Thank you for your reply.
Luckily I have the chance to put in 47,220,470 or even 2200u BGs but I don't want to destroy my pcb by soldering/desoldering different caps again and again. I read the whole topic (others as well) and I saw the posts about the value and realized that the value itself has a positive effect on the distortion, so practically high capacitance BGs would be the best choice. This raised my question why did you choose much smaller ones? These were available or something else affected your choice? I'm really curious about your reasons if there were any apart from the availability of the caps.
Of course the decision is mine but any help is much appreciated.
There is such a thing as too much capacitance. Once exceeded the sound turns slow and compressed.
I usually install sockets and then I can swap the caps without need to solder them each time. Distortion is one thing, but I don't think it translates directly into sound signature. I'm of the opinion that the actual type and capacitance value is much more critical when it comes to sound signature. Black Gates are not perfect, but for some reason they present the signature I kinda like.
I started building amps with a lot of capacitance like everybody else. I thought the more, the better. Then Junji Kimura came up with Gaincard and surprisingly, he had only 2000uF of filtering caps in 50W amp. That was completely different approach. I found it interesting though and did my own experiments which confirmed that the amp with much less capacitance can actually sound better: more transparency, detail and speed. With high efficiency drivers it did not affect bass performance at all. I actually preferred the bass this way.
So if it worked for amp, I thought it may as well work for other stages and I started experimenting with much smaller caps in my version of TDA1543 DAC. And it worked. Since then I always use as little capacitance as possible.
As I do not have a stock of BG, I used Mundorf smooth foil (not raw) which you use for cross overs normally with 15uf. Sounds much better than was in before. I think the sound signature is more important than the actual value. Maybe Kaisei is nowadays a good choice as well... or PP.
I'm getting ready of setting up a dam1941 with a couple of Salas ultrabib for the analog and a Reflektor mini for the digital. I'm thinking to use some Mills MRA05 resistors as suggested, but they are not cheap; Which value do you think is more appropriate?
I was thinking two 3.9ohm for the Ubib (159mA) and 1ohm or 1.2ohm for the reflector.
I was thinking two 3.9ohm for the Ubib (159mA) and 1ohm or 1.2ohm for the reflector.
Last edited:
I use 1941 with UBiBs.
Rset resistors (Ohmite AG 5W):
1R8 for Vcc5D and Vee5a (100mA load, 225mA shunt)
1R2 for Vcc5a (350mA load, 140mA shunt).
Rset resistors (Ohmite AG 5W):
1R8 for Vcc5D and Vee5a (100mA load, 225mA shunt)
1R2 for Vcc5a (350mA load, 140mA shunt).
Hi everyone, I just had an idea for a project:
I can buy for a good price an Acuhorn Streamer, nice machine, but the dac is not enough for what I want. And It don't have digital usb input.
So I'm thinking if could be possible to replace the dam1021 with dam1941?
I understand that I the sizes are not the same, so I should change the enclosuer, but I could re-use the raspberry controller and power unit.
What do you think? Could work? Would worth it?
Thanks,
I can buy for a good price an Acuhorn Streamer, nice machine, but the dac is not enough for what I want. And It don't have digital usb input.
So I'm thinking if could be possible to replace the dam1021 with dam1941?
I understand that I the sizes are not the same, so I should change the enclosuer, but I could re-use the raspberry controller and power unit.
What do you think? Could work? Would worth it?
Thanks,
Crossfeed
Crossfeed
Quoting the manual:
Crossfeed is a function added in dac1421/dac1541, and the dam1941 is based on the dac1541. So it support crossfeed. For those who don't know crossfeed, it's a function to be used with headphones to reduce the with of the soundstage to simulate speakers. The dac1541/dam1941 do it using digital filters that mixed the two channels.
The“XFEED” button selects the Crossfeed mode for Headphones, it’s only active when using the Headphones. The Crossfeed function is used to make the stereo image smaller, to make it sound more like the stereo image from speakers. dac1541 “XFEED” LED color shows the current crossfeed mode selected for the headphones
dam1941 Pin Description J4,
Now,
I could not detect any difference at first try.
Actually, in order to simulate a natural crossfeed, more has to be done.
There are three mechanisms that a person uses to locate and externalize sources of sound:
First, the sound of a source to the right side of the listener (e.g., the right loudspeaker in a stereo setup) not only reaches the right ear, but attenuated and delayed, is also heard by the left ear. The level of attenuation and the delay time of this crossfeed signal provide important directional information. For localization of a sound source, the delays of the frequencies below 2 kHz are the most important, and therefore should have a natural 300 microseconds delay.
Second, the soundwaves are partly absorbed and partly reflected by the listener's head. Especially, the reflections at the ear pinnae interfere with the soundwaves that directly enter the ear canal and amplify or attenuate specific frequency components. As these reflections depend on the direction of the soundwave, the "color" of the sound changes with the direction of the source.
