life lesson on ceramic capacitors in signal path

mazurek · 21st February 2006, 05:39 PM

My active crossovers I built sounded sort of thin to me, really good on some things, not so much on others. I removed the 10 pf ceramic bandwidth limiting capacitors in the feedback path, and problem is gone. I had thought that their frequency of operation was so far outside of audio band that I was safe from their supposed problems. Luckily no oscillation, as I don't have room to install mica capacitors.

In retrospect, if I really need bandwidth limiting caps, I maybe should have chosen smaller resistor values to obtain larger capacitor values (and have more choice in quality capacitors).

By the way, In my opinion, I think modern rock/metal is the best way to find problems with speakers, my overpriced retail speakers and bad crossover all sounded great with acoustic material.

jcx · 22nd February 2006, 02:58 AM

not all ceramic is bad, np0 or c0g are generally better than mica and approach film quality on measurable distortion and dielectric adsorption, with higher feedback R many op amps require some feedback C to cancel destabilizing parasitic C at their inputs, often only a few pF is required - don't even think about using hi K ceramics like Z5U in signal apps

generally harmonically "simple" music with single voice, few instruments gives best sound with distorting systems - challenging music is well recorded (without the compression common in Rock and other pop formats) and harmonically complex - classical orchestral music with real dynamic range is the hardest to get right

Rock/Metal usually relies on distortion in tube instrument amps as a part of the musical "voice" and may have further compression used to keep average levels high - its very hard to judge when such pre-distorted source is getting a little more added by your system - but the high average levels will certainly test your excursion limits and amp clipping levels

klitgt · 22nd February 2006, 07:29 AM

Quote:

Originally posted by mazurek
My active crossovers I built sounded sort of thin to me, really good on some things, not so much on others. I removed the 10 pf ceramic bandwidth limiting capacitors in the feedback path, and problem is gone. I had thought that their frequency of operation was so far outside of audio band that I was safe from their supposed problems. Luckily no oscillation, as I don't have room to install mica capacitors.

Can you attach a diagram to visualize what we are talking about. It is an interesting topic that many can benefit from if they understand what it is all about. I think I have an idea but not quite sure

Cheers

AndrewT · 22nd February 2006, 09:06 AM

Hi,
will polystyrene fit the space? even standing end on to the PCB.

mazurek · 22nd February 2006, 01:34 PM

Can you get polystyrene in small values, and where?

Interesting comments jcx, appreciate the feedback. Too busy for a real reply today.

I attached a copy of the pcb for those interested, I will write up this project more thoroughly later, the documentation is not great.

Nordic · 22nd February 2006, 04:45 PM

RS components sells FSC Series Polystyrene Capacitors from 10pf upwards to .01uf, Capacitance tolerance ±2·5% or 1pF, 40V though, pretty expensive and 10 minimum per order

Bobken · 22nd February 2006, 05:49 PM

Hi,

If you have a choice here, go for extended foil polystyrenes as their end caps are attached to a greater area of the foil, and (generally) will sound better, too.

FSC-EX versions are extended foils, but you may not get them in the lowish values you are after. The FSCs mentioned above aren't very far behind them, though, in the sound-stakes.

Sonically, these extended-foils are one of the finest caps for small values, with only (non-magnetic leaded, which are rather rare) silver micas being close. So long as their slightly higher inductance (due to their rolled-up construction) isn't significant, and for most audio situations it won't be, I would choose these polystyrenes over all other caps, for relatively low pF values like compensation caps etc.

It was the accidental discovery of the obvious differences in sound between polystyrenes and ceramics (in a MC head amp) some 30+ years ago, which started me out on 'the road to ruin' of listening to all passive components in my entire system, and choosing accordingly!

Regards,

AndrewT · 22nd February 2006, 07:11 PM

Hi,
Rapid sell from 10pF upwards in qty of 1. But post from UK will kill you. I'm sure someone local must offer a similar service.

TwoSpoons · 23rd February 2006, 12:55 AM

I can't believe you would delete some caps, not replace them, then blame the sound change on the fact they were ceramic. The sound change is because the caps are missing !

davidsrsb · 23rd February 2006, 02:00 AM

A 10pF ceramic will have a very good dielectric and is very close to an ideal capacitor.

I suspect that you have a stability problem somewhere.

