Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › making my own volume attenuator?
New Posts  All Forums:Forum Nav:

making my own volume attenuator?

post #1 of 30
3/9/10 at 7:05pm
Thread Starter 
hi there,

I like to listen at very low volume to my uber-sensitive 32Ω/104dB cd1k(same drivers as the cd3k), and I currently control volume digitally(in 64float) w/ a Griffin Powermate



I use a Firestone Spitfire DAC fed w/ toslink, so that's my only option...it was so handy to set the Powermate on the (hardware accelerated) analog output of a soundcard, but over S/PDIF that's a no go.

anyway, if I connect my headphones directly to my DAC and attenuate in my audio player(over KS) to 0-3%...the sound is way too bright/edgy, I'd guess that the LT1028AC DAC LPF opamps are grossly underpowered?

it goes a lot better w/ the infamous shure attenuator, but there's some major stereo imbalance at low volume:



and the Burson HA-160 amp that's bound to ship to me anytime soon has a stepped attenuator...so it's also vastly innacurate for low volume listening.

anyway! I understand that stepless analog volume attenuation is a terrible idea, but that happens to be exactly what I need/want

I'd like to DIY something like this, w/ 1/8" input/outputs: Ortofon HD-Q7 headphone amplifier



the ALPS RK50 is just way overpriced...what would be a good bang for bucks solution to my problem? I'd also need a knob, because the ALPS would appear to be sold naked?

do you need to supply an ALPS pot w/ power at all? also, the Shure attenuator is apparently using resistors...it makes the trebles less shrill and gives a more balanced sound than digital attenuation, is that how an ALPS works? the Burson stepped attenuator is full of them: High Precision Volume Control (HPVC)

maybe it's actually the impedance change(due to the resistors in the Shure "toy") that's making the sound more balanced than directly from the DAC line-out? I'd appreciate any insight, thanks in advance

BTW, the Spitfire line-out is rated for 2Vrms/100Ω
Reply
post #2 of 30
3/9/10 at 7:25pm
eeeeh.

I would be VERY careful with using a simple pot on the output of something. it inevatably screws up the output impedance (if that bothers you, or even matters in your app)

If you are interested in this sort of thing, I would czek out dsavistik's webpages.

Somewhere in there he mentions the idea of using an autoformer directly off of the output of a CD player. This effectively allows for volume control as well as low output impedance.

A pair of nice autoformers (with switches) can be had for a hair less than an RK50. After that its all the same junk of a case, knob, and jacks.
Reply
post #3 of 30
3/9/10 at 7:37pm
Thread Starter 
oh......well, apparently there's no "hi-fi" volume attenuator being sold anyway? only crappy ones for portable use(w/ major stereo imbalance at low volume)?

the Shure attenuator gives a far less shrill sound than directly on the DAC line out + 0-3% digital attenuation, my cd1k seems to enjoy the impedance change due to the resistors within the Shure attenuator(which is most likely made of a variable resistor) I'd guess?

ah well, all the possible solutions seem to have drawbacks...I guess my best option will be to set the Burson amp stepped attenuator as low as possible and manage digital attenuation to 75/100% in my audio players.

I read some posts from Ti Kan that didn't make analog volume attenuation too appealing
Reply
post #4 of 30
3/9/10 at 7:42pm
there are MANY posts here that deal with attenuation (mostly in the analog domain; analog controlled or digitally controlled).

not sure there needs to be another thread. see if you can live with the various kinds of volume control (pots, relays, LDRs, solid state r/2r). you are right, regular pots are not the be-all end-all and there are many ways to skin this cat.
Reply
post #5 of 30
3/9/10 at 7:53pm
Thread Starter 
yes yes yes, I've read them all...they're full of ppl repeating every 2 posts "there are MANY posts here that deal with attenuation [..] not sure there needs to be another thread"

and they all end w/ "I wish there were some hi-fi volume attenuators".
Reply
post #6 of 30
3/9/10 at 7:58pm
it gets very very subjective the more money you throw at it and you trade some things for others when you pick a model (relays vs solid state vs pot and so on).

I've never seen any single answer that everyone agrees on. its almost a 3rd rail, in that respect

not sure what else to tell you. there are lots of ways to go and no clear winner. just trade-offs; so you have to pick what you want to live with.

I noticed you mentioned low level tracking as one must-have. pots suffer from that but none of the other technologies do. solid state tracks well, relays and R's track each other well and ldr's *can* be made (with effort) to stay in track.

you can even change your approach; use variable gain instead of attenuation. depending on the amp topology it may or may not be easy to change it to support variable gain.
Reply
post #7 of 30
3/9/10 at 10:07pm
While there are tons of posts on here talking about pots vs. stepped attenuators vs. digital attenuators etc, they all have a high output impedance basically meaning that you need an amplifier with them. The pot used in the inline volume control you linked to goes up to 100ohms. By contrast, the lowest pot you'll find discussed around here is 10kohms.

You could make your own inline volume control by finding a decent 100ohm pot and wiring it up with a couple jacks. This isn't quite ideal but it'll do what you want. Another approach would be to make a simple amp with a decent audio pot on the input, maybe a Cmoy.

A final, rough idea would be to make an impedance adapter that puts a fixed resistor in line with the headphone cable (you could probably find some premade ones). It would lower the volume by a fixed amount allowing you to use higher volume settings on your computer which make fix the shrill sound. Fixed resistors add very little, if any, coloration to the sound.
Reply
post #8 of 30
3/9/10 at 10:23pm
You could make a stepped attenuator with custom values on the low end to give better adjustment at low levels. There is a blank 23 step switch on ebay for $14.99:
23 Steps Ladder Type Potential Divider / Switch
There are some web sites that show how calculate the resistor values.
Reply
post #9 of 30
3/9/10 at 10:40pm
uhm... still huge impedance issues
Reply
post #10 of 30
3/9/10 at 10:44pm
passive vol controls kind of suck...

buffer them. that's what all serious implementations do.

once you are buffered, you can throw impedance to the wind (so to speak) since you are low enough to not care.
Reply
post #11 of 30
3/9/10 at 10:49pm
Use 2 small mono log pots, one for each channel.
Reply
post #12 of 30
3/9/10 at 10:53pm
Thread Starter 
ok, thanks for the suggestions!
Quote:
Originally Posted by linuxworks View Post
you can throw impedance to the wind [..] since you are low enough to not care.
hah, well the DAC lineout is given to be 2Vrms/100Ω, my cd1k is 32Ω/104dB...and god knows about the Shure attenuator

that's what a soundcard manufacturer has to say about the impedance mismatch when you plug headphones to a line-out(click on "technotes"): Auzentech, Inc. X-Fi Home Theater HD
Quote:
You can see from this graph that when connecting the Headphone amplifier, it shows a stable output in all range of frequency, but when connecting headphone to the line output, due to the headphone impedance curve, it will affect the output.


I'm not sure that I don't care about impedance...the Shure thingie even at max volume sounds far more balanced on my cd1k than the straight DAC output..
Reply
post #13 of 30
3/9/10 at 11:01pm
if you have a buffer, you've done what you can to 'not care' as much about the load.

I dont' believe in having amps 'participate' in playing with the load. they should just drive the line and the load and drive it well. many amps you can choose from (even portable amps are good enough to be buffers).

you can't change the fact that headphones are reactive. but that has nothing to do with using a buffer. if you use a buffer, you've done what you can and at least attempted a proper solution. if you go unbuffered, you take your chances and probably have a very suboptimal result.
Reply
post #14 of 30
3/9/10 at 11:13pm
Thread Starter 
oh ok...well I wanted to know more about that Shure "toy" but their server is down: Shure > PA235

I still wonder why w/ this thing at max volume, I get a much more balanced sound than w/o....got to be an impedance mismatch of some sort.

some amp manufacturers keep boasting about buffers that automatically match the input/output impedances...OTOH, many soundcards can drive headphones on their line-out..
Reply
post #15 of 30
3/9/10 at 11:21pm
Overall recommendation - dude, stop spending money on tweaking everything in sight, get something decent to begin with.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › making my own volume attenuator?