Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!

Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  
previous: Jim Wheeler Re: Polarity of non-polarized caps -- 2/28/2000 10:31 AM View Thread

Re: Polarity of non-polarized caps - long

2/28/2000 10:41 PM
Re: Polarity of non-polarized caps - long
Okay, here's the (long) reply from an engineer at SBElectronics, makers of Sprague Orange Drop caps (written by them, not me). I was specifically asking about the 715 and 716 caps. Some of this was discussed a few months ago on the WeberVST board. -- Dan  
You raise some interesting points, and the answers I give you may be somewhat controversial in the world of "audiophiles". I will back up my opinions with technical explanations; feel free to ask more detail if you need to.  
Some background.....  
First of all, our process makes no differentiation as to which lead is connected to the outside foil. If you have a large enough sample of our parts you will find the outside foil is connected to a given lead for half the units. Should you decide it really is important, the best way to determine which lead is connected to the outside foil is to build a test oscillator that uses the capacitor in question as a resonating or frequency determining element. One side of the capacitor must be at the circuit "common" or ground potential. Adjacent to the capacitor one would install a metal plate connected to a very high impedance [>10megohm] amplifier. If the outside foil is connected to the circuit common, there will be little if any signal capacitively coupled to the test plate. If the inside foil is grounded, the voltage changes on the outside foil will be coupled to the high impedance amplifier.  
Second, just because you may have capacitors with an "outside foil" marking is no guarantee that the marked side is really the "outside foil"!!  
Third, some capacitors are constructed with 2 or more capacitors internally connected in series to gain higher voltage ratings. If this is the case the concept of "outside foil" breaks down completely. [Standard value 715/716P capacitors do not use series construction until Rated DC voltages are higher than 600V.]  
Rationale for use of "outside foil" CONCEPT:  
To me the logic is clear, use the outside foil as a shield. The CONCEPT works even if one terminal of the capacitor is not grounded. If I was to pay attention to which lead was the outside foil, my rule would be: "Connect the outside foil terminal of the capacitor to the lowest impedance side of the circuit".  
For coupling applications this concurs with your quoted admonition to "orient the capacitor so that the outer foil connects to the "incoming signal" and the inner foil to the "output" side of the circuit". Signal sources are almost always lower impedance than signal loads.  
For tone control circuits it is not as simple to define the signal in and signal out terminals of the capacitor. Tone control circuits operate with significant AC voltages across them. For the case where one end of the capacitor is grounded [simple treble cut circuit] the rule you quote breaks down. The signal source should not be connected to the outside foil in this case, the outside foil should be connected to ground.  
I do not believe use of the outside foil concept is meaningful for audio applications. There are so many more important issues!!!! ***************************************************************  
In coupling applications:  
For coupling to occur, the AC voltage across the capacitor is near zero. that means that the AC impedance to ground at frequencies of interest is nearly the same at both ends of the capacitor. The thing most folks forget when talking about the "value" of correct outside foil connection is the impedance of the capacitor itself. Lets say the plate load resistor of the source stage is 100K ohms, and the grid resistance of the load stage is 1meg. For grins pick a coupling capacitance value of 47nf. At 60Hz, the capacitor impedance is 56K, a small fraction of the 1M grid resistance. Doing a simplistic analysis [ignoring tube plate resistance] the total impedance looking back from the grid to the plate load is: the square root of (100K squared + 56K squared)or about 115K.  
This means that if one got the coupling capacitor too close to the heater wiring the hum pickup would be nearly the same, regardless of which lead was the outside foil. The fix is to dress all high impedance signal carrying wire and components as far as possible from undesired sources. Keep all high impedance signal runs as short as possible, or run them in shielded cable. Better yet, remove undesired sources from the environment.... such as running the tube heaters with regulated DC. From experience, that works very well. I have an old tube type HP signal generator that had interference problems at lower harmonics of 60Hz. DC on all heaters completely eliminated the problem.  
For tone control applications I cannot be as specific without the circuit diagram at hand, but the analysis is the same. The wire dress is MUCH more important than outside foil orientation.  
The reason the capacitor outside foil orientation has remained so alive in audiophile culture is its simplicity.  
A comparative situation is the union concept of seniority. While it is true that experience in a job will make a motivated employee more valuable to a business, there is no guarantee that the motivation component is present. Yet seniority is so easy to measure it remains a primary parameter in a union shop determining who gets laid off first or gets an advancement in pay or grade.....  
The best news about consideration of capacitor "outside foil" is that implementation of its concepts correctly [or even incorrectly...] can do no harm!  
I hope this discussion has helped.  
If you have more specific questions just ask.  
Good luck with your project(s)!!