Because the dam1021 is fine as is, but as it's a DIY product, some people want to modify it, just to do something, you might have noticed that trend on some of your own projects....
I wouldn't rail him for the Vref flaws... It certainly wasn't nice, but it is DIY and we wouldn't blame Soren for a bug in the firmware that needs reflashing, would we?... I know he charged you for repairs but everyone else here who got in on rev1 seemed to have no problem making those changes themselves. He's not a native speaker so I don't think we must take it at face value when he uses the word "idiot"...
You've clearly been waiting for him to implement multi-channel features all this time. I have no idea if it's technically impossible as Soren claims it is now, but it's obviously a real bummer. To be fair, Soren has done a pretty decent job on other aspects of engineering. I don't know if this DAC sounds better than the Apple 3.5mm dongle (or if any DAC does), but at least it's a reasonable attempt at R2R when there is nothing else affordable out there. If he's ethically corrupt, there would be no word left for MSB and the like, except maybe "idiot".
To be very blunt, I don't think leveling accusations at Soren will be of any use in your case. I think he's a very competent engineer, and maybe even extraordinary in the company of other "audiophile engineers"; many have been much more lavish in their praise of the soekris DACs. If he were to do anything right now, he would be admitting to his "ethical corruptions". Even a full refund can't offset that image.
I certainly don't think it's a happy situation that Soren told everyone multichannel crossover or something like that would work with dam1021 but later claims it to be impossible or in any case never going to be implemented. You probably got the worst of it out of everyone on this thread. I think anyone who isn't infatuated with the Soekris DAC can sympathize with you to some degree. We might take it for granted when all the large retailers have generous return policies nowadays, but it would still be kind of him if Soren ever decides to take your boards for his experiments in exchange for some amount of store credit, especially after all this time and many iterations in the product line. It's a damn shame that your boards have been sitting around for all those years
DIY means different things to different people. When you call a product "reference", I think most would assume that one wouldn't have to break out the soldering iron to repair fundamental issues with the board itself. Hypex calls their ncore amp modules DIY- do you know anyone who had to replace or add components to their boards to bring them up to snuff? Soren has admitted publicly on this forum after several years that the V1 had issues. I don't believe the buyer should have to fix these problems in order to have what was advertised. DIY doesn't mean "fix it yourself".
When you tell people, without it turns out, even testing yourself, that the product can be used in a multichannel configuration, then the truth comes out that it can't, what is that called? The only reason I bought these 4 boards was because I was told by the seller that they would work for this application. We know now they can't be. The right thing to do would be to take them back, that is without question what any other legitimate business would do.
I don't expect anything at this point from Soekris Engineering. Their true colors have been shown to all. I get that people using these as a stereo dac are happy with them. I am glad people have been able to get this dac to perform to their satisfaction. I have $1250US work of paperweights because they won't work as advertised in my application, despite being told they would. Silly me for believing the vendor.
I appreciate your understanding of my side of the situation.
Yes, most people use the dam1021 as is, you don't need to modify it....
An integrated buffer makes it a ready solution for people driving headphones. In the absence of such needs it only subtracts transparency.
PS improvements work equally well on other dacs as well; using a single transformer is always a compromise.
These measures and many others have been implemented in my other dacs and were essential to make the DAM comparable. It is not realistic to expect miracles from the default setup.
Just to make it clear:
The dam1021 WILL work just fine in a multichannel configurations, just feed multiple boards I2S audio and take the outputs, just like any DAC chip.
It also do have the hardware for doing the digital crossovers using the onboard IIR filters sections. But I never promised a smart user interface for it, you need to design the filters yourself and add them to the filter file. Its not an easy task, so the best way is probably to use an external DSP board with a good user interface.
Anybody claiming something else do not understand the board, or worse, is lying.
The dam1021 WILL work just fine in a multichannel configurations, just feed multiple boards I2S audio and take the outputs, just like any DAC chip.
It also do have the hardware for doing the digital crossovers using the onboard IIR filters sections. But I never promised a smart user interface for it, you need to design the filters yourself and add them to the filter file. Its not an easy task, so the best way is probably to use an external DSP board with a good user interface.
Anybody claiming something else do not understand the board, or worse, is lying.
Usually I supply the DAC with batteries, so how many rails do I need?
Can I use the direct output without any opamp and buffer?
It should be the best way if it will drive 10K pot.
I wouldn't use the digital volume control, I want a bit perfect signal at DAC level.
There is no better volume control than the already built in.
