DAC volume control conundrum

[cat6man]

Hi folks,

I'm upgrading my system and have been on a DAC search for a few months now. I'll regale you with the saga sometime in the future, but right now I need some help/advice.

I'm down to two DACs, the TotalDAC D1-dual and an MSB platinum (options TBD).
Until I heard the D1-dual, the MSB was the best I had heard in my system. I have yet to do a direct comparison between the two but I am certain I could be happy with either.

My setup is pretty simple: the DAC outputs need to feed two paths, one for speakers and one for headphones.
The speaker path is about 8m of audioquest single ended cable to VTL deluxe 300 monoblocks to Nola metro-grand speakers.
The headphone path is balanced out of DAC to Headamp BHSE to Stax 009.

My problem: digital volume control of 20-30dB is the spawn of the devil, at least as compared with no digital attenuation. I assume this surprises absolutely no one on this forum (I've been lurking for years).

My recollection of the MSB in my system (compared that day with Ayre, Bricasti, Romulus) was that the MSB volume control (platinum version, switched resistors I assume) significantly whomped the others (caveat: Romulus did not have it's analog volume output stage installed, so volume was cut in JRiver).

Right now I have the TotalDAC and with volume control set to 0 or that vicinity, the sound on my Stax is as good as I have ever heard........a clear keeper.
However, with the VTL/Nola, I need to add about 30db attenuation on the DAC and the magic says bye-bye.
To corroborate my theory, I did some additional listening with the Stax setup with the DAC at 0db and 28dB of digital attenuation, with the BHSE volume control used to match the volume at the Stax...............no contest as expected, the 0dB setting on the DAC magical and the 28dB setting with a flattened soundstage, less air, detail and low end.

As many have already advised, a good preamp will solve this problem.................but I wanted to simplify, reduce components and cables, etc.
So option 1 is to compare the MSB directly to the TotalDAC and see if 1) it is as good or better with the Stax and 2) assuming the first part, see if the MSB
volume control solves my problem and is much more transparent (I expect it will be but need to check).

The MSB option, of course, comes as a cost which is not a lot different than adding a good preamp.
Last night I did a little searching and found a potential solution that I wanted to run by folks here.
What do you folks think of the Luminous Audio Axiom II 'Walker Mod' as a passive, single input, single output volume setting for the single ended speaker path?
I imagine setting it to an approximate value of somewhere in the 20db-30db range and using the TotalDAC volume control for the remote controlled final 5-10db of volume adjustment.

Do you think that would get me most of the way there (i.e. buy back the big losses in sound quality due to 30db of digital attenuation)?

Cheers and thanks in advance for the advice.

 

[Empirical]

Right now I have the TotalDAC and with volume control set to 0 or that vicinity, the sound on my Stax is as good as I have ever heard........a clear keeper.
However, with the VTL/Nola, I need to add about 30db attenuation on the DAC and the magic says bye-bye.
To corroborate my theory, I did some additional listening with the Stax setup with the DAC at 0db and 28dB of digital attenuation, with the BHSE volume control used to match the volume at the Stax...............no contest as expected, the 0dB setting on the DAC magical and the 28dB setting with a flattened soundstage, less air, detail and low end.

As many have already advised, a good preamp will solve this problem.................but I wanted to simplify, reduce components and cables, etc.

You might consider a transformer linestage. A good one like Music First will actually make things sound better because of the galvanic isolation it provides. If you are running balanced cables, it will improve SQ even more.

Steve N.
Empirical Audio

 

[audioarcher]

Try out a Exasound e20 with the .82 pico second clock option. It has a unique volume control according to the designer. It is neither digital based nor resistor based. He has created software to lower the dac chips output voltage instead of lowering it in the digital domain. I have one and it works great. There is no loss of sound quality at any volume level. It has an 80 db volume control range in .5 db steps and individual control of each channel for balance control. It's overall sound quality is also one of the best I've heard. They are based in Canada but they have a 30 day free trial excluding shipping cost. Highly recommended!

 

[microstrip]

You might consider a transformer linestage. (...)

Excellent advice. A transformer based solution should solve your problem. Lundahl transformers has high quality line models with several windings that can be configured by the user to obtain the desired attenuation. They are used in some famous DACs.

 

[Empirical]

Try out a Exasound e20 with the .82 pico second clock option. It has a unique volume control according to the designer. It is neither digital based nor resistor based. He has created software to lower the dac chips output voltage instead of lowering it in the digital domain.

They are using the volume control built into the Sabre chip. Its not a custom hardware implementation.

 

[audioarcher]

They are using the volume control built into the Sabre chip. Its not a custom hardware implementation.

