Valve DAC from Linear Audio volume 13

Actually, assuming everything works as it should, an audio recording device may be a better tool than a scope. You could play back a DSD512 recording of silence and of very soft music and record the output with RV3 and RV4 both set to 0, both to 1/10 of their maximum value, both set to 2/10 and so on and compare the recordings afterwards and listen which you like best.
 
The 0ohm resistors are basically a selector for whether you use master or bit clock and only one should be installed - take a look at the schematics.
Thanks, Ray - I don't think that's correct.

By the manual, master or bit clock are selected by the jumper settings.

By the schematic, I only see the 0 ohm resistors function in connecting the Amanero to the clock circuit or not.
 
Actually, assuming everything works as it should, an audio recording device may be a better tool than a scope. You could play back a DSD512 recording of silence and of very soft music and record the output with RV3 and RV4 both set to 0, both to 1/10 of their maximum value, both set to 2/10 and so on and compare the recordings afterwards and listen which you like best.

Thanks, Marcel - that's a great suggestion!

As you said, tapping the resistors R33 and so on could be dangerous with my equipment.

But what about the input legs of caps C1 and C3 for final adjustment of trimming clock and clockn to be in sync?

So, that would be adjustment by recording/ear with 1/10 steps of the trimmers set in tandem, and then finally syncing the positive and negative clock by scope.

Does this sound like a good plan?


Can you explain why there would be crosstalk with the resistors mounted when nothing is connected at P13? The only effect I can see is those unused antenna legs being a bit longer.
 
Thanks, Ray - I don't think that's correct.

By the manual, master or bit clock are selected by the jumper settings.

By the schematic, I only see the 0 ohm resistors function in connecting the Amanero to the clock circuit or not.
Probably a minor misunderstanding but if you look at page 3 of the schematic R58 & R59 (the 0ohm resistors that were both soldered by my friend) select which clock you use from the Amanero header (P13) though, as you're not using P13, Marcel is suggest you remove both resistors. IIRC the jumpers are to set the clock divider if you use the MCLK - I'm a bit rusty on this and don't have time to refresh myself on it now as grub is up.
 
As you said, tapping the resistors R33 and so on could be dangerous with my equipment.

But what about the input legs of caps C1 and C3 for final adjustment of trimming clock and clockn to be in sync?

That would be more awkward than R33 and R34, as C1 and C3 are SMD devices mounted on the backside of the PCB that have fairly high voltages across them.

So, that would be adjustment by recording/ear with 1/10 steps of the trimmers set in tandem, and then finally syncing the positive and negative clock by scope.

Does this sound like a good plan?

I would skip the scope part completely. The most critical clocks go straight to the bottom E88CCs (bottom in the schematic) without anything adjustable in the path anyway. The adjustment is just meant to ensure that the data signals to the upper E88CCs switch at a moment when the tail current is off, so the DAC behaves as a true RTZ DAC. That should be audible (and certainly was audible on my valve DAC prototype); if it isn't, it doesn't matter whether the DAC truly works in RTZ mode.

Can you explain why there would be crosstalk with the resistors mounted when nothing is connected at P13? The only effect I can see is those unused antenna legs being a bit longer.

Precisely that. If R48, R57 ... R59 and P13 are mounted, you have one of P13's vertical pins connected to the bit clock adjacent to a pin connected to a data signal and another one sandwiched between pins connected to data signals. That is bound to give some capacitive crosstalk - not a lot, but if you can avoid it, it's best to avoid it.
 
I am rebuilding the whole chain from scratch and working at the pre and power stages at the moment. At some point in 2023, I'll need a new DAC. Instead of getting something conventional, this DAC opens another level of experimentation. As I'll have a low distortion system, adding valve dac can give a more realistic sound. Primarily for jazz. I'll keep this thread on my radar and if decide to build will make inquiries about key board availability/abandoned projects or print new ones.
 
  • Like
Reactions: 1 user
If you should decide to build the original design, you have to know that presumably due to the chip shortage, the lead time for the FPGA module is long (the raw DSD version has no FPGA module). Before COVID-19, when you ordered something from Trenz Electronic, it usually got delivered in a few days (here in the Netherlands anyway). Nowadays they have almost nothing in stock, don't specify lead times on their site anymore and the last time I ordered an FPGA module, it took about half a year before it was delivered.
 
  • Like
Reactions: 1 user
I noticed a little noise on my Valve DAC so tried to null it with the potentiometers but I'm out of travel on one channel so it looks as though I'm going to have to take the main board out and substitute the revised components - I'm still using the original BOM values. I'll also install multiturn pots to give a bit more precision to the adjustments.
 
Has anyone built one with external power supplies? I want to build a second one that puts the transformer and a Maida regulator into a separate enclosure, then there is a 7-pin umbilical: -300 and ground from Maida; 6.3VAC for heaters; 6.3VAC for the digital side; and a safety/chassis wire. Would there be a downside into running a one-meter umbilical with the 300VDC from the Maida with two AC lines?