Question about analog controls

[nightflier]

(This may have been covered in another thread, if so, please include a link)

If manufacturers have such a hard time creating silent analog controls for volume, balance, etc., wouldn't a digital control be more quiet, precise, durable and efficient? I'm thinking of the Conrad Johnson SC-26 for example. Now it isn't one of their highest rated models, but I would think it would be pretty good.

And if it's just an interface issue where owners "prefer" analog dials, couldn't manufacturers just create an analog dial on the front of the component but with a digital interface inside the unit?

I'm not an electrical engineer, so I don't know all the technical details about this, but I was just wondering.

[Feanor]

(This may have been covered in another thread, if so, please include a link)
...
And if it's just an interface issue where owners "prefer" analog dials, couldn't manufacturers just create an analog dial on the front of the component but with a digital interface inside the unit?

I'm not an electrical engineer, so I don't know all the technical details about this, but I was just wondering.

For example the volume knob on my Sonic Frontiers LINE 1 which is, in fact, a digital control with with an analog dial just like you suggest ...



My former Adcom GFP-750 had a selector knob that was also actually digital
...

[nightflier]

Exactly.

So why don't more manufacturers do this?

[Feanor]

Exactly.

So why don't more manufacturers do this?

But are you off an age to remember the slider fad of the mid-80's? Horizontal or vertical sliders replaced knobs on a lot of stuff for volume, balance, tone controls, etc. That was brutal. Worse than buttons, IMO. Mostly it was low to mid-range consumer stuff, fortunately.

[basite]

primare also uses the digital volume control & input selectors, and probably more brands will do this, but sometimes you just don't notice that they use it.

most HT receivers also use a digital volume control.

[Mr Peabody]

Doesn't digital get criticized for having it's own noise? Krell and many of the CJ preamps use a step attenuator where each step has a different resistor value. I believe this has become one of the most widely accepted high end volume control designs.

[nightflier]

But are you off an age to remember the slider fad of the mid-80's? Horizontal or vertical sliders replaced knobs on a lot of stuff for volume, balance, tone controls, etc. That was brutal. Worse than buttons, IMO. Mostly it was low to mid-range consumer stuff, fortunately.

I think that was something that crossed over from the pro audio world, but I can't imagine this improved on the rotary controls in any way. Come to think of it, the Phillips receiver my dad had there in those years, with horizontal controls, did get really dusty over time. I would guess that a rotary dial would stay cleaner simply because of gravity & surface area factors alone. Vertical dials (like on equalizers) might be the better option, but I still think rotary dials would be better.

That's why I'm thinking that digital controls, especially those w/o dials as front-ends would be the best choice in higher end components. Actually, are there any components that use the "feather-touch" type controls? These would be the best, right?

Mr. Peabody, where did you read that digital controls add noise? Is this common knowledge? Did I miss that article somewhere?

[Mr Peabody]

No, that was more of a sincere question. Digital is usually pretty well isolated and there was that business where it became big to be able to turn off the display on CD players. Even on the power conditioners they isolate digital and analog outlets from each other. I'll have to do some checking.

[Feanor]

No, that was more of a sincere question. Digital is usually pretty well isolated and there was that business where it became big to be able to turn off the display on CD players. Even on the power conditioners they isolate digital and analog outlets from each other. I'll have to do some checking.

Is there a noise difference between "digital" circuits that generate EMI or RFI -- such as coding, decoding, and amp switching signals -- versus simple logic gates which I presume is the nature of source/output selectors and volume controls?

[FLZapped]

(This may have been covered in another thread, if so, please include a link)

If manufacturers have such a hard time creating silent analog controls for volume, balance, etc., wouldn't a digital control be more quiet, precise, durable and efficient? I'm thinking of the Conrad Johnson SC-26 for example. Now it isn't one of their highest rated models, but I would think it would be pretty good.

And if it's just an interface issue where owners "prefer" analog dials, couldn't manufacturers just create an analog dial on the front of the component but with a digital interface inside the unit?

I'm not an electrical engineer, so I don't know all the technical details about this, but I was just wondering.

It's not all that difficult to do and may be more wide-spread than you realize. (Very big in the recodring/live console market)

Perhaps the biggest knock against digital controls is the zipper noise and response lag they have if the corresponding analog control is turned too fast.

Once in a while, they may also forget where they are set to if the programmers haven't made sure to check the control setting every so often.

-Bruce

[nightflier]

zipper noise and response lag they have if the corresponding analog control is turned too fast.

So there is noise involved? I'm also surprised that there is lag compared to the analog control. I thought it would be just as fast, but I can see where this could be a problem.

Very interesting.

[Mr Peabody]

http://www.elecdesign.com/Articles/A...2531/2531.html

http://www.elecdesign.com/Articles/A...376/12376.html

http://www.edn.com/article/CA207102.html

Here's a couple links to some info on digital volume controls. I found some others but they were either advertising their design or really technical. The above mention the lag, zipper and show an example of using the resistors.

What's interesting to me after looking at numerous "digital" designs, are some of them I wouldn't have thought as being digital, more a type of hybrid. I guess it depends on what the main circuit type is.

[nightflier]

So according to one of these articles "...designers can reduce zipper noise by converting the random change in volume versus time into a defined pattern where volume changes occur only at zero crossings. Consequently, no abrupt volume changes will transpire. This technique works by inserting a zero-crossing window detector in the Chip Select (/CS) line of the digital potentiometer, delaying the device update until the audio signal reaches the zero-crossing window."

So is this expensive?

[hermanv]

As far as I know a big problem with digital "integrated" volume controls is that they use silicon resistors, perhaps not the best sounding choice or silicon diodes operated at a "knee" voltage which then requires fairly small (tiny) AC signals

There are two other techniques, one uses a four quadrant multiplier (The Levinson No.28 does this a DC value times an AC value), but multipliers tend to be noisy and linearity is so-so. The last technique I know of a is a light dependent resistor, the digital signal sets a DC current through a light source (usually an LED) and then uses a selenium light dependent resistor as the gain element.

None of these ideas hold up to Vishay discrete resistors and silver or gold contacts. They do keep getting better and my knowledge could be out of date.

Motor driven conventional pots could appear as a digital volume control to an end user. I've heard some decent pots, but to my ears they couldn't compete with the Vishay and dry contacts

[nightflier]

The last technique I know of a is a light dependent resistor, the digital signal sets a DC current through a light source (usually an LED) and then uses a selenium light dependent resistor as the gain element.

I did not intend for this to be the case, but this is starting to be quite a complicated subject, the more I read about it. I do think that the best solution is what hermanv suggested above. If I understand this correctly, the user-knob is analog, but instead of rubbing along against a receiving plate (excuse the layman technology), it sends a light signal to a receptor that then measures the increase or decrease. This way there is no friction involved at all. Am I understanding this correctly? If so, I could see this being quite expensive to develop and mass-produce in a manufactured component.

[hermanv]

There's never a free lunch.

There are controls known as incremental and absolute shaft encoders (Digikey has them reasonably priced). These devices put out a digital code corresponding to shaft position and feel just like a pot. This code can then be wired to an R/R2 ladder network with relays or digital analog switches or to a four quadrant multiplier to set the gain.

These same device can be used to set current in an LED for the photo-coupler idea.
See for example http://www1.silonex.com/audiohm/leve...ol.html#config

If you read the application notes carefully, you will see that distortion products while not bad aren't up to the best audiophile stuff.

There is at least one commercial passive pre-amp that uses optrols, I thinks it's called the lightspeed. Here's a link to another audio forum (is this allowed?) http://www.diyaudio.com/forums/showt...80194-p-3.html where this is discussed at some length.