what's a servo out for?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

kiira

Well-known member
Joined
Aug 2, 2004
Messages
536
Location
Baltimore - Blobsville USA
My api312 board from Fabio has a circuit and output for "servo out"... what does that do? I've also seen designs like the Jensen twin-servo... are they related? I'm ordering parts for stuff (yech!) so I need to decide if the servo thingy is cool.

thanks

Kiira
 
[quote author="Family Hoof"]Servo removes DC offset actively, as opposed to just passively blocking it with a cap. Yes, the Jensen design is named for this topology which it makes use of.[/quote]

I see, thanks. Is there some practical advantage to using it? Perhaps a link explaining that anyone knows about?

Kiira
 
Run a search or two on the forum.

There have been a number of discussions. It's a little contentious to some people, a problem is that some folks seem to over-simplify things and assume that all servos are either God's gift to audio, or the work of Satan.

The concept is graspable, but the subtle details are not so apparent. It's been discussed a few times here.

Keith
 
One of the things you often hear is that servos eliminate capacitors in the signal path. In a way this is true, but it leads to the impression that there are consequently no more effects of capacitor imperfections as a result. This is false---the servo itself has one or more capacitors, and when configured in the circuit those caps have similar effects. What the servo achieves is an overall highpass response. It may be more convenient and possibly more flexible, particularly when the alternative is a monster electrolytic with its large stray capacitance and inductance and board-hogging profile, but there are a host of other issues that come with the servo.

As the old saying goes, there's no capacitor like no capacitor. But servos do not eliminate capacitors.
 
> the servo itself has one or more capacitors, and when configured in the circuit those caps have similar effects

Yes.... but the cap can be a lower value (thus a better type at same/less cost/size) and typically its signal and flaws are resistively divided before going back into the signal path.

OTOH, you have added an amplifier to leverage-up the cap's effect, an amp that may have flaws even in the limited work it needs to do.
 
[quote author="PRR"]> the servo itself has one or more capacitors, and when configured in the circuit those caps have similar effects

Yes.... but the cap can be a lower value (thus a better type at same/less cost/size) and typically its signal and flaws are resistively divided before going back into the signal path.

OTOH, you have added an amplifier to leverage-up the cap's effect, an amp that may have flaws even in the limited work it needs to do.[/quote]

Yes, like noise.

But if you don't need much servo action the noise can be managed. There are also nonlinear system approaches---I don't know if anyone has implemented them in pro equipment as yet.
 
.
Kiira, with those boards, you can use any of the outputs, but don't need all of them. If you have out trafos like Cinemag CMOQ-2B you don't need to fit the servo parts (labelled with an 'S', like RS1 etc.) or the cap output (U, like CU1 etc.). If you want both cap and trafo, may need to put a switch or switched relay in there. Not sure if you can use both trafo and servo outs, or if a switch would be needed there too. 20uf Solens from Parts Xpress were nice for CU1 and CU2 (one cap to replace both), but I prefer the trafo output, so removed caps. BTW 'U' section is unbalanced out.

Sorry if you knew,
Paul
 
[quote author="Svart"]"There are also nonlinear system approaches"

Enlighten us? :thumb:[/quote]

Some aspects I may file on, if I ever get my ass in gear, so I shouldn't talk about. But consider having a hybrid approach---a ~zero-droop track-hold comes into play to take out big offsets, and a more conventional servo takes care of the residual.

The zero-droop track-hold could be made with an A/D + latch + D/A.
 
[quote author="adrianh"]"there's no capacitor like no capacitor"

Wow sounds like a transformer to me.
And yes my transformers block DC
so I do not record DC to the DAW.[/quote]

There are people who hate caps enough to use inductors everywhere possible---Tim de Paravicini I believe did a phono preamp in which the RIAA EQ was done with L's.
 
All this effort to get rid of the last little bit of DC..... what's wrong with a bit of offset? :razz: Improve low level distortion! A little bit of current never hurt anyone! :wink:
 
Is there a calculation for coupling with inductors or is there a value that
works well in general ?

I have a pinout drawing of a QE ca137 DOA drawn as a sum amp that came from Indigo Ranch and the input shows an inductor but no value ?


Lance
 
You're talking about choke-coupling? The inductive reactance (Xl) should be greater--preferably much greater--than the source impedance at the lowest frequency to be amplified. The formula for inductive reactance is
Xl = 2pi F L (where F is frequency in cycles per second, L is inductance in Henries and Xl is inductive reactance in Ohms). Ignoring shunting effects of other circuit elements, the signal will be -3dB at the frequency where Xl equals the source impedance.

Choke-coupling is expensive and often impractical, which is why it was never widely used. Usually, it's cheaper to use a higher-voltage power supply and couple with resistance and capacitance (R-C coupling). However, transformer coupling has always been popular where impedance transformation is required, since it can perform that function as well as block DC.
 
Thanks Dave!

I now remember reading about this in my naval basic electronics book.

My mind starts to race with questions....

... and I guess inductors and chokes are like transformers where the quality varies considerably with construction and materials.

Still, isn't it easier to do DIY inductors of good quality than DIY transformers?

Lance
 
Back
Top