mini-board size 2) completely hardware-programmed and 3) makes music. If I get all three, then it will indeed be a very successful project for me.
In the meantime, not to be smarmy, but why don't you show us some of your designs?
It's easy to say don't listen to other people, try it for yourself, but everyone has to start somewhere. Would you have me do it exactly your way? You might as well make the board then, and send me one. Until that time, I'm pretty happy with the choices I made --- yes, after a lot of reading...data sheets, books, and forums. If it sounds bad, well, then I learned something, and I will try again. As a scientist, I know I learn the most from my mistakes. I don't, as a matter of personal policy, try to do everything perfect the first time. Why? Because I have found, more often than not, I end up not finishing anything that way. I had three goals for this project 1) fits ExpressPCB mini-board size 2) completely hardware-programmed and 3) makes music. If I get all three, then it will indeed be a very successful project for me.
In the meantime, not to be smarmy, but why don't you show us some of your designs?
You asked for comments so I gave you some. Now you’re annoyed.
Oh really? I don’t see any series resistors on the datasheets of the three major chips you have chosen. Only the CS8416 datasheet shows a ferrite and it’s not on the input of a regulator.
It sounds like another diyAudio Expert in the making. I guess there’s money to be made selling PCBs and kits to other diyAudiophools.
If you believe the ground plane should be on the bottom why did you fill every available square millimeter on the top with ground plane?
Who said all the traces have to be on the component side?
Good, that’s why I said, if it were my project, I would build on the experience and start again.
I did. I showed you the layout of my current project, a 4xPCM1704 OS/NOS DAC. It has many more components and was far more challenging to layout on a MiniBoard than your design and it has a full ground plane on the component side and voltage planes on the foil side. The two DAC chips per channel can be wired as singles, parallel pairs, or differential pairs. All timing comes from a synchronous counter clocked by either a local oscillator or S/PDIF. The oscillator also exports a clock to slave the CDP. Digital input can be 8x oversampled (24- or 32-bit frames) or non-oversampled (left/right channels time-aligned or staggered.) It also has eight voltage regulators and an input transformer.
Incidentally, the layout I posted is the second version. I am still not happy with the results so I am doing it again. Maybe I should have posted details of the project when I started it to get others to fund it by selling them PCBs. Nah, I do DIY to study, experiment, and learn. I don’t need help from the diyAudio Experts to design or layout a DAC. Why would I want my designs to look like everybody else’s?
Oh really? I don’t see any series resistors on the datasheets of the three major chips you have chosen. Only the CS8416 datasheet shows a ferrite and it’s not on the input of a regulator.
It sounds like another diyAudio Expert in the making. I guess there’s money to be made selling PCBs and kits to other diyAudiophools.
If you believe the ground plane should be on the bottom why did you fill every available square millimeter on the top with ground plane?
Who said all the traces have to be on the component side?
Good, that’s why I said, if it were my project, I would build on the experience and start again.
I did. I showed you the layout of my current project, a 4xPCM1704 OS/NOS DAC. It has many more components and was far more challenging to layout on a MiniBoard than your design and it has a full ground plane on the component side and voltage planes on the foil side. The two DAC chips per channel can be wired as singles, parallel pairs, or differential pairs. All timing comes from a synchronous counter clocked by either a local oscillator or S/PDIF. The oscillator also exports a clock to slave the CDP. Digital input can be 8x oversampled (24- or 32-bit frames) or non-oversampled (left/right channels time-aligned or staggered.) It also has eight voltage regulators and an input transformer.
Incidentally, the layout I posted is the second version. I am still not happy with the results so I am doing it again. Maybe I should have posted details of the project when I started it to get others to fund it by selling them PCBs. Nah, I do DIY to study, experiment, and learn. I don’t need help from the diyAudio Experts to design or layout a DAC. Why would I want my designs to look like everybody else’s?
I assume you're talking about R19-R21? If so, those are shown on p. 17 of the data sheet for CS8416 Eval board.
No argument here. You could have just said this straight up.
This is either plain silliness or blatant hostility. You really think I'm trying to make money? This is my first DAC project --- Hell, it even says so in the thread title. I explicitly said I would be willing to give people boards at cost, if they want them, and with the understanding that I can't guarantee it will work. (You do know what "at cost" means, right?)
On the advice of others. I'll simply leave it at that.
I have no idea. Did I say this?
I haven't even ordered the boards, let alone put one together, and you want me to start over?
I'm glad for you. Again, I'll refer you to the thread title.
I'm really not sure why you have this notion that I'm trying to profit from this, because I have never said that. I'm trying to get help to build my own working DAC, and if other people happen to be interested, I'll sell them the board at cost. I already said that, though. Let's leave it at that, shall we?
But you want mine to look like yours, is that it?
Hey guys, cool it. No need getting worked up - it just results in wasteful use of energy. Energy is always in short supply !
I'm being very serious when I say that all our members should do some Pranayama which is controlled deep breathing when seated in a specific posture. It helps to increase blood circulation and relax the nerves and in my opinion also helps to to get 'above' a certain stressful situation. It's very simple but its effects are quite magical.
I would have suggested meditation which put very "simply" is sitting and doing nothing - including the mind !
However it is much harder to do this than it appears!
