So I've been that thinking about using this DAC for a Linkwitz lx521 project. I would use the minidsp nanodigi for DSP processing and then use 4 of these DACs to decode the outputs to get 8 analog channels. To control volume I would use the digital volume controls integrated in the DAC.
My question is what would be the best way to control the volume on all 4 DACs? Would it be to use a single pot and connect the volume pins in parallel? Or would I have to use a 4way potentiometer to make sure they all read the correct resistance?
Would it be possible to directly control the pin with an active voltage stage instead, like from an arduino or similar circuit? This would allow me to use remote volume control and would be very convenient.
My question is what would be the best way to control the volume on all 4 DACs? Would it be to use a single pot and connect the volume pins in parallel? Or would I have to use a 4way potentiometer to make sure they all read the correct resistance?
Would it be possible to directly control the pin with an active voltage stage instead, like from an arduino or similar circuit? This would allow me to use remote volume control and would be very convenient.
IMHO the best (most accurate) way would be to use an Arduino and send the same serial command to all 4 boards. Both the MEGA and the DUE boards have plenty of serial ports to accomodate them. But I would suggest waiting for the new firmware that will enable the isolated serial port on the DAM.
Put everything together this weekend, with I2S source, it works. However, I have a question about 3.3v isolator PS, because I don't have it, is it OK? Also I learned that it can be 5V instead, can someone confirm it? Does it need a high quality PS?
BTW the DAC sounds really good, even though mine is 0.05 version, stock filter and just a temporary setup. Maybe part of the reason is I have ACF tube buffer right after it. Will update the filter once the adapter is delivered.
I found this board is available from Chinese TaoBao site, (not a replication) , not sure an authorized reseller or not, and, those guys are talking about to develop a backplane board to accommodate this DAC, USB interface and PS, should be an interesting one.
BTW the DAC sounds really good, even though mine is 0.05 version, stock filter and just a temporary setup. Maybe part of the reason is I have ACF tube buffer right after it. Will update the filter once the adapter is delivered.
I found this board is available from Chinese TaoBao site, (not a replication) , not sure an authorized reseller or not, and, those guys are talking about to develop a backplane board to accommodate this DAC, USB interface and PS, should be an interesting one.
A last series of measurements ... I have found that the Fireface 800 has a second pair of inputs with variable (analog) gain which has less gain but has a nice looking noise floor at around -130dB (input shorted and averaged samples):
Measurements were made with 0dBFS=-28dBu ~ 0.03V RMS. So 32 dB below full scale SE-out, or 41 dB below FS balanced.
First the noise spectrum of the switched off DAM connected via the balanced output. Somehow some mild noise enters the system.
Next the noise, balanced, DAM switched on no signal (over shorted input):
Finally a 1kHz -64.5dB signal (over DAM-off, and second over DAM-on no signal). The plot of the difference of the 1khz signal and DAM-on, no signal, (blue) looks already better.
Now we look at the balanced mono output taken from the SE connectors.
First DAM powered off over noise floor (above) and DAM powered on, no signal (below). The reason that the "powered on"-plot looks better is, that with power on the SE input has a residence of 625Ohm to ground, whereas switched off the resistance is high. The plots indicate also that my signal path, two chinch cables in a cinch to stereo TRS adapter, has room for improvement
When I have some time again I will work on that an see how much remains.
Now the 1kHz -60dB signal over the DAM-powered-on noise. The difference (blue) looks really nice.
Finally a word to the measurement. The plots are from an average of 100 samples, this removes (and hides) most of the random part of the noise. The non-randon part is quite reproducible. Here a plot of the difference of two independent measurements the DAM-powered-off measurement in the balanced-mono case. I admit the levels here are probably beyond any practical relevance, but it is interesting to see.
Measurements were made with 0dBFS=-28dBu ~ 0.03V RMS. So 32 dB below full scale SE-out, or 41 dB below FS balanced.
First the noise spectrum of the switched off DAM connected via the balanced output. Somehow some mild noise enters the system.
Next the noise, balanced, DAM switched on no signal (over shorted input):
Finally a 1kHz -64.5dB signal (over DAM-off, and second over DAM-on no signal). The plot of the difference of the 1khz signal and DAM-on, no signal, (blue) looks already better.
Now we look at the balanced mono output taken from the SE connectors.
First DAM powered off over noise floor (above) and DAM powered on, no signal (below). The reason that the "powered on"-plot looks better is, that with power on the SE input has a residence of 625Ohm to ground, whereas switched off the resistance is high. The plots indicate also that my signal path, two chinch cables in a cinch to stereo TRS adapter, has room for improvement
When I have some time again I will work on that an see how much remains.
Now the 1kHz -60dB signal over the DAM-powered-on noise. The difference (blue) looks really nice.
