Coupling Cap polarity..?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Anyone? I'm building green pre right now, comparing green schematic with the original, they have different polarity connection in coupling cap, is it doesn't matter?
 
> what happened if it connected in reverse ?

Film caps, all the same.

Electrolytics: burst.
 
film caps don't have polarity.(poly-styren-ester-propylen etc)
if you reverse polarized cap on the output of an op amp you will pass any DC, that is present at the opamp's output, to the next stage (generally speaking). the rule is to put it the way that will not pass DC. if not sure try it. connect multimeter after the cap. if there is DC it is wrong way around. there is usually high value resistor (10-100k) after the cap to help discharge the coupling cap.
best

 
Thanks Syn...so what's the rule about the polarity of Electrolytics coupling cap? i saw most of circuit  put the negative at the output side, but this one ? ???
 

Attachments

  • out cap.JPG
    out cap.JPG
    40.7 KB · Views: 100
Simonsez
              you are welcome. The rule is that there is no DC after the cap ;D . That might be a mistake on the schematic. Try it the way i described above, you will not burn anything. Just don't connect any equipment to its output before you make  sure that you have lost all the DC. Multimeter's one probe after the cap the other probe to gnd. If there is DC turn the cap the other way around there should be no DC.
cheers
 
I don't subscribe to an outside foil polarity theory, but there is some truth in the article, although the author should have analysed the subject with a more solid scientific approach. In most decoupling and coupling applications, capacitors behave as virtual short-circuits AC wise; as a consequence there is no significant difference between one armature and the other; both orientations will radiate the same electrostatic field. This is not true in filter applications, where one of the armatures may be at ground potential and the other carrying a significant AC signal. If you take the example of a Sallen & Key LP filter, the cap that goes from +in to gnd should have its outer armature to ground; that would prevent this cap radiating and the risk of crosstalk or incorrect behaviour. The answer is not as easy for the other cap, the one that goes from the output to the resistor node, because there is signal on both sides, so some analysis is required, but in most cases, connecting the outer armature to the output will make sure that the effects of parasitic capacitance to ground are minimised.
In the case of a plate-to-grid coupling cap, there should be no difference, unless the Fc is voluntarily high; an example is the Vox AC30's Brilliant channel that has a 500pF cap between the 1st plate and the 500k volume pot, which makes it a 600Hz hipass filter. If the outer foil is connected to the pot, any electrostatic interference below 600Hz will be transmitted to the pot unattenuated. It is clear that connecting the outer foil to the plate will protect against interference because the plate impedance is lower (50-100k) and the outer foil will act as a shield to the inner foil.
Some decoupling caps may require a little more attention and more analysis of the signals and the environment.
One of the recurrent problems I see is that some guys want to give quick and dirty answers to any problem, but in fact one cannot get away with doing his homework; these ready-made answers may be right in many cases, but it is so much more reassuring to actually assess and quantify the problem.
 
schmidlin said:
I suppose you guys don't subscribe to the outside foil polarity theory?
http://www.aikenamps.com/OutsideFoil.htm
Seems the polarity on the shemo is on target if so.
The first sentence from the OutsideFoil.htm starts with "Some non-electrolytic" which rules out any further referencing to it as the caps in question are electrolytic, and Simonsez's question was/is about particular situation shown on the schematic above and about electrolytic cap orientation. OutsideFoil.htm is very good info but another story IMHO.
As far as the homework goes, yes, there is plenty left in my case. However, "quick and dirty" does the job in this case,  IMHO. Simonsez wants his pre up and running, and I tried to give him just that, what he asked for.
Best
Milos
 
Syn, you gave an appropriate reply.  I suppose I lumped you in with PRR's reply, risking hijacking the thread.  I assumed audio folks would be more sensitive to this outside foil talk (keeps things quieter, supposedly), and wonder if I am wasting my time determining blocking cap foil sides in amplifiers.  If you all don't care, I certainly won't.