Now I use it direct into a poweramp. Better than with a highend preamp between. It only adds distortion anyway Without buffer the output impedance is 640ohm.
I think it is all in the manual? At least been mentioned many times in this thread.
Now I use it direct into a poweramp. Better than with a highend preamp between. It only adds distortion anyway
Without buffer the output impedance is 640ohm. I think it is all in the manual? At least been mentioned many times in this thread.
The buffer is there not just for headphones but for everyone requiring (professional) balanced outputs.
Practically all the music availible has run through some (de)balancing stages, since balanced lines are the standard in pro audio. Nothing bad in principle about it.
The DAM1021 buffer doesn't sound very good to my ears, but I see no need to remove it, as its presence doesn't affect the unbalanced output.
I don't think it's a good idea to remove the NPO output filtering capacitor. It serves a necessary function, is out of the audio band and these caps are well suited for filtering.
Practically all the music availible has run through some (de)balancing stages, since balanced lines are the standard in pro audio. Nothing bad in principle about it.
The DAM1021 buffer doesn't sound very good to my ears, but I see no need to remove it, as its presence doesn't affect the unbalanced output.
I don't think it's a good idea to remove the NPO output filtering capacitor. It serves a necessary function, is out of the audio band and these caps are well suited for filtering.
There is no better volume control than the already built in.
Now I use it direct into a poweramp. Better than with a highend preamp between. It only adds distortion anyway
I think it is all in the manual? At least been mentioned many times in this thread.
This is your opinion that I respect but I don't agree, I would never use a digital volume control. Usually I don't use any preamp but simply passive components that don't add any distortion.
If you set the digital volume control to -60dB you are losing 10 bit depth so you are listening to a 17 bit DAC (27-10).
I'm sure that there is all in this thread, but I have no time to read all the posts so I ask some suggestions from who is already running the DAC.
Last edited:
The DAM1021 is designed to be used with a digital volume control. It is the most transparent way to use it.
Do you have any idea what a potmeter does with the low level signal? Not to mention issues with stereo tracking and dust over time. I tried a lot, pots, stepper switches and relais based volume control. Everything degrades the audio signal a lot IMHO.
So I also vote for digital volume control, but not at +60 dB. I advise to set up amp gain so that 0 or -10 dB is the max required level. Then typically you would go down another 20 or 30 dB in normal listening. Max 40 dB in low level listening or so...
This way you still have enough bits...
So I also vote for digital volume control, but not at +60 dB. I advise to set up amp gain so that 0 or -10 dB is the max required level. Then typically you would go down another 20 or 30 dB in normal listening. Max 40 dB in low level listening or so...
This way you still have enough bits...
I lastly worked on one of these dacs more than a year ago and memory is not what it used to be. Just a warning my advice may be inaccurate.
Batteries would be perfect. You have the option of supplying either a single bipolar voltage between +/-7 and +/-15v, or individual rails.
The most important rails are the supplies for the R-2R shift registers and the clock supply.
The shift register supplies are +/-4v, initially pre-regulated to +/-5v by IC regulators (U5/U7) and final regulation is achieved by opamps with a tl431 reference.
Some builders have bypassed the entire on-board regulation and supplied 3v3 from batteries suffering a decreased maximum output voltage. I have only removed the 5v regulators and added pass transistors to the opamp regulators, so PS to the shift registers is still 4v.
The 3v3 regulator for the clock is just next to it and very easy to remove.
Some have also supplied the FPGA with an alternative 1v2 regulator.
I respect your opinion but I wouldn't use the digital volume control neither in my Trust 50 Euro desktop speaker system connected to the PC I use to work.
I'm happy with my Shallco rotary switch and shunt resistors, no room for digital crap.
Your suggestions are very appreciated, but I wouldn't rebuild much of the DAC, since is a finished device I expect it works fine without heavy tweaking.
It's not the DAC of my life, I'm designing my own DAC and my own FIFO with a totally different architecture.
I have ordered this DAC to compare it with other devices and mostly to measure the phase noise of the clock.
Maybe if I find the time to do a heavy tweaking, I will throw out all the front-end keeping the DAC switches and the ladder network only, then I will feed it with my fixed oscillators and my FIFO buffer to understand the sonic result with a different architecture.
If I was you, I’d give it an ear or two—since it’s all here already, so the time and money investment should be minimal...
or be consequent and ditch all digital crap and rely on a reel-tape [emoji39] but that’s for different threads [emoji846]
Yes, of course, it's sign-magnitude !Sign-magnitude. It was explicitly created as an alternative to sigma-delta DACs.
Short circuit in some brain cells