MiniDSP -> Passive preamp/potentiometer. Impedance matching? - Page 2 - diyAudio
Go Back   Home > Forums > Source & Line > Digital Line Level
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Thread Tools Search this Thread
Old 25th August 2011, 02:58 PM   #11
diyAudio Member
Join Date: Aug 2010
Interesting discussion!

You will definitely loose resolution (bits) when attenuating in digital domain, however as qusb mentioned you will also loose resolution attenuating in the analog domain.

For me the question is, is analog attenuation (if done right!) really by far lossier as qusb mentioned?

As a first approach I will take a look to the datasheets (e.g. PGA, Codec,...): e.g. THD+N at -60dB.

Next step will be doing some measurements. I am about to set up a "preamplifier" with miniDSP 2x8 and "optional" analog attenuation (will try different things like PGA, Pot, stepped attenuator, w or w/o buffer) and plan to do some measurements (Though might take some weeks or months, 2 kids are so time consuming ;-)). Lets see if my equippment can find out some obvious differences for low output signal levels. One thing is clear: analog attenuation if (far) more effort! Build in attenuation of miniDSP is really easy.
  Reply With Quote
Old 25th August 2011, 06:58 PM   #12
diyAudio Member
Join Date: Sep 2009
Thanks for all your replies, yeah as I said "perhaps i'm just paranoid", what I meant by that statement is that I'm not "religious" about my opinions at all
But as I said, my biggest concern was the effect (if at all noticable) when using high efficiency drivers, like the Beyma TPL150 with it's 99dB/1w/1m efficiency, considering it's a 10dB "head start" over many typical non-pro drivers, and 10dB is often mentioned in discussions around digital attenuation.

If I understand correctly you loose 1 bit for every 6dB of attenuation in the digital domain, but with 24bits processing, you can attenuate -48dB and still have 16bits left, so yes, I was being paranoid And then there's dithering methods, not sure if that applies to minidsp or not.
Double-blind study would probably confirm what you're all saying, that it's irrelevant and unnoticable and that loss of dynamic range with analogue attenuation is equally lossy as digital.

curryman : That would be a interesting experiment! I seem to recall Meridian using some kind of hybrid at one point, that is hybrid digital/analog attenuation. There's probably lots of ways of doing this, perhaps even mechanically (i.e. linking the "digital" pot (the one connected to the MINIDSP board) mechanically with an "analogue" pot would perhaps be one way of doing it. Question is if it's worth the hassle.

For my own project I'll be sure to go the digital-domain route since I'm a newbie at DIY audio and I don't want to become overwhelmed by complexity
  Reply With Quote
Old 25th August 2011, 08:02 PM   #13
qusp is offline qusp  Australia
diyAudio Member
qusp's Avatar
Join Date: Oct 2009
Location: Brisbane, Australia
curryman, for sure its the lesser of the 2 evils, ive been using 40 and 64bit attenuation on my sabre dac and usb interface for years now and theres no turning back. good luck even getting close with a 200 dollar pot. with these methods an ADC is used to read a reference voltage applied through a resistor, so you can use a cheapie carbon pot, a relay based attenuator, or as ive done as well you can make a simple stepped attenuator, single layer with smd resistors and this works really well, the quality of the resistive element matters nothing to the sound, but the action and feel is nicer. it takes a little getting used to the perfect channel matching all the way down, a hard ask for an analogue pot of any value
  Reply With Quote
Old 26th August 2011, 06:07 AM   #14
tnargs is offline tnargs  Australia
diyAudio Member
Join Date: Sep 2006
Originally Posted by curryman View Post
...however as qusb mentioned you will also loose resolution attenuating in the analog domain....
actually that was me.
  Reply With Quote
Old 26th August 2011, 07:11 AM   #15
diyAudio Member
Join Date: Aug 2010
Hi all,
I totally agree that digital attenuation is the far easier solution (only one simple pot) and automatically gives good channel matching which is not easy to achieve with pots or stepped attenuators (PGA might be better) and I also will definately go this way first!

However looking specifically at the miniDSP solution of digital attenuation (by the way: done in DSP or in AD/DA Codec? I'll ask miniDSP devteam) it is not working with 64bits like sabre DAC and the DAC is not 32bit resolution. Don't get me wrong: I don't claim that analog is better! I am not sure if digital is better either. I just want to get a feeling and understanding by doing measurements and learning the theory behind. In the end it's definitely some kind of paranoia

In general gain structure is very important doing signal processing in the digital domain. Just thinking about the tweeter channel: put a highpass filter, say 2,5 kHz to a typical music signal not much signal energy is left. If you attenuate this signal by 40dB you might get into trouble even with 24bit resolution. Thus it makes sense to think about it a little bit ;-)

@tnargs: sorry, missed it

Last edited by curryman; 26th August 2011 at 07:12 AM. Reason: layout
  Reply With Quote
Old 26th August 2011, 03:04 PM   #16
qusp is offline qusp  Australia
diyAudio Member
qusp's Avatar
Join Date: Oct 2009
Location: Brisbane, Australia
oh sure, its not as advanced in the minidsp, but there was thorough testing done at a recent audio meet here in australia with minidsp specifically and the guys there ended up not being able to discern a difference between high end preamp with analogue attenuation and the internal digital pot. so i think its wise you try that first.

btw the sabre uses 40bit and i've been using that the longest, my multichannel usb ->i2s interface (titan) however uses 64bits, this is only a recent addition to my system, I cant discern the difference between them as you would expect, but as titan is the source i find it more convenient to use that, its also multichannel and allows me to control all dacs simultaneously.

oh and yeah i was just tnargs's backup
  Reply With Quote


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

Similar Threads
Thread Thread Starter Forum Replies Last Post
Impedance matching; phono > pot > A/D convertor? Aleksunder Analog Line Level 4 25th July 2007 06:29 PM
Passive Crossovers >> Capacitors Question Beggar Multi-Way 7 15th December 2005 03:20 AM
Need help understanding transformer impedance ratios and impedance matching percy Tubes / Valves 5 28th February 2005 08:35 PM
Matching a preamp to high output impedance source ainami Tubes / Valves 5 29th November 2004 04:02 PM
Matching BOSOZ > SOZ impedance Roberto Amato Pass Labs 2 30th August 2002 10:14 PM

New To Site? Need Help?

All times are GMT. The time now is 06:55 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2