Best back at ya,
Schmidlin
 
Ok, that schematic above is from 9098i circuit diagram ( downloaded from groupdiy gmail), and after reviewing again i don't thing it's a drawing error/mistake. there is a lot electrolytic cap coupling/blocking there with different polarity connection. sometimes + at the output side, sometimes negative. I think there is another explanation to this (or it doesn't matter?).  Right now i connect the + at the output side and no dc at the output. Problem solved but still leave a question in my head...
 
@simonsez what is the DC offset before the output capacitor?  someone like PRR or abbey road d enfer
would be much appreciated to give us some more insight into this. that mic pre is the part of the console and i'm not sure that is meant to go to the outside world like that.on the other hand green is meant to go to the outside world and it has caps polarity reversed...
http://www.groupdiy.com/index.php?topic=36210.0 will give some answers...

@schmidlin about outer foil- i'm not sure, but i have not noticed or measured any difference. this statement is limited to low V circuits like passive LC filters where i used polystyrenes and checked for the orientation. i'll re check. probably with higher currents and/or voltages things behave different. i don't have much knowledge/experience with tube stuff at all, and i have no interest in power or guitar amplifiers at this moment. i'm building a couple of tube comps right now so i'll give it an experiment with this...

cheers

m
 
syn said:
the rule is to put it the way that will not pass DC.

Strange, where I live they (i.e. electrolytic caps) don't pass DC either way round. If they would they would only pass one half-wave if your music.
Only problem is that they want to have a DC potential the right way around them. If you get it reversed (and above about 1 - 1.5V) electrochemical reduction starts to set in and damage your capacitor. If the voltage is the wrong way round and even higher they will go KABOOM and you can have a look inside.

The reason that most schematics have in the outputs the + side towards the op-amp is that people think "that op-amp has an offset voltage" and voltage most mean positive. Depending on the op-amp it can also be negative. That's one of the reasons some circuits (like the 9098i mentioned above) add DC offset on purpose. Only then you can know which way round to put in your caps. If course there are other areas in a circuit where you have sufficient DC offset and where it's thus easy to figure out which way round to put in the electrolytic capacitors.
Concerning an output stage I'd advice to build your stage so that you can have the - side towards the op-amps (or whatever there is in your circuit) just in case somebody applies +48V phantom power to your outputs.

Olaf
 
syn said:
if you reverse polarized cap on the output of an op amp you will pass any DC, that is present at the opamp's output, to the next stage (generally speaking).

No.

Like olafmatt says, that's not how polarized capacitors work.

JDB.
[and don't start about varicaps; that's another case altogether]
 
olafmatt, jdbakker thank you very much indeed, now i understand much better how they work.
still i am confused by the fact that i have DC after the coupling cap roughly equal to its offset value
if i reverse cap polarity. it does not open up. it does pass audio. it does pass DC. i can measure it.
BigBloak (CFB, ~-1VDCoffset, no output transformer connected) into 4700uFNichiconPW/25V, reverse polarity of the output cap and it will just do the above.
that is how i got the (wrong) impression that if you reverse the polarity it will pass DC.
@olafmatt: what is the reason that you reversed polarity of the output caps on you supergreen compared to the original schematic?
just trying to put this on a right place in my head for good.
cheers
 
There may be two reasons why you measure DC at the output:
a) the cap is leaky
b) the discharge is not complete, and I tend to think this is the case. 4700uF with 100k gives a time-constant of 470seconds. It means it will take about 8 minutes to reduce the voltage down to 30%.
I suggest you power up the beast, discharge the cap with a low value resistor (100R) from output to ground for a few seconds and re-do your measurement. Please note that the voltage may go up momentarily because of DA (dielectric absorption).
 
abbey road d enfer thank you very much, i'll do just that later today and report back.
best
m
 
abbey road d enfer said:
b) the discharge is not complete, and I tend to think this is the case. 4700uF with 100k gives a time-constant of 470seconds. It means it will take about 8 minutes to reduce the voltage down to 30%.
I suggest you power up the beast, discharge the cap with a low value resistor (100R) from output to ground for a few seconds and re-do your measurement. Please note that the voltage may go up momentarily because of DA (dielectric absorption).
abbey road d enfer, again, thank you very much indeed. that is it .damn thing takes forever to fully discharge. lesson learned.
thank you everyone
 
Back
Top