Output stage for the Soekris R2R

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi folks,

my dam1021-12 will arrive tomorrow - or so... I want to use the unbuffered se outputs to feed an output stage that offers both, se and bal outputs. Adding gain would be a nice feature. As I cannot make my own boards I would rather like to buy a kit or at least a pcb plus additional information (BOM, schematics, etc.).

Do you guys have solutions that you're willing to share? Btw. I don't like tubes...


Best,
Dominik
 
Hi folks,

my dam1021-12 will arrive tomorrow - or so... I want to use the unbuffered se outputs to feed an output stage that offers both, se and bal outputs. Adding gain would be a nice feature. As I cannot make my own boards I would rather like to buy a kit or at least a pcb plus additional information (BOM, schematics, etc.).

Do you guys have solutions that you're willing to share? Btw. I don't like tubes...


Best,
Dominik

Hi Dominik,

Why you need more gain, since most of the times you have attenuator in front of amplifier anyway. Do you really need more gain? If you want to use headphones then watch for Salas DCG3 in this forum. He makes spectacular designs. DCG3 is discrete preamp and headphone amp with gain of 3 times.
 
Last edited:
Wow, thanks for the fast replys!

I plan using a "The Wire BAL-BAL" as an amp. So, balanced outs of the buffer stage are needed. I thought it would noisewise be better, to have some gain (2x or maybe 3x) right after the DAC. Am I wrong?

Of course I will be using the volume control auf the dam module only.
 
Wow, thanks for the fast replys!

I plan using a "The Wire BAL-BAL" as an amp. So, balanced outs of the buffer stage are needed. I thought it would noisewise be better, to have some gain (2x or maybe 3x) right after the DAC. Am I wrong?

Of course I will be using the volume control auf the dam module only.

I've such a configuration, but I have avoid using the buffered output of the DAM, because the OPA1632 used in the Wire BAL-BAL have the same capabilities of doing BAL out of a SE signal when tying Vin- to the ground (figure 13 of the datasheet : http://www.ti.com/lit/ds/symlink/opa1632.pdf ) .

I've also raised the input impedance to 10k ohms (it's 1k ohm default for the Wire).
 
Last edited:
Wow, thanks for the fast replys!

I plan using a "The Wire BAL-BAL" as an amp. So, balanced outs of the buffer stage are needed. I thought it would noisewise be better, to have some gain (2x or maybe 3x) right after the DAC. Am I wrong?

Of course I will be using the volume control auf the dam module only.

See my setup, except I'm using a pair of dam1021 boards for balanced raw output feeding the BAL-BAL rather than the buffered outputs of a single board. I solely use the dam based volume control. It sounds absolutely incredible with hd800's.

I've set the BAL-BAL to 2x gain using 10k resistors on the inputs and 20k resistors for the negative feedback because the stock config of 1k resistors put too much load on the dam1021's raw output (Zout of 625R), although it should be fine using buffered output which is significantly lower at 20R balanced.

EDIT: Indeed GourouLubrik is correct, its way more suitable to use the BAL-BAL's OPA1632 to convert the single ended to balanced in a single board configuration. You'll still need to change the BAL-BAL to 10k input impedance though for the same reasons we did.
 
Last edited:
You guys are absolutely right! As you mentioned the low input impedance of "the wire" is a problem when connected to a dam1021 (using the RAW output). Although SE to BAL conversion is possible within "the wire" I decided to go for a dedicated output stage, that offers many advantages, such as SE and BAL output and a low output impedance of about 10R. Besides "The Wire" can use it's full potential.
 
Hi,
one question: does anything speak against using an analog volume pot after the unbuffered output? I would use a 50k motorized pot, which should work well with the 600-something Ohms output impedance of the dam, followed by a Calvin hybrid buffer.
Normally I would not see a problem here, but most people use the digital attenuation... so I'm just checking...
Thanks
Florian
 
Hi,
one question: does anything speak against using an analog volume pot after the unbuffered output? I would use a 50k motorized pot, which should work well with the 600-something Ohms output impedance of the dam, followed by a Calvin hybrid buffer.
Normally I would not see a problem here, but most people use the digital attenuation... so I'm just checking...
Thanks
Florian

Why you what to use analog attenuation when you have bit perfect digital available? For me what you want to do does not make sense at all and I do not see any benefit of that. Moreover it will just add lot of distortion and channel imbalance in low volumes levels.
 
