[cjl]

   
It can... badly. Have a look here, fig. 4.32, 4.33. And that's 3Vrms into 1K.

That's still around 0.01% THD, which isn't audible, and its worst at extremely high frequencies (where it really, really isn't audible). Sure, it's pretty mediocre by modern electronics standards, but you aren't going to hear any serious flaws in your music just because the preamp is adding 0.005%THD to most of the spectrum and 0.02% to the top octave.

 

[ab initio]

   
It can... badly. Have a look here, fig. 4.32, 4.33. And that's 3Vrms into 1K.

 
Take a look at the scale bars! This is given as a log plot of the percent THD and all values are much less than 0.1 %.
 

 
I'm not sure that would be audible at all.
 
Cheers

 

[Steve Eddy]

It can... badly. Have a look here, fig. 4.32, 4.33. And that's 3Vrms into 1K.

Into 600 ohms, it would likely hit about 0.02% at 20kHz. That wouldn't even bring it into the realm of audibility.

se

 

[00940]

Quote:

   
Take a look at the scale bars! This is given as a log plot of the percent THD and all values are much less than 0.1 %.

 
Wow, I would never have figured out by myself how to read those graphs 

 
 
I don't have the time to get into yet another debate. When I wrote a significant increase, I had in mind an increase by an order of magnitude. When he read significant increase, Steve Eddy had in mind an audible increase. Not the same thing and let's leave it at that.
 
 

Into 600 ohms, it would likely hit about 0.02% at 20kHz. That wouldn't even bring it into the realm of audibility.
 

 
It's already 0.025% with a 1K load in Self's setup (3Vrms @20khz). I suppose you mean 0.02% at 2Vrms, a standard dac output ?
 
S. Groner has the measurements at +20dBu btw, if you're interested.

 

[Steve Eddy]

It's already 0.025% with a 1K load in Self's setup (3Vrms @20khz). I suppose you mean 0.02% at 2Vrms, a standard dac output ?

No, I meant 3 VRMS. And I erred, I meant to say about 0.03% THD, not 0.02%. And again, that's at 20kHz. What's the second harmonic of 20kHz? Can you hear 40kHz?

Sure, if you're just wanting to play the numbers, the TL071 doesn't fare as well as others into 600 ohms. But outside the numbers, is it really meaningful?

S. Groner has the measurements at +20dBu btw, if you're interested.

Thanks, but I think these numbers make my point just fine.

se

 

[Audio Jester]

What have I done... (@, @ )!

 

[SilverEars]

What have I done... (@, @ )!

This is what you get for asking questions here.

   Hope you learned your lesson.  
Yup, you created this ambiance.  

  kidding.  

 

[Reggy]

The ratio of 10 is a minimum, a lot more isn't a problem usually. The ratio is more useful to see if an output impedance of a DAC/preamp is too high; it's not very helpful when the output impedance is very low.

Btw, the output impedance alone is not a good indication of the ability of the DAC to drive an amp. Many DAC have rather weak output buffers (or even no dedicated buffer) and are not able to drive loads under 10k properly or, if they can, distortion will rise significantly at lower loads.

Professionnal gear is usually designed to drive 600R inputs if needed, hifi gear is more usually designed for a minimum of 10K.

Your 10:1 amplifier rule is really for devices that need power. There are electrical dampening issues that can pop up, but what happens is the output impedance of one device will create a voltage divider when paired to the input impedance (or driver impedance) of another device. If the impedances of your headphones and amplifier are the same, then the voltage output of the amplifier will essentially be cut in half. From a power perspective, this makes it more difficult to drive a pair of headphones even if your amp can supply a lot of power.

With a DAC, you actually want a theoretically infinite input impedance on the amp (input not output). The output voltage of the DAC is already small, and this pairing ensures that the signal isn't degraded by the connection. At 133:1, the pairing will be good.

My thoughts:
a] The 600 Ohm matched output to input system is ancient history from the broadcasting & pro audio industries. The only equipment made in the last 50 years using this system are legacy or boutique units.
b] The typical concern with the 10:1 ratio is when using high output impedance vacuum tube equipment to drive low input impedance solid state equipment.
c] There are no 'electrical dampening issues' in any interconnect system.

Hey guys, I read your posts which were useful, and I am asking your opinion on my setup which has 8:1 factor. I am wondering if a passive preamp to accompany my Ares DAC would lead to poor dynamics and losses would it? I will be using a Denafrips amplifier in conjunction with my Genelec 8040b's, however I need a volume control as it doesn't have one. I am debating on whether passive or active preamp is the way to go.

The specs of the Ares DAC and Genelec 8040B are as follows:

Analog Output: XLR at 4.4Vrms +-10% (Peak voltage calculated to be 6.22 volts)
Output Impedance of 1250 Ohms

Genelec 8040B Specs:
Crossover Section:
10 kOhm input impedance
Input level for maximum short term output of 100 cB SPL @ 1 M: Adjustable +6 to -6 dB (or 0.39 volts to 1.55 volts).

If you could take a look that would be extremely helpful to get your opinion on whether it could drive my active studio monitors. Thanks.

 

[drtechno]

Hey guys, I read your posts which were useful, and I am asking your opinion on my setup which has 8:1 factor. I am wondering if a passive preamp to accompany my Ares DAC would lead to poor dynamics and losses would it? I will be using a Denafrips amplifier in conjunction with my Genelec 8040b's, however I need a volume control as it doesn't have one. I am debating on whether passive or active preamp is the way to go.

The specs of the Ares DAC and Genelec 8040B are as follows:

Analog Output: XLR at 4.4Vrms +-10% (Peak voltage calculated to be 6.22 volts)
Output Impedance of 1250 Ohms

Genelec 8040B Specs:
Crossover Section:
10 kOhm input impedance
Input level for maximum short term output of 100 cB SPL @ 1 M: Adjustable +6 to -6 dB (or 0.39 volts to 1.55 volts).

If you could take a look that would be extremely helpful to get your opinion on whether it could drive my active studio monitors. Thanks.

passive 10K attenuator will give you the maximum bandwidth at the lowest noise on most electrically balanced signal lines.

 

[Reggy]

passive 10K attenuator will give you the maximum bandwidth at the lowest noise on most electrically balanced signal lines.

Thanks, but have you had any experience with the limitations of a passive pre-amp, if you read my original concern/question you will see my Source/Load situation that may need to be addressed (ie 10K pre-amp does nothing if it may not work adequately with the potential impedance matching problems I have). I still believe in the abundance of knowledge in the audio world I can still get a great working system given my choice of DAC.

This link may be useful to read, I just need some direction in where to find the right pre-amp:

http://audiokarma.org/forums/index.php?threads/passive-pre-table-regular-vs-haller-shunt.556704/

To anyone with an acquired knowledge of the combinations of source output impedance, potentiometer impedance and amplifier input impedance an how to properly implement these parameters and recommend the right product.

Thanks.

 

[drtechno]

Thanks, but have you had any experience with the limitations of a passive pre-amp, if you read my original concern/question you will see my Source/Load situation that may need to be addressed (ie 10K pre-amp does nothing if it may not work adequately with the potential impedance matching problems I have). I still believe in the abundance of knowledge in the audio world I can still get a great working system given my choice of DAC.

This link may be useful to read, I just need some direction in where to find the right pre-amp:

http://audiokarma.org/forums/index.php?threads/passive-pre-table-regular-vs-haller-shunt.556704/

To anyone with an acquired knowledge of the combinations of source output impedance, potentiometer impedance and amplifier input impedance an how to properly implement these parameters and recommend the right product.

Thanks.

just use a 10K balanced passive attenuator. otherwise by a passive monitor controller. There won't be any loss, because the dac is not driving into a lower impedance load that would cause lack of bass response. btw the thread you are referring to is for an unbalanced circuit. The two types that are used are the PI atternuator ( also called "o" attenuator), and "T" type. Typically T-types are used for adjustable attenuators. The monitors are 10K on the input. That is why it states 10K. Also, you have to understand a couple of things about the output stages, the output impedance is the smallest impedance that will guarantee the specifications. Its always better to find the optimal load, on tl071, tl074, types 10K seems to be the optimal load and the norm in electrically balanced systems, but this is not a fixed point somewhere as 2K-10K is the final resulting impedance in an balanced chain (in your system, the terminating resistance on the output of the dac in parallel to the input of the genlecs that we put in parallel an attenuator). I've seen zillions of times interconnecting different pro gear, 600 ohms is the norm for transformer balanced systems this impedance interconnect is a fixed point. Btw, I wire radio stations, pa systems, and pro recording studios for the past 20 years. yes there are other impedances, but 10K is not going to cause the tl071 to pull excessive current and cause white noise, low impedance will. There's maybe 35 different studio setups I've done and about half of them have a 10K attenuators because of these powered monitors. I've hooked up your model before this way, along with Yamaha hs80, Neumann KH80... etc, all of those self powers are like this. So if I go by the 10X rule, the terminating resistance is 12.5K on the dac's output, then 10K is parallel, to another 10K and if we do the parallel formula, we get 3.57K as our ending impedance.

Just giving you some professional advice, recommendations, and a little bit of whats going on in the background for you....

edit: here is some light reading on the subject:
http://www.radio-electronics.com/in...n/attenuators/balanced-rf-attenuator-pads.php

and

https://en.wikipedia.org/wiki/Attenuator_(electronics)

Unless you want to build one out of some high end resistors (like z-foil resistors), a standard, off the shelf 10k attenuator is going to work.

 

[tansand]

Thanks for that, drtechno, that cleared something up for me. Reggy, the XLR luminous is a Pi attenuator, and the RCA ones are shunt/haller volume controls.

drtechno, is it possible that some of the passive monitor controllers are just a regular volume control across the positive and negative wires, rather than a Pi variable attenuator? Would that work?

 

[71 dB]

Wow, I would never have figured out by myself how to read those graphs.

Being able to read these graphs is part of literacy, more precisely scientific literacy.
A lot of information is given in this form to people and if they can't understand it, the information doesn't come over.
Lack of scientific literacy is a sign of failed education system.

 

[Speedskater]

Stereophile magazine often tests the line outputs of components that were designed to work into a 10K load with a 600 Ohm load (why they do this , I don't know). But most components do almost OK, even with this heavy overload. So don't put too much into that 10 to 1 ratio.

 

[tansand]

Being able to read these graphs is part of literacy, more precisely scientific literacy.
A lot of information is given in this form to people and if they can't understand it, the information doesn't come over.
Lack of scientific literacy is a sign of failed education system.

Scientific literacy isn't neccessary, most people are fine without it. Claiming otherwise is just some kind of reflexive elitism.