That's interesting. Sounds transparent to me.

 

[Empirical]

Excellent advice. A transformer based solution should solve your problem. Lundahl transformers has high quality line models with several windings that can be configured by the user to obtain the desired attenuation. They are used in some famous DACs.

Here is the best on the planet, and not too expensive:

http://www.ebay.com/itm/Teramoto-FINEMET-Premium-Transformer-Volume-Control-TVC-with-Seiden-Switch-/200870484580?pt=Vintage_Electronics_R2&hash=item2ec4d05a64

Steve N.

 

[DonH50]

Why not a standard resistive stepped attenuator at the output of the DAC? No worries about transformer noise pickup, frequency limiting, saturation and hysteresis, or ringing...

 

[microstrip]

Why not a standard resistive stepped attenuator at the output of the DAC? No worries about transformer noise pickup, frequency limiting, saturation and hysteresis, or ringing...

Just a subjective opinion, but I have owned VTL MB450s (quite similar to the MB300) and still own a very good passive attenuator - the ProPASSION built with Vishay foil resistors that I had tried several times with the VTL. Unfortunately the VTL tube amplifiers, although having high impedance (RCA only input ,around 220 kohm, if I remember well) never sounded good with a resistive stepped attenuator, but sounded very good with the Steve McCormak FlexConnect transformer/stepdown unbalancer. The passive made it sound overall less dynamic and veiled at low levels. YMMV, surely.

 

[cat6man]

Thanks. I never considered transformer options at all.
The McCormack sounds interesting but specs at -12dB attenuation, and that may not be sufficient since the totaldac needed to be around -30db or lower.
The MSB ran around -25dB if I remember correctly, which is consistent with the 2V single ended output of the MSB versus the 3V for the totaldac.

Anyone with an MSB able to comment on the sound of the MSB attenuator (platinum, signature or diamond) at -20 to -30?

 

[cat6man]

Just a subjective opinion, but I have owned VTL MB450s (quite similar to the MB300) and still own a very good passive attenuator - the ProPASSION built with Vishay foil resistors that I had tried several times with the VTL. Unfortunately the VTL tube amplifiers, although having high impedance (RCA only input ,around 220 kohm, if I remember well) never sounded good with a resistive stepped attenuator, but sounded very good with the Steve McCormak FlexConnect transformer/stepdown unbalancer. The passive made it sound overall less dynamic and veiled at low levels. YMMV, surely.

Excellent point and reminds me that everything matters and EMMV (everyone's mileasge may vary).

 

[Empirical]

Why not a standard resistive stepped attenuator at the output of the DAC? No worries about transformer noise pickup, frequency limiting, saturation and hysteresis, or ringing...

Because resistive attenuators rob the amps of transient current needed for good dynamics. You get good detail, but you sacrifice dynamics. They can also make a nice low-pass filter, robbing you of the HF information depending on the capacitance and length of interconnects. Transformers have none of these drawbacks.

Steve N.

 

[Empirical]

Thanks. I never considered transformer options at all.
The McCormack sounds interesting but specs at -12dB attenuation, and that may not be sufficient since the totaldac needed to be around -30db or lower.
The MSB ran around -25dB if I remember correctly, which is consistent with the 2V single ended output of the MSB versus the 3V for the totaldac.

Anyone with an MSB able to comment on the sound of the MSB attenuator (platinum, signature or diamond) at -20 to -30?

Forget about all other attenuators. I am telling you that the ebay transformers are the best on the planet. You will not find anything better. I have a all-silver wound $8K Music First TVC and the ebay transformers beat it. Its a no-brainer.
Steve N.

 

[Alrainbow]

Correct me if I am wrong , but wouldn't you need four of them in a balance configuration ?

 

[DonH50]

Because resistive attenuators rob the amps of transient current needed for good dynamics. You get good detail, but you sacrifice dynamics. They can also make a nice low-pass filter, robbing you of the HF information depending on the capacitance and length of interconnects. Transformers have none of these drawbacks.

Steve N.

A resistive attenuator before the amp robs it of dynamics? That makes no sense to me given the relative impedances involved but whatever. Transformers can also have limited bandwidth on both ends (LF and HF), by the way. And present a reactive load and source to the preamp/amp that could cause other artifacts, and they have hysteresis etc. Whether any of that matters for the transformers under discussion I have no idea. I keep stepping into religious debates, sorry for the diversion. Need to stay the heck off this place...

 

[Empirical]

Correct me if I am wrong , but wouldn't you need four of them in a balance configuration ?

You can do with 2. These are autoformers I believe, so I would want a 1:1 after it.

Steve N.

