My layout will have wiring as short or shorter than that...but it will use a bit of an unconventional layout. I can try to draw up a diagram at some point. Your wiring job is really great, though!
Thanks for the answers, Russ. I figured the through-holes would be around 18 guage, but I wanted to double check...and I agree, 18 guage is way more than enough, haha.
I haven't looked too closely at the TXD (really, only the photo on your TXD page). I'll look more deeply in to this, but I have a few questions: TXD is single channel, correct (left+ and - or right+ and -)? Also, I'm guessing there is no low pass filtering done.
-edit- ooh, re-imaged TXD. Will this have load balancing resistors for using the Opus in dual differential mode?
Thanks for the answers, Russ. I figured the through-holes would be around 18 guage, but I wanted to double check...and I agree, 18 guage is way more than enough, haha.
I haven't looked too closely at the TXD (really, only the photo on your TXD page). I'll look more deeply in to this, but I have a few questions: TXD is single channel, correct (left+ and - or right+ and -)? Also, I'm guessing there is no low pass filtering done.
-edit- ooh, re-imaged TXD. Will this have load balancing resistors for using the Opus in dual differential mode?
fierce_freak said:
It is a single balanced channel yes. There are 2 spots (C7 and C8) which enable optional filtering. The filter can easily be calculated once you know the feedback resistor value. I use 240pf there with 390R but you could easily go up to 1nf with the same feedback resistor (RF).
It seriously is an incredible headphone amp.
:EDIT: of course you have to rig your cans for balanced operation, but once you do you won't want anything else....:EDIT:
Cheers!
Russ
Ok, I'm thinking out case sizes for my Opus as I plan to recase it when I change the layout. The Opus TXD-inspired module will be in the same size format as the other modules? Will this take a supply like your bipolar power supply or the LCDPS? Lastly, do you think a greater (or any) improvement will be had by which of the following two options: using independent supplies for each boards power inputs, or one supply each for analog and digital sections, both supplies being extremely regulated and remarkably low ripple.
I know this is going to really be getting down to that last Nth percent of performance, but I'm willing to go to those lengths for actual improvements, however small.
I know this is going to really be getting down to that last Nth percent of performance, but I'm willing to go to those lengths for actual improvements, however small.
fierce_freak said:
1) Yes. Thats the plan.
2) Bipolar (LCBPS), but you actually can actually make the LCDPS act Bipolar if you want, but the LCBPS is a better/more practical fit.
3) Isolating the modules from each other by giving them their own PS should help at least a little. The is also true where boards have both analog and digital supplies. But I doubt this will be a huge difference. Still it won't hurt a bit. The fact that the modules each have local VREGS at the critical spots and close to the pins makes external power supplies less critical then they would be otherwise.
Cheers!
Russ
One nice thing about the WM8741 (if we can ever get them) is that it supports the native time multiplexed DSP format most DSPs use. DSP mode A and B in the WM8741 data sheet. This means the DSP does not have to convert to I2S or use an I2S peripheral.
I some point I will be playing with DSP filters and DRC.
Cheers!
Russ
I some point I will be playing with DSP filters and DRC.
Cheers!
Russ
Thanks
Russ/Brian, Thanks for the answer. Using a cheap donor receiver for the decoding is probably just as easy anyway.
I am interested in DRC and filters though, because right now, to play with FIR filters the best way now is with a PC and multichannel soundcard but that does not integrate as well.
Russ/Brian, Thanks for the answer. Using a cheap donor receiver for the decoding is probably just as easy anyway.
I am interested in DRC and filters though, because right now, to play with FIR filters the best way now is with a PC and multichannel soundcard but that does not integrate as well.
Russ White said:
Awesome, thanks Russ. I'm out of town on vacation, so I don't have access to my modules at the moment. How tall are the LCDPS and LCBPS with heatsinks installed? Also, I'm thinking of adding additional seperation between the analog and digital supplies by using something like this. Since this circuit calls for 24v toroids capable of 2a, what VA rating would I be looking at? Almost 100?
Lastly, if using a seperate supply for each board, should I be tying grounds together, or would this remove any possible benefit?
An externally hosted image should be here but it was not working when we last tested it.
I just realized as I was on the brink of assembling my cd player project that I don't understand 1 thing in the hook-up diagram. On the right where it shows connection for unbalanced (cd-rom drive has only 2 conductor out) is says "use R1" I have no idea what this means.
Please help
Darren
fierce_freak said:
No problem.
I have never used anything like that, but based on your stated requirement 100VA should be plenty.
Tying the grounds is mandatory (unless you use some sort of isolator chip, which is not really needed here), and no, it is not a problem so long as they all just meet at one spot. Tying the grounds does not negate the value of individually regulated circuits.
Cheers!
Russ
Re: COD vs Opus
There is no accounting for personal tastes, but from my point of view Opus is very definitely better than the passive and buffered COD.
It is a well established fact that current sources really must see a low impedance to work effectively. The TXD-IV solution I am using has an input impedance in the tens of milliohms. A resistor (or any other impedance) will cause the output of the DAC itself to swing (voltage wise) too much. This is not good. You want the DAC output to stay as static as possible voltage wise.
That said, COD with active I/V is at least as good as Opus, and I really would be hard pressed to choose between them.
Cheers!
Russ
hbarki said:How does the sound of Opus compared with COD (passive I/V stage)?
There is no accounting for personal tastes, but from my point of view Opus is very definitely better than the passive and buffered COD.
It is a well established fact that current sources really must see a low impedance to work effectively. The TXD-IV solution I am using has an input impedance in the tens of milliohms. A resistor (or any other impedance) will cause the output of the DAC itself to swing (voltage wise) too much. This is not good. You want the DAC output to stay as static as possible voltage wise.
That said, COD with active I/V is at least as good as Opus, and I really would be hard pressed to choose between them.
Cheers!
Russ
SRC4192 vs AD1896
Hi Russ,
I was wondering why you went with the SRC4192 instead of the AD1896? A while ago on DIYAUdio there was a discussion of the relative merits of the 1896 and the 4192, and iirc, the conclusion was that the PLL on the 4192 was set too high, and thus the AD1896 was the better part at jitter rejection. I'm asking because I'm contemplating ordering the metronome board, but I'm not sure if I want you to solder the SRC4192 in place or send it to me without that chip installed, and I'll install one of the AD1896s I have on it instead, since they are pin compatible.
Thanks!
Brad
Hi Russ,
I was wondering why you went with the SRC4192 instead of the AD1896? A while ago on DIYAUdio there was a discussion of the relative merits of the 1896 and the 4192, and iirc, the conclusion was that the PLL on the 4192 was set too high, and thus the AD1896 was the better part at jitter rejection. I'm asking because I'm contemplating ordering the metronome board, but I'm not sure if I want you to solder the SRC4192 in place or send it to me without that chip installed, and I'll install one of the AD1896s I have on it instead, since they are pin compatible.
Thanks!
Brad
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.