[Occam]

......
On second thought, forget I asked that. I think I'm just going to bow out of this discussion and the cable discussion.

se

HUZZAH!!!!
The drama queen exits, stage left, to sit alone in the dark 

[Scott F.]

Oh....and I was soooo looking forward to your Karmic retribution 



signed,
wanker²

[JoshK]

Maybe we differ over our definitions of linear but I see it as meaning a straight line. If the impedance plot varies from a straight line I feel justified in describing it as non-linear.

This is quite often confused, but when engineers, mathematicians and scientistist speak of linearity they do not mean a line, they mean a linear system.  A linear system has more to do with how elements are introduced into the system.  Nonlinear has to do with certain interactions that make the addition of a new element not as simple as its addition.  As far as I understand it, EE's speak of linearity as to whether the "element" can be modeled as a simple series or parrallel introduction of simple capacitive, inductive and resistance components rather than the reactive components interacting with the reactive components already in the system in a non additive way. 

[art]

ESL, which is largely determined by lead spacing, can be kept to a minimum by avoiding axial leaded capacitors and using radial capacitors with the smallest lead spacing.

se

NO!!!!!!!!!!!!!!

You are assuming that the cap has zero winding inductance. And you use an electrolytic cap as an example.

Good grief.

[Steve Eddy]

Kind of hard to bow out of a thread when people continue to reply to your posts (and initiate personal attacks against you to boot).

NO!!!!!!!!!!!!!!

You are assuming that the cap has zero winding inductance.

No. It simply assumes that inductance due to the windings is minor compared to inductance due to lead spacing.

And you use an electrolytic cap as an example.

Good grief.

Good grief? Why's that? Do you have any idea how incredibly thin the dielectric is in an aluminum electrolytic capacitor? It's on the order of 15 angstroms per volt. That's 5.8 x 10^-8 inch. So for say a 60 volt capacitor, you're looking at a spacing between foils of only about 0.0000035". With such a microscopically small spacing between the foils, they're simply not going to produce much in the way of inductance.

I didn't just pull this notion of lead spacing dominating the capacitor's ESL out of my ass. It came from Characteristics of Aluminum Electrolytic Capacitors by United Chemi-Con.

To wit:

The inductance of a capacitor is a constant and is due primarily to the capacitor terminal spacing. The inductance runs anywhere from 4nH for miniature radial capacitors to as high as 34nH for large can capacitors.

So if you have an issue with what I said about ESL and lead spacing, take it up with the folks at United ChemiCon.

se

[art]

Wish I had an old Mepco catalogue to 'splain it to you.

Then you could take it up with them. And have them 'splain to you the techniques for lowering ESL. If it was as easy as you seem to believe.................

[bwaslo]

As far as I understand it, EE's speak of linearity as to whether the "element" can be modeled as a simple series or parrallel introduction of simple capacitive...

Sort of.  But the way it is usually talked about is in terms of spectral effects.  That is, if you put N frequency tones into a Linear system, at the output you'll see no other frequencies than the original ones you put in.  But a Non-linear system will give you new output frequency tones that weren't in the input, i.e., distortion products.

Using the term "linear" to mean "response like a straight line" over some variable can be a bit hard to keep consistent for reactive components.  For instance, the impedance of a perfect capacitor won't graph as a straight line if plotted as Impedance vs. Frequency, it will have a warp to it.  But if plotted on a log-frequency scale (Impedance vs. Log Frequency, which is probably more appropriate seeing as ears tend to hear frequency changes in octaves rather than in Hz), then the plot would  be a straight line, sloping downward, after all. 

If the impedance were to be  straight and also be flat with frequency (or log-frequency) -- the same impedance at all frequencies -- then you don't have a capacitor at all, you have a resistor, which is pretty easy to come by with a constant number of ohms or milliohms over a wide bandwidth.  Putting that across a power supply output has its problems, though, since it will have the same number of milliohms at DC as at audio frequencies, and draw a lot of current, probably generating smoke.  The whole point of using a capacitor bypass is it has near infinite impedance at DC and near-DC frequencies, but low impedance at higher frequencies of interest.  A capacitor can't be flat and level over all frequencies, that wouldn't be a capacitor at all.  All of which I think was understood by Steve Eddy and Paul Hynes, I just thought it needed clarification in general.

HUZZAH!!!!
The drama queen exits

that comment seemed uncalled-for.  And rather drama-queen like, itself.

[Steve Eddy]

Wish I had an old Mepco catalogue to 'splain it to you.

Why do you need an old catalogue to 'splain it to me?

se

[Steve Eddy]

All of which I think was understood by Steve Eddy and Paul Hynes, I just thought it needed clarification in general.

Thanks, Bill.

Yeah, as I said I'm sure Paul and I both understood what each other meant but didn't quite see eye to eye on terminology.

se

[john curl]

Paul is in track.  Electrolytic caps, even the best quality, can be detected in many audio circuits, if they are located too close to the active audio electronics.  It takes a huge cap to properly bypass a battery in any case.  Dielectric absorption can be as high as 10% in electrolytic caps.  A fave of mine is to take a NEW large valued 200V electrolytic cap that has been stored on a shelf for awhile.  Then short the terminals with a screwdriver.   Take care, sparks can fly!

[DaveC113]

Paul is in track.  Electrolytic caps, even the best quality, can be detected in many audio circuits, if they are located too close to the active audio electronics.  It takes a huge cap to properly bypass a battery in any case.  Dielectric absorption can be as high as 10% in electrolytic caps.  A fave of mine is to take a NEW large valued 200V electrolytic cap that has been stored on a shelf for awhile.  Then short the terminals with a screwdriver.   Take care, sparks can fly!

What would you do to match caps to a battery power supply? The battery is 12V powering an amp that draws about 30W max.

If the topic is too big maybe point me in the right direction so I can read about it somewhere.

Thanks,
Dave

[john curl]

It is almost impossible, except at high frequencies. 

[Imperial]

How about this: http://www.pyramidcaraudio.com/itempage.asp?model=RDCAP12
Got a really low esr 0.0016 Ohm, and most likely will not be fazed by a draw of 30w..
Use Tantalum wire to connect battery to capacitor, to keep ESR low. Tantalum is very low in resistance. Use thicker wire rather than thinner = less resistance...
Hook up yer amp... Should work nicely this...

Imperial

[art]

Why do you need an old catalogue to 'splain it to me?

se

Why.........??????.........cuz it has such nice purty piktures, even you would not be able to argue with. Besides, you seem to have a penchant for quoting stuff from spec sheets. Good enough for you..........good enough for me.

BTW......how do 'splain that the ESL of damn near every radial leaded stacked-foil cap is about the same......and it has nothing to do with lead spacing.

Say "Good night, Gracie......."