Finally a word to the measurement. The plots are from an average of 100 samples, this removes (and hides) most of the random part of the noise. The non-randon part is quite reproducible. Here a plot of the difference of two independent measurements the DAM-powered-off measurement in the balanced-mono case. I admit the levels here are probably beyond any practical relevance, but it is interesting to see.
If you take the 1k -60dB peak as reference, in the balanced mono plot it is at about local -39dB, the third harmonic about local -108dB so real-129dB. The noise floor is local -122dB, so real -143dB. This corresponds well with the direct measurements. The additional gain has then "only" the benefit that we do not operate at the theoretical lower limit oft the ADC and below the specified noise - thus we measure more precise, which is confirmed by the clear result of the difference sig/noise.
If you take the specs of the fireface as reference you would expect the peak being at -24 dB local minus some loss in the signal path. I will at some time try to pin down where this difference comes from.
Doh, just an update guys, turns out I had the potmeter disabled. I read the suggestion to set the volume to V-99, but I was doing so outside of the uManager prompt.
I found Soren's prior post for setting 'set volume = -99' in the umanager and straight after power cycling the dam the pot started working. Great!
Turns out sometime ago I was setting the volume via the set volume command, as I wasn't yet aware of being able to do it by simply inputting V commands blindly outside of uManager, turns out this disables the pot.
Yet another reason why we need official documentation.
I found Soren's prior post for setting 'set volume = -99' in the umanager and straight after power cycling the dam the pot started working. Great!
Turns out sometime ago I was setting the volume via the set volume command, as I wasn't yet aware of being able to do it by simply inputting V commands blindly outside of uManager, turns out this disables the pot.
Yet another reason why we need official documentation.
Last edited:
RAW output impedance matching?
Hi all. Planning to connect this amp to a DIY headphones amp. Both the DAC and amp are being built so I get to choose the right pot for the amp input.
What's the optimal input impedance for the headphones amp to match the DAC's raw SE output? Is 47.6K ohms suitable?
p.s. A late thank-you to @Priidik for your help
Hi all. Planning to connect this amp to a DIY headphones amp. Both the DAC and amp are being built so I get to choose the right pot for the amp input.
What's the optimal input impedance for the headphones amp to match the DAC's raw SE output? Is 47.6K ohms suitable?
p.s. A late thank-you to @Priidik for your help
Doh, just an update guys, turns out I had the potmeter disabled. I read the suggestion to set the volume to V-99, but I was doing so outside of the uManager prompt.
I found Soren's prior post for setting 'set volume = -99' in the umanager and straight after power cycling the dam the pot started working. Great!
Turns out sometime ago I was setting the volume via the set volume command, as I wasn't yet aware of being able to do it by simply inputting V commands blindly outside of uManager, turns out this disables the pot.
Yet another reason why we need official documentation.
Great! I know that feeling of relief.....
I ordered a board some time ago but have not gotten around to building it yet as I'm in the process of finishing a couple other projects before starting this one. Great story I know... anyways progress made here with regards to filters, add-on boards and implementations is really impressive and has me very excited. Very big thanks to those who have supported the development of this amazing and unique DAC offering. And extra thanks to Soren for making it available to us.
In my opinion, Soren NEVER release FW to support DSD.
I still waiting product manual since january. Soren knows that in EU is abslutely ilegal to sell a comercial product without manual, but Soren disobeys its obligations.
Regrads
https://hifiduino.wordpress.com/2015/03/16/soekris-dam-1021-r-2r-dac-users-guide/
Great manual.... what's missing?
While I agree it's taking Soren a fair while to update us on the firmware status, or even an official manual, there should be enough information to implement the boards.
If you're having difficulty getting things to work just ask on the implementation thread and folks will be able to help you with it.
I'd much rather have the built dac on my desk right now making sweet music than to wait another year or two before bureaucrats can be satisfied with some over-polished documentation. This is why diya rules.
Also grats to the hifiduino guy, i used his guide and it went smooth.
I was told this dac would be able to be used in an 8 channel active system so I bought 4 of them. There has been no documentation on exactly how to do this...so they are still sitting here in the box several months after I bought them. If I had been told upfront it would be months and months before any documentation would be available I would have waited. Still no word on when....While I appreciate the work that went into them I am starting to wonder if I have made an expensive mistake...
It seems we are not talking about the same thing. My opinion is referred to the manufacturer´s obligations and the support ofered to his products. In this sense Soren seems to be regrettable, just a disaster.
In the other hand, we can take care about the consumer´s enthusiasm, those who really give the product´s support.
Regards
I am waiting to get more boards since the firmware to run the DAM crossover function is not released yet. Søren did say that he was excited to get this feature running for his own personal use, so hopefully it will be out soon and well tested by him. I would be happy for more updates so that enthusiasm does not wane for the product. I am very close to settling for minidsp as an alternative.