I like tinkering as much as the next guy but I can't help wondering, do one really need a buffer stage for the dam?
I have a couple of Hypex nCore amps with differential input. So, XLR connections but you can drive them just fine with a single ended source. Input impedance is close to 100kohm, I think, much higher than 10k anyway. The dam's output impedance of 600'ish ohms shouldn't pose any problem...
Although I have been wanting to build some buffer stages, I can't really justify it if there is nothing to be gained from it. My speakers are very efficient, no extra amplification of the signal is needed.

So, is there anything for me to gain with a buffer stage or should I just be happy and go with the RAW output from the dam? It'll sure make for a more compact, easier, cheaper and more efficient build. I'm waiting for the 1231 (Pi-dam) to become available.
 
Why you what to use analog attenuation when you have bit perfect digital available? For me what you want to do does not make sense at all and I do not see any benefit of that. Moreover it will just add lot of distortion and channel imbalance in low volumes levels.

I know, it would be more convenient, but digital volume controls are normally not bit perfect. Usually bit depth is lost the more you attenuate. Is there somewhere explained how this works in the dam1021? I am happy for any pointers. I only recall someone in the dam-thread measuring THD at different volume settings, and THD rises substantially with increasing attenuation.

My thinking is just: Yes, there is increased distortion and channel imbalance with analog potis, but this is neglible with quality parts. Probably as neglible as the increased distortion that was measured with the dam digital attenuation.
 
I like tinkering as much as the next guy but I can't help wondering, do one really need a buffer stage for the dam?
I have a couple of Hypex nCore amps with differential input. So, XLR connections but you can drive them just fine with a single ended source. Input impedance is close to 100kohm, I think, much higher than 10k anyway. The dam's output impedance of 600'ish ohms shouldn't pose any problem...
Although I have been wanting to build some buffer stages, I can't really justify it if there is nothing to be gained from it. My speakers are very efficient, no extra amplification of the signal is needed.

So, is there anything for me to gain with a buffer stage or should I just be happy and go with the RAW output from the dam? It'll sure make for a more compact, easier, cheaper and more efficient build. I'm waiting for the 1231 (Pi-dam) to become available.

Maybe it depends. If you have long runs of cable from your DAC to your poweramp, maybe you can benefit from a low impedance output. If not, then it should be ok, after all the 100k input impedance is over 100 times larger than the output impedance.
I heard that the ncores benefit from balanced drive, so maybe you consider to add a balancing output transformer after your DAC. In this case a buffer would be totally justified because output transformers very much like low impedance sources.
 
I know, it would be more convenient, but digital volume controls are normally not bit perfect. Usually bit depth is lost the more you attenuate. Is there somewhere explained how this works in the dam1021? I am happy for any pointers. I only recall someone in the dam-thread measuring THD at different volume settings, and THD rises substantially with increasing attenuation.

My thinking is just: Yes, there is increased distortion and channel imbalance with analog potis, but this is neglible with quality parts. Probably as neglible as the increased distortion that was measured with the dam digital attenuation.

That is true and I agree with that. However as far as I can recall Soren states that with 28bit depth of soekris at -70db level stil first 16bit are played. You can find it somewhere in soekris related thread.
 
They have differential input, so you can just have signal ground as the second input and that'll give you the balanced benefits. My current dac is using trafos on the output but the dam doesn't seem to need it?

you can just use ground as second input, and this should work. While the dam does not need a trafo output, a trafo would balance the output impedances for true balanced operation, and apparently the hypex amps benefit from this. But, if it works for you without buffer and trafo, then it should be also no problem.
 
Since the 1231 hasn't been launched yet, I'm still waiting. At the moment I'm planning the coming build. My old dac uses trafos...
I have schematics for a tube buffer, a DCB1 PCB and hooking a trafo up is a easy thing to do. However, if there is nothing to be gained, why spend the money? I try to focus on building stuff that improves the sound... ;)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.