A NOS 192/24 DAC with the PCM1794 (and WaveIO USB input)

As I understand, the LF80 and LF33 are the 8.0v and 3.3v standard regulators. I would imagine the new shunt regulators are to replace these. The 3 legs of the shunt reg will be input voltage, output voltage and gnd and will be the same layout as for a standard 3 leg 7805 reg. If you look at the standard surface mounted LF regulators, you'll see that they have 2 legs presumably for voltage in and out, then the body tag at the other end is also connected to gnd. Removal of the wima caps on the board is only because there are also caps doing the same job on the shunt reg.
Take that all with a pinch of salt as I haven't done it myself, yet, but that would be my guess. Hopefully enough to point you in the right direction?
 
Sovereign,

You need to look closely at my pictures in post 259.
I used the soldering points of the pink capacitors because they are easier to solder than the SMD soldering connections of the standard regulators.
And also because these caps need to be omitted when using the shunts, also the electrolytics downstream of the regulators need to be removed. Otherwise it will deminish the effect of the shunt regulators.

You need to remove the regulators from the board, they are the black IC's with code LF80 and LF33 That come soldered on the board.

Regards,
 
LF type

see picture
 

Attachments

  • GE3DPAK05-40.jpg
    GE3DPAK05-40.jpg
    5.1 KB · Views: 1,057
I have taken a part of the drawings of Doede, you can clearly see how the LF regulators are connected.
Look special for c12 and c16 they are connected to LF80 and LF33 to out and ground.
The tent labs reg. goes in place of the LF80 and LF33
 

Attachments

  • LF connection.jpg
    LF connection.jpg
    64.6 KB · Views: 1,032
If you've removed the original RF reg, you will have 3 solder points. One large one which is gnd, then V-in and V-out.
Set your multimeter to continuity (where it beeps when there's a connection) and do some probing to work out what's what.
Hold 1 probe onto the large gnd from the original reg, then probe the 2 holes from the pink cap. This will tell you which cap hole is gnd. Then put probe 1 into the other cap hole and probe the 2other points from the original reg. The one that beeps is the V-out, the other is V-in.
Once you've worked out what's what, join the V-in to the V-in of your shunt reg, then join the gnd to either of the gnd points you've identified and join the V-out to either the original regs V-out point or the V+ of the pink cap.
Make sense?
 
I could not have explained it better. Thanks!

So the Vin of the shunt will be soldered to the pad of the original regulator. Than Vout and ground can be soldered on the spot of the wima caps.
It is also possible to use all the pads from the original regulator but that will make a weak soldering connection so I will not advise to do so.

Good luck!
 
Supersurfer, you mentioned before about removing electrolytic caps if the shunt regs are being used. I've looked at your pictures which seem to show 1 of the 47uf caps removed per channel from the space where there are 4 of them together. Is this one removed just for easier fitting of the shunt regs or is this the one you mentioned to remove to make the shunt reg perform better?
Thanks,
James
 
Excellent, thanks. Is that because those local electrolytic caps can cause an oscillation? Is there something better I could put in their place?

Maybe a question for Doede, but is there a reason why a solid polymer cap like an Oscon or Nichicon NS isn't used for the digital supplies rather than a normal electrolytic? In my limited understanding, I thought these were ideal for digital decoupling. Do the Wima film caps mean they're not necessary?
 
As you can see on the picture, Guido Tent also make use of a Wima cap, close to the 3 pins of his shunt regulator.
Perhaps this is a cap on the output of the shunt regulator,if so then you don't need an extra cap between shunt reg. and dac.
It would be double up.
 

Attachments

  • page29_1.jpg
    page29_1.jpg
    41.2 KB · Views: 756
Quick quuestion regarding the quality of the ribbon cables in general, and the ones flowing out of the WaveIO USB card in particular.

I do not care about the ribbon cable used for the LEDs, but there is another one used to transport the I2S signals to the DDDAC1543 mkII SPDIF & I2S receiver board.

There are only 4 signals carried by this ribbon cable ( DATA, FS, BCK and GND), so I would like to know if it is not better to use 4 high quality cables (let say for example Teflon insulated) instead of a ribbon cable, in order to protect as much as possible the signal coming out of the WaveIO board ?

Thanks for your advices,