Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Digital Volume Control, Bad??
New Posts  All Forums:Forum Nav:

Digital Volume Control, Bad??

post #1 of 11
12/21/09 at 2:49pm
Thread Starter 
What's the drawback to using this for vol control? How does it control volume?

LED Display/Digi Remote Control DIY PCB w/parts /CS3310 - eBay (item 290379114391 end time Jan-06-10 17:47:43 PST)
Reply
post #2 of 11
12/21/09 at 4:16pm
So judging by the Cirrus CS3310 datasheet, it looks like its use adds to the THD at the input and potentially increases cross talk. Also, because the CS3310 doesn't exclusively attenuate the signal, it could color the sound somewhat.

Honestly, the fact that it's bigger than a stepped attenuator would be my biggest complaint.

From what I gather, it works by having that microcontroller trigger some relays for source selection. The outputs of those relays are wired up to the appropriate pins on the CS3310. Internally, the CS3310 chooses (via a multiplexer) the appropriate resistor values for the feedback loop and source of an internal opamp to give the correct attenuation or gain. The output from the CS3310 is then exposed as through-hole pins on the board.

This works based on the same principles used to set the gain in opamp based amplifiers, such as the CMoy. To use the CMoy as an analogy, in reference to Tangent's CMoy schematic, R2 and R3 are selected by the control register of the CS3310 and potentially some other internal logic.
Reply
post #3 of 11
12/21/09 at 4:29pm
Thread Starter 
Quote:
Originally Posted by jezz View Post
So judging by the Cirrus CS3310 datasheet, it looks like its use adds to the THD at the input and potentially increases cross talk. Also, because the CS3310 doesn't exclusively attenuate the signal, it could color the sound somewhat.

Honestly, the fact that it's bigger than a stepped attenuator would be my biggest complaint.

From what I gather, it works by having that microcontroller trigger some relays for source selection. The outputs of those relays are wired up to the appropriate pins on the CS3310. Internally, the CS3310 chooses (via a multiplexer) the appropriate resistor values for the feedback loop and source of an internal opamp to give the correct attenuation or gain. The output from the CS3310 is then exposed as through-hole pins on the board.

This works based on the same principles used to set the gain in opamp based amplifiers, such as the CMoy. To use the CMoy as an analogy, in reference to Tangent's CMoy schematic, R2 and R3 are selected by the control register of the CS3310 and potentially some other internal logic.

Hmm, Thanx!

The size is not an issue, it would be attached to the front plate. May affect where the Jacks go, I'll think about that one.

The exposure of the PINS seems a non issue to me. The alternate way would be to run Long input wires from the inputs to the attenuator, most likely mounted on the front of the amp. So, Short traces seems better there. I'm interested in the coloration obviously. I'm also not interested in it doing any gain as well. You sure about the gain possibility? Does the input pass thru the chip do you think?
Reply
post #4 of 11
12/21/09 at 5:11pm
If you get one, switch the chip from the Crystal Semi to the TI PGA2311. Pin for pin compatible and better sounding.
Reply
post #5 of 11
12/21/09 at 5:20pm
Thread Starter 
Quote:
Originally Posted by BrianDonegan View Post
If you get one, switch the chip from the Crystal Semi to the TI PGA2311. Pin for pin compatible and better sounding.
Since you look like you're a TP fan, what is your assessment of this gadget?
Reply
post #6 of 11
12/21/09 at 5:50pm
Quote:
Since you look like you're a TP fan
Co-owner.

Quote:
what is your assessment of this gadget?
Well, I don't understand why it is so big... could be a LOT smaller. After using the CS/PGA chips for a while (Kookaburra, then some products we never released), we have stopped altogether. They sound very good at lower volume levels, but once they approach the 0dB mark (and above) they get hissy. I can't really see the layout, but judging by the component locations, probably less then ideal.

There are a few DIY designs floating around if you search for PGA2310 and PGA2311.
Reply
post #7 of 11
12/21/09 at 10:16pm
Thread Starter 
Quote:
Originally Posted by BrianDonegan View Post
Co-owner.
Heh, I totally didn't read your name when you posted! Just noticed the Avatar!


Quote:
Well, I don't understand why it is so big... could be a LOT smaller. After using the CS/PGA chips for a while (Kookaburra, then some products we never released), we have stopped altogether. They sound very good at lower volume levels, but once they approach the 0dB mark (and above) they get hissy. I can't really see the layout, but judging by the component locations, probably less then ideal.

There are a few DIY designs floating around if you search for PGA2310 and PGA2311.
Interesting, thanx!
Reply
post #8 of 11
12/21/09 at 10:45pm
Quote:
I can't really see the layout, but judging by the component locations, probably less then ideal.
Actually, after having another look, the components layout doesn't look that bad. I should have spent a little more time looking before commenting.
Reply
post #9 of 11
12/22/09 at 1:51am
Quote:
Originally Posted by les_garten View Post
You sure about the gain possibility? Does the input pass thru the chip do you think?
Yeah, I'm pretty sure about the gain. Check out the datasheet; register settings of 11000000b (0xC0 = 192) and up will set up the opamp to give between 0dB and +31.5dB of gain. It even says, in the input code definition on page 8, "Gain or Attenuation (dB)." Basically, you only attenuate for about 3/4 of the range of the acceptable input values, and the other 1/4 is gain.

I might ask the seller if they can leave the CS3310 unpopulated to add in the PGA2311 yourself, or perhaps ask BrianDonegan if he has any Kookaburra boards lying around .
Reply
post #10 of 11
12/22/09 at 2:10am
Quote:
Originally Posted by jezz View Post
because the CS3310 doesn't exclusively attenuate the signal, it could color the sound somewhat.
Yes. The CS3310 and PGA231x are basically digitally-controlled op-amp gain blocks. That means there are additional op-amps in your circuit that you can't control, other than switching between these two brands of chips. If you don't like either chips' sound, you're stuck.

There are purely attenuating digital volume control chips, which are essentially digital stepped attenuators. The DS18xx series from Maxim, for instance. How close they come to traditional SAs in sound, I can't say because I didn't spend much time listening to them.
Reply
post #11 of 11
12/22/09 at 2:15am
Thread Starter 
Quote:
Originally Posted by tangent View Post
Yes. The CS3310 and PGA231x are basically digitally-controlled op-amp gain blocks. That means there are additional op-amps in your circuit that you can't control, other than switching between these two brands of chips. If you don't like either chips' sound, you're stuck.

There are purely attenuating digital volume control chips, which are essentially digital stepped attenuators. The DS18xx series from Maxim, for instance. How close they come to traditional SAs in sound, I can't say because I didn't spend much time listening to them.
Hey, this is exactly what I was gleaning from the info you guys posted here. Don't think I'm interested in having an OPAMP in line. Thanx!
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Digital Volume Control, Bad??