This means pre-programming each board mono left and mono right before daisy chaining?
Some random thoughts.
On the volume control;
When I connect the DAM directly to the power amp I usually end up with volume settings V-15 to V-20. In view of
Perhaps a future revision could realize the coarse part of the volume control outside the ladder.
On crossover/momo:
I would prefer to keep the paths of the analog signals short. So ideally something like aktive speakers with digital input. For that the DAM would need
1.) to have the crossover for one channel on one board (with the current approach, one board handle a dedicated frequency range for both channels).
2.) that two DAM boards can run synchronous and communicate controls over some meters distance.
On the volume control;
When I connect the DAM directly to the power amp I usually end up with volume settings V-15 to V-20. In view of
the realization of the volume control in the DAC ladder is convenient but a waste of resources. The expensive MSB resistors are never used, and the specs are thus (theoretically) worse than if you attenuate after the ladder (for the volume only, no precision resistors are needed) or lower Vreff.
Perhaps a future revision could realize the coarse part of the volume control outside the ladder.
On crossover/momo:
I would prefer to keep the paths of the analog signals short. So ideally something like aktive speakers with digital input. For that the DAM would need
1.) to have the crossover for one channel on one board (with the current approach, one board handle a dedicated frequency range for both channels).
2.) that two DAM boards can run synchronous and communicate controls over some meters distance.
That could work to some extent if only a physical input switch and volume pot are used for dam1021 control, assuming DAM ignores all irrelevant chatter on serial line, like R and L messages.
For driving it with a microcontroller it gets hairy. With the current firmware we have to go into umanager to obtain the filter list. With daisy chained setup the downstream devices would have to echo the umanager output from the first device without acting on it. It gets ugly pretty fast...
And to change default settings or load new filters and firmware updates the completed device would have to be physically taken apart...
I'm getting more inclined towards using dedicated serial host ports for each dam board.
Any official recommendations?
As stated long time ago, the dam1021 resistor network direct out is a voltage source with 625 ohm in series. If you want to lower the output level you simply just shunt the output with a suitable resistor....
Now you want to push beyond what the dam1021 was designed for, it can't be everything to everybody and still be simple and low cost....
A follow up here. This happens from time to time. Quite annoying
Did a check in umanager today and this is what it says when it becomes silent.
Seems as it loads a "silent" filter or something (L000). I am unable to make it sing again. Have to turn power off and on again.
Attachments
I have never observed anything like that. L000 could be caused by unstable clock. Are you sure it's the dam1021, not your player or your raspberry pi ??
I have uploaded updated firmware, still 0.99 but dated 20150826. Can you please check to see if the problem you observed is still there ?
An easy solution could be to give each board an individual id-number as parameter to be set in uManager once.
Then make two variants of the commands, e.g.
"set2" would only affect DAM no 2, whereas "set" would affect all DAMs.
For settings where it makes no sense to send a command to all DAMs the unnumbered command should be disabled if the DAM has an id number.
The only problem is , that I have the impression the DAM does not always react on commands (but perhaps I only typed something wrong ... hard to say if you get no echo of what you typed in play mode). If this happens with daisy chained DAMs they end up in different states.
Lets face it, there is no elegant way to control multiple devices over a single RS232 line. The second set of commands will not help if there are more than two dacs for crossover use. And it would not be that trivial to make it work reliably. It is going to take a bit of extra hardware to make multi-dam setup fully functional.