 

[Empirical]

A resistive attenuator before the amp robs it of dynamics? That makes no sense to me given the relative impedances involved but whatever. Transformers can also have limited bandwidth on both ends (LF and HF), by the way. And present a reactive load and source to the preamp/amp that could cause other artifacts, and they have hysteresis etc. Whether any of that matters for the transformers under discussion I have no idea. I keep stepping into religious debates, sorry for the diversion. Need to stay the heck off this place...

This has nothing to do with impedance. Has everything to do with the active input devices and how they react to transient voltages. Most amps need some current to be dynamic, not just voltage. I know what I'm talking about here. I have 30+ years experience with lots of amps and different passive attenuators and active preamps.

Hysteresis with these transformers is a non-issue. These are virtually impossible to saturate. Like I said, they are simply the best. They use a new core technology.

Steve N.

 

[cat6man]

Conundrum temporarily solved. I just purchased a Pass Aleph P for a good price and will give up my quixotic search for the perfect DAC volume control. I really didn't want to get into MSB price range if possible, but you all know how dangerous it can be to hear something that really makes a difference to your ears................and down the rabbit hole we go.

If I can get most of the great TotalDAC D1-dual sound I hear with my stax setup by adding the Pass to my vtl/nola, I keep the DAC and tweak and optimize at a later time.

I've spent enough time on this particular roller coaster, found that I could happily live with both the TotalDAC D1-Dual and MSB, so it is time to go back to listening and enjoying music mode. I'll let you know how the Aleph P works next weekend.

Cheers and thanks for the advice.

 

[dallasjustice]

I know you have resolved your issue. I just noticed this thread so I could offer my experience with the totaldac.

I think you may have to do some "gain matching" whenever you decide you want to use digital volume control. This is a system thing so you have to consider how the entire system gets better or worse with a specific implementation. My experience with digital volume control is that dithering is an artform, not just a number of bits game. I think it's best to listen carefully and decide what it is that you are losing by attenuating -20 or -30db. I don't think it will be naturalness or spaciousness when it comes to how Vincent does his volume control. If anything, you would hear a loss of system dynamics. However, unless your room has something north of 100db of dynamic range, you will not hear any loss of dynamics. Most rooms don't have any dynamic range north of 90db. This is easy to measure with a microphone. It's simple math.

However, not all dithered volume controls sound the same. I know Bob Katz has dealt with this topic extensively and I've done my own blind ABX comparing totaldac's digital volume control against the Jriver 64bit dithered volume control. My local dealer could pick out the totaldac volume control everytime when only attenuating -11db in my ABX tests.

It's funny because the folks that like to criticize digital volume control don't seem to acknowledge the loss of transparency that even the best pre-amps will introduced when interposed between a DAC with a GREAT analog output stage (like the totaldac). I've done my homework. I compared super nice pre-amps against the totaldac direct, and the difference is pretty big in favor of the totaldac.

 

[Empirical]

I know you have resolved your issue. I just noticed this thread so I could offer my experience with the totaldac.

I think you may have to do some "gain matching" whenever you decide you want to use digital volume control. This is a system thing so you have to consider how the entire system gets better or worse with a specific implementation. My experience with digital volume control is that dithering is an artform, not just a number of bits game. I think it's best to listen carefully and decide what it is that you are losing by attenuating -20 or -30db. I don't think it will be naturalness or spaciousness when it comes to how Vincent does his volume control. If anything, you would hear a loss of system dynamics. However, unless your room has something north of 100db of dynamic range, you will not hear any loss of dynamics. Most rooms don't have any dynamic range north of 90db. This is easy to measure with a microphone. It's simple math.

However, not all dithered volume controls sound the same. I know Bob Katz has dealt with this topic extensively and I've done my own blind ABX comparing totaldac's digital volume control against the Jriver 64bit dithered volume control. My local dealer could pick out the totaldac volume control everytime when only attenuating -11db in my ABX tests.

It's funny because the folks that like to criticize digital volume control don't seem to acknowledge the loss of transparency that even the best pre-amps will introduced when interposed between a DAC with a GREAT analog output stage (like the totaldac). I've done my homework. I compared super nice pre-amps against the totaldac direct, and the difference is pretty big in favor of the totaldac.

This is definitely the best way to do digital audio, but there is one caveat: the output driver and volume control technology in the DAC must be up to snuff. A weak output driver using op-amps will kill this, as well as a poorly implemented volume control technology. Using 100% DSP volume control is also not advised. Usually you get some degradation in detail.

This is the reason that I developed a post-DAC transformer selector buffer, so that one could easily bring-in home theater or vinyl AFTER a direct-connected DAC.

Steve N.