Can you really blank out your mind for even 5 minutes with absolutely no thoughts flitting across your mind ? You should just "see" a white space around you. Of course this question does not apply to guys who already practice meditation.
I'm no expert either. I am told that the thought process gets clearer and better if one practises meditation . Any experts out there? Made a better pcb ( or other ) design after meditation ?
Cheers.
I'm being very serious when I say that all our members should do some Pranayama which is controlled deep breathing when seated in a specific posture. It helps to increase blood circulation and relax the nerves and in my opinion also helps to to get 'above' a certain stressful situation. It's very simple but its effects are quite magical.
I would have suggested meditation which put very "simply" is sitting and doing nothing - including the mind !
However it is much harder to do this than it appears!
Can you really blank out your mind for even 5 minutes with absolutely no thoughts flitting across your mind ? You should just "see" a white space around you. Of course this question does not apply to guys who already practice meditation.
I'm no expert either. I am told that the thought process gets clearer and better if one practises meditation . Any experts out there? Made a better pcb ( or other ) design after meditation ?
Cheers.
ezkcdude said:
Not at all. I think it should look like your design: A design based on your knowledge and inspiration, not a collection circuit fragments you copied from here and there without understanding their purpose.
Neither your schematic nor the first version of your layout has ferrites. Someone here says you should have them and, as if by magic, the next version has ferrites but they are in an unorthodox location. Perhaps in between versions of your PCB layout you researched the application of ferrites in digital circuits and determined that ferrites are most beneficial on the inputs of the voltage regulators. If that’s your considered design decision then you ought to stick with it. If the next version of your layout has ferrites in a different location it means you don’t have a clue and are just copying others because it’s easier than learning.
The same with series resistors. If they are beneficial on the outputs of the CS8416, why aren’t also beneficial on the outputs of the AD1896? You say you copied the idea from a Cirrus evaluation board. Did the evaluation board also have an AD1896?
By the way, I’ve studied your circuit in detail and have discovered an error I would classify as a “show stopper” but I’m sure you know what you are doing and have good reasons for your choices. After all, you are showing all the characteristics of a diyAudio Expert. I’m sure your newfound digital audio expertise will be welcomed in this forum.
But wait, there's more. I found another showstopper. I can already smell the stench of burning ICs. Where there are two there are probably more but why should I waste my time identifying the errors in your design when you reject my suggestions.
So, how do you suggest I learn? Don't I actually have to design and build something first? What's the alternative you are suggesting here? I have tried to gather all the information I can from as many different sources as possible. For a first project, I don't see how you can expect more than that.
I assume they are beneficial because they were in that data sheet, and not in the other one for AD1896. Don't the engineers for those companies have some knowledge, too? Like I said before, I can use 0-ohm resistors there if it makes an improvement. If you have differing opinions about it, why not just share them with us, instead of being so coy and snooty about it.
Look, I don't really understand why you are being so cynical and mean-spirited about the whole thing. Evidently, you have it in your head that all these DIYers are out to make money. You may be on to something, though. I think I'm going to quit my job as a postdoc, forget about applying for faculty jobs this fall, and become a full-time DIYer who makes tens of dollars off selling crappy one-off DAC's to people. Sounds like a great plan to me.
Hey ezkcdude,
You are pretty much holding up your cool..........well almost !
Ulas is bugged by something. Not something we all necessarily know about. Let him cool off.
If you say things like ".....I'm going to quit my job as a postdoc, forget about applying for faculty jobs this fall, and become a full-time DIYer who makes tens of dollars off selling crappy one-off DAC's .........." , you aren't going to help the 'cooling off process'.
Your are however on the right track. Since you are starting off with little knowledge , do it even if you don't understand it all and the outcome will make you wiser. The circuits you looked at must have worked earlier.
Don't loose your cool. You are doing great .
Cheers.
You are pretty much holding up your cool..........well almost !
Ulas is bugged by something. Not something we all necessarily know about. Let him cool off.
If you say things like ".....I'm going to quit my job as a postdoc, forget about applying for faculty jobs this fall, and become a full-time DIYer who makes tens of dollars off selling crappy one-off DAC's .........." , you aren't going to help the 'cooling off process'.
Your are however on the right track. Since you are starting off with little knowledge , do it even if you don't understand it all and the outcome will make you wiser. The circuits you looked at must have worked earlier.
Don't loose your cool. You are doing great .
Cheers.
Hi Rfbrw,
It could also be the other way round !
Cheers.
---------------------------------------------------------
If you are logged on to the DIYaudio forum and you aren't smiling, you are missing the basic reason why it exists ! So help it get better.............start smiling.
🙂
It could also be the other way round !
Cheers.
---------------------------------------------------------
If you are logged on to the DIYaudio forum and you aren't smiling, you are missing the basic reason why it exists ! So help it get better.............start smiling.
🙂
A couple more little things on the schematic.
- There seems to be a cap blocking the power on Vdd on the 1794
- I'd add reverse protection diodes to the LM317's.
- There seems to be a cap blocking the power on Vdd on the 1794
- I'd add reverse protection diodes to the LM317's.
ashok said:
At this relatively simple level, there is no real reason for floundering about in the dark in the hope of coming up with something that works. The rules are relatively straightforward and once you have it clear in your own mind there should be no problems. It is when one seeks to mix and match what others have done simply because others have done it that the magic smoke escapes.
- Status
- Not open for further replies.