21st February 2006, 05:39 PM	#1
mazurek diyAudio Member Join Date: Aug 2005	life lesson on ceramic capacitors in signal path My active crossovers I built sounded sort of thin to me, really good on some things, not so much on others. I removed the 10 pf ceramic bandwidth limiting capacitors in the feedback path, and problem is gone. I had thought that their frequency of operation was so far outside of audio band that I was safe from their supposed problems. Luckily no oscillation, as I don't have room to install mica capacitors. In retrospect, if I really need bandwidth limiting caps, I maybe should have chosen smaller resistor values to obtain larger capacitor values (and have more choice in quality capacitors). By the way, In my opinion, I think modern rock/metal is the best way to find problems with speakers, my overpriced retail speakers and bad crossover all sounded great with acoustic material.

22nd February 2006, 05:49 PM	#7
Bobken diyAudio Member Join Date: Dec 2002 Location: UK	Hi, If you have a choice here, go for extended foil polystyrenes as their end caps are attached to a greater area of the foil, and (generally) will sound better, too. FSC-EX versions are extended foils, but you may not get them in the lowish values you are after. The FSCs mentioned above aren't very far behind them, though, in the sound-stakes. Sonically, these extended-foils are one of the finest caps for small values, with only (non-magnetic leaded, which are rather rare) silver micas being close. So long as their slightly higher inductance (due to their rolled-up construction) isn't significant, and for most audio situations it won't be, I would choose these polystyrenes over all other caps, for relatively low pF values like compensation caps etc. It was the accidental discovery of the obvious differences in sound between polystyrenes and ceramics (in a MC head amp) some 30+ years ago, which started me out on 'the road to ruin' of listening to all passive components in my entire system, and choosing accordingly! Regards, __________________ Bob

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
what is the difference between silver mica capacitors and ceramic capacitors ??	prorms	Solid State	1	6th May 2008 12:38 PM
Good or bad to use parallel capacitors of same value in signal path...	2004ex	Tubes / Valves	1	25th May 2005 11:38 AM
Tantalum Capacitors in Signal Path	schnullimaus	Parts	9	30th April 2005 11:07 AM
Aleph P 1.7: 2k pot is in the signal path?	swede	Pass Labs	14	29th December 2002 06:29 PM
15uf signal path capacitors for BOSOZ	ralf	Pass Labs	5	24th April 2002 07:51 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

22nd February 2006, 02:58 AM	#2
jcx diyAudio Member Join Date: Feb 2003 Location: ..	not all ceramic is bad, np0 or c0g are generally better than mica and approach film quality on measurable distortion and dielectric adsorption, with higher feedback R many op amps require some feedback C to cancel destabilizing parasitic C at their inputs, often only a few pF is required - don't even think about using hi K ceramics like Z5U in signal apps generally harmonically "simple" music with single voice, few instruments gives best sound with distorting systems - challenging music is well recorded (without the compression common in Rock and other pop formats) and harmonically complex - classical orchestral music with real dynamic range is the hardest to get right Rock/Metal usually relies on distortion in tube instrument amps as a part of the musical "voice" and may have further compression used to keep average levels high - its very hard to judge when such pre-distorted source is getting a little more added by your system - but the high average levels will certainly test your excursion limits and amp clipping levels

22nd February 2006, 09:06 AM	#4
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Hi, will polystyrene fit the space? even standing end on to the PCB.

22nd February 2006, 04:45 PM	#6
Nordic diyAudio Member Join Date: Sep 2005	RS components sells FSC Series Polystyrene Capacitors from 10pf upwards to .01uf, Capacitance tolerance ±2·5% or 1pF, 40V though, pretty expensive and 10 minimum per order

22nd February 2006, 07:11 PM	#8
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Hi, Rapid sell from 10pF upwards in qty of 1. But post from UK will kill you. I'm sure someone local must offer a similar service.

23rd February 2006, 12:55 AM	#9
TwoSpoons diyAudio Member Join Date: Oct 2004	I can't believe you would delete some caps, not replace them, then blame the sound change on the fact they were ceramic. The sound change is because the caps are missing !

23rd February 2006, 02:00 AM	#10
davidsrsb diyAudio Member Join Date: Dec 2005 Location: Kuala Lumpur	A 10pF ceramic will have a very good dielectric and is very close to an ideal capacitor. I suspect that you have a stability problem somewhere.