Third, reflections of soundwaves on the walls of the listening room produce reverberation that conveys an extra feeling of space.
The information obtained by these mechanisms is further refined by movements of the head. Changes in sound levels, delay times and sound colour refine our sense of direction.
All these mechanisms are missing when we hear music using headphones. The transducers are directly coupled to our ears. The sound of the right (left) transducer will not reach the left (right) ear and the reflections on the oracles have changed and hardly interfere with the original soundwave. Moreover, the sound-sources are attached to our head, so head movements no longer add information. Reverberation is not present.
The lower limit for directionality is twice the distance between the ears, or about a foot, or Fl = 1100 Hz. This means that the low frequency listening environment for speakers can be approximated with headphones by crossfeeding about a factor of unity of the opposite channel's signal into the other channel-about doubling the sound intensity below Fl.
Although this is an approximation, it is reasonably close to the value used in other designs-Jan Meier used 650 Hz. The original Linkwitz paper used 700 Hz., also.
The high frequency listening environment for speakers can be approximated with headphones by crossfeeding about a factor 1/4 of the opposite channel's signal into the other channel, which is reduced by a factor of 1/4, such that it is still unity sound intensity. Again, this is an approximation, but compares favorably with the original Linkwitz paper which used -3db for the high frequency crossfeed signal ratio.
Reiterating the above reasoning, the stereo system speaker environment can be approximated with headphones by:
The frequency domain is div
Technically it should be easy to simulate the normal amounts of transaural crosstalk which could do fascinating things to the surrealistic image you get with phones. Soekris has all the technology needed to do it. Just attenuate the signal, roll off the high end ,delay it .1 msec and feed it to the contra lateral ear.
Split the stereo inputs into 4 signal lines. Delay, filter and attenuate one line from each channel and mix it into the opposite channel. It wouldn't take any longer to build the circuit paths than it took to type this. You could fine tune all the parameters while you listened.
Head Related Transfer Functions is what you should use instead of crossfeed filters. You ought to simulate a perfect stereo-dipole, such as they are used in ambiophonic presentation. This will make the sound appear to come from in front of you, with a very wide stage where sound sources are presented with pin-point precision.
HRTF is a psychoacoustic term which has to do with the way we determine direction of sounds. It consists of the major components Interaural Level Difference (and the related comb-filtering) and Interaural Time Difference: ILD and ITD, and also of pinna-related phenomena like phase-shifts and the like. There are many resources on the internet if you want to learn more.
OT. I suggest Sören to look into this as a market opportunity.
Crossfeed
Quoting the manual:
Crossfeed is a function added in dac1421/dac1541, and the dam1941 is based on the dac1541. So it support crossfeed. For those who don't know crossfeed, it's a function to be used with headphones to reduce the with of the soundstage to simulate speakers. The dac1541/dam1941 do it using digital filters that mixed the two channels.
The“XFEED” button selects the Crossfeed mode for Headphones, it’s only active when using the Headphones. The Crossfeed function is used to make the stereo image smaller, to make it sound more like the stereo image from speakers. dac1541 “XFEED” LED color shows the current crossfeed mode selected for the headphones
- Off Crossfeed circuit disabled Green Small Crossfeed,
- -12 dB sent to other channel Orange Medium Crossfeed,
- -8 dB sent to other channel Red Large Crossfeed,
- -5 dB sent to other channel.
dam1941 Pin Description J4,
- pin 16: input XFeed Select Switch to uC.
- pin 04 input: Control Signal to Enable Headphone Output
Now,
I could not detect any difference at first try.
Actually, in order to simulate a natural crossfeed, more has to be done.
There are three mechanisms that a person uses to locate and externalize sources of sound:
First, the sound of a source to the right side of the listener (e.g., the right loudspeaker in a stereo setup) not only reaches the right ear, but attenuated and delayed, is also heard by the left ear. The level of attenuation and the delay time of this crossfeed signal provide important directional information. For localization of a sound source, the delays of the frequencies below 2 kHz are the most important, and therefore should have a natural 300 microseconds delay.
Second, the soundwaves are partly absorbed and partly reflected by the listener's head. Especially, the reflections at the ear pinnae interfere with the soundwaves that directly enter the ear canal and amplify or attenuate specific frequency components. As these reflections depend on the direction of the soundwave, the "color" of the sound changes with the direction of the source.
Third, reflections of soundwaves on the walls of the listening room produce reverberation that conveys an extra feeling of space.
The information obtained by these mechanisms is further refined by movements of the head. Changes in sound levels, delay times and sound colour refine our sense of direction.
All these mechanisms are missing when we hear music using headphones. The transducers are directly coupled to our ears. The sound of the right (left) transducer will not reach the left (right) ear and the reflections on the oracles have changed and hardly interfere with the original soundwave. Moreover, the sound-sources are attached to our head, so head movements no longer add information. Reverberation is not present.
The lower limit for directionality is twice the distance between the ears, or about a foot, or Fl = 1100 Hz. This means that the low frequency listening environment for speakers can be approximated with headphones by crossfeeding about a factor of unity of the opposite channel's signal into the other channel-about doubling the sound intensity below Fl.
Although this is an approximation, it is reasonably close to the value used in other designs-Jan Meier used 650 Hz. The original Linkwitz paper used 700 Hz., also.
The high frequency listening environment for speakers can be approximated with headphones by crossfeeding about a factor 1/4 of the opposite channel's signal into the other channel, which is reduced by a factor of 1/4, such that it is still unity sound intensity. Again, this is an approximation, but compares favorably with the original Linkwitz paper which used -3db for the high frequency crossfeed signal ratio.
Reiterating the above reasoning, the stereo system speaker environment can be approximated with headphones by:
The frequency domain is div
Technically it should be easy to simulate the normal amounts of transaural crosstalk which could do fascinating things to the surrealistic image you get with phones. Soekris has all the technology needed to do it. Just attenuate the signal, roll off the high end ,delay it .1 msec and feed it to the contra lateral ear.
Split the stereo inputs into 4 signal lines. Delay, filter and attenuate one line from each channel and mix it into the opposite channel. It wouldn't take any longer to build the circuit paths than it took to type this. You could fine tune all the parameters while you listened.
Head Related Transfer Functions is what you should use instead of crossfeed filters. You ought to simulate a perfect stereo-dipole, such as they are used in ambiophonic presentation. This will make the sound appear to come from in front of you, with a very wide stage where sound sources are presented with pin-point precision.
HRTF is a psychoacoustic term which has to do with the way we determine direction of sounds. It consists of the major components Interaural Level Difference (and the related comb-filtering) and Interaural Time Difference: ILD and ITD, and also of pinna-related phenomena like phase-shifts and the like. There are many resources on the internet if you want to learn more.
OT. I suggest Sören to look into this as a market opportunity.
You might be doing something wrong, the crossfeed filters works just fine. The easiest way to test is with an early Beatles recording, they have a extreme stereo image....
I have made some adjustments to my DAM1921. Totally sharged by lt3045 boards. Tried to change the caps described earlier in this thread. Big caps changed to FG 2200uF, and after that FG 100/25 and changed the ceramics for FG 22/25. This was significant better with clarity in sound and micro details, air between instruments. But still a little edgy and sharp. So I will try some more cap-settings.
I have ordered a Salas triplet and think it will be better than lt3045.
I use a AD797AN buffer stage on output. Nice sounding and works good. I have tried different buffers and may go back to raw but like to be able to connect to "anything" and it sound good. Next I will try a tube buffer but is a little suspicious with needed coupling caps..
I have ordered a Salas triplet and think it will be better than lt3045.
I use a AD797AN buffer stage on output. Nice sounding and works good. I have tried different buffers and may go back to raw but like to be able to connect to "anything" and it sound good. Next I will try a tube buffer but is a little suspicious with needed coupling caps..
Attachments
Does anyone have a suggested parts list for populating the three PCBs (dsp1971, buf1961 and pwr1951) that are manufactured for DAM1921/1941?
Thx!
Sameer
Thx!
Sameer
Accessory PCB parts list
Found it!
http://www.soekris.dk/download/dam1941/dam1941_fileset.pdf
Thanks,
Sameer
Found it!
http://www.soekris.dk/download/dam1941/dam1941_fileset.pdf
Thanks,
Sameer
1941 parallel mode
Hi everyboby,
I see in the DAM1941 user manual :
"The dam1941 can be set for “Parallel” mode, in that case it will be single ended outputs with an
output impedance of 313 ohms"
How I do to connect my DAM1941 in parallel mode ?
Thanks for ansewering and best regards
Hi everyboby,
I see in the DAM1941 user manual :
"The dam1941 can be set for “Parallel” mode, in that case it will be single ended outputs with an
output impedance of 313 ohms"
How I do to connect my DAM1941 in parallel mode ?
Thanks for ansewering and best regards
As per manual, in uManager do a "set mode=parallel". Then parallel the two resistors string per channel, ie, short the + and - outputs....
That function was added after the first release, so it's not documented very good. It was discussed earlier on in this thread.
Will shorten the pin 5-7 and pin 13-15 with a jumper and then use the pin 8-10-12-14 like in the picture works too?
Attachments
Last edited:
I think many would like you to that instead of saying that something is not well documented (happens several times before) see a comment like "I just updated the document in the soekris homepage so that you, dear customer, can use my product in the most satisfying way and I'm sorry for the inconvenience and I aim to do better in the future" 🙂
//