Ok this has probably asked a million times but i can't seam to find any real answers to what I'm asking.
I acquired some leak sandwich speakers (mini and 2way) they still have what looks like the original bipolar electrolytic capacitors that must be getting on for 40 years old now. I understand that replacing them with film caps will do them no end of favours. I also have an stereo 70 amp, in need of a rebuild, that has high value speaker coupling capacitors standard are 2200µf but can, i believe, be pushed to about 4700µf. questions are.
1. Speakers:
The cap in parallel with the woofer does it need to be a really good poly type?
The tweeter cap as it is directly in the signal path it should be a very good poly cap?
I understand the multiple caps paralleled are better?
The woofer inductor in the 2way are bypassed with a large carbon comp 22 Ohm resistor then a 1µf capacitor. Reading from left to right on the positive line: supply-inductor-woofer.
does this need to be a poly type cap and does the resistor need changing.
2. Amp:
What should i be using to couple the speakers. At the moment it has polarised electrolytics in it. i believe that in this place theoretically i should have a film cap but that wold be massive.
3. what is the right way of paralleling capacitors to get the value you want. If i can get two/three/four values the same to add up to what I want then good. But what if I can't and i have to use say one smaller value and one larger value. I only ask because i read somewhere that if the values are different but too close then it wont work as well. can't find it now.
Finally should i be bypassing any of these caps with ?% smaller film caps.
Thanks
Tom
I acquired some leak sandwich speakers (mini and 2way) they still have what looks like the original bipolar electrolytic capacitors that must be getting on for 40 years old now. I understand that replacing them with film caps will do them no end of favours. I also have an stereo 70 amp, in need of a rebuild, that has high value speaker coupling capacitors standard are 2200µf but can, i believe, be pushed to about 4700µf. questions are.
1. Speakers:
The cap in parallel with the woofer does it need to be a really good poly type?
The tweeter cap as it is directly in the signal path it should be a very good poly cap?
I understand the multiple caps paralleled are better?
The woofer inductor in the 2way are bypassed with a large carbon comp 22 Ohm resistor then a 1µf capacitor. Reading from left to right on the positive line: supply-inductor-woofer.
does this need to be a poly type cap and does the resistor need changing.
2. Amp:
What should i be using to couple the speakers. At the moment it has polarised electrolytics in it. i believe that in this place theoretically i should have a film cap but that wold be massive.
3. what is the right way of paralleling capacitors to get the value you want. If i can get two/three/four values the same to add up to what I want then good. But what if I can't and i have to use say one smaller value and one larger value. I only ask because i read somewhere that if the values are different but too close then it wont work as well. can't find it now.
Finally should i be bypassing any of these caps with ?% smaller film caps.
Thanks
Tom
Replace, replace, replace !
It's the electrolyte that dries ...some say ! I'm using some bipolars from the '70s too,
and they're still fine ! Different is the output coupling cap of the amp ,which has to be perfect . A good industrial type aluminum I think is best ( like the old ones on some Philips amps );replace also the supply cap and all the old ones ; probably you'll hear a different sound , more richer in the bass .
For the cap on the tweeter , go for 250 V or higher . But you could surprisingly find that
a 50 cent bipolar suits to your taste
( IMHO !!!)
It's the electrolyte that dries ...some say ! I'm using some bipolars from the '70s too,
and they're still fine ! Different is the output coupling cap of the amp ,which has to be perfect . A good industrial type aluminum I think is best ( like the old ones on some Philips amps );replace also the supply cap and all the old ones ; probably you'll hear a different sound , more richer in the bass .
For the cap on the tweeter , go for 250 V or higher . But you could surprisingly find that
a 50 cent bipolar suits to your taste
( IMHO !!!)Tom, everything depends on exact speaker model. Generally, replacing electrolytics to poly will bring the improvements, but if speaker response is not flat enough, replacing electrolytic in mids-highs with poly-type capacitors can make sound more exact, without smoothing, but can highlite, for example peak and make your high-mids very unpleasant. So, you need to do some corrections in crossover to change speaker response. Same job as engineering speakers (proper equipment, software, etc.). Same with bass: it can be more exact with poly, but you can loose some "volume" - sound can be a bit "dry".
There is no exact answer - yes or no, you need to try.
Place a Solen across your woofers, and some CrossCaps, ClarityPX, or Erse PulseX caps for starters on the tweeter.
I'd use a CrossCap for the 1uF in that notch on the woofer.
The large 'Lytics in an amp should be okay.
All components are in the signal path, series or parallel, so that archaic thought process is incorrect.
Later,
Wolf
I'd use a CrossCap for the 1uF in that notch on the woofer.
The large 'Lytics in an amp should be okay.
All components are in the signal path, series or parallel, so that archaic thought process is incorrect.
Later,
Wolf
Last edited:
The spec for electros is about 7 years. Usually, unless abused by high current or high temp, about 10 years is their life. Poly caps will be much closer to the designed spec. 5% or better, where electro's are 20% on a good day, 50% on a bad. That can be a large part of the sound change when caps in filters are changed.
The difference between an electro and a mylar is big. A small step to poly-pro. Tiny improvement to exotic, tight wrapped audio caps. Let your wallet be your guide. I use the 1% Daytons.
In the amp, yes, you can put in larger main caps and lower ESR. If you do, you should really think about a larger rectifier. The change to a hexfred rectifier and larger low ESR caps made a very large and measurable difference in my DH-120. Over 8 dB lower noise floor.
The difference between an electro and a mylar is big. A small step to poly-pro. Tiny improvement to exotic, tight wrapped audio caps. Let your wallet be your guide. I use the 1% Daytons.
In the amp, yes, you can put in larger main caps and lower ESR. If you do, you should really think about a larger rectifier. The change to a hexfred rectifier and larger low ESR caps made a very large and measurable difference in my DH-120. Over 8 dB lower noise floor.
Hi,
1a) No. A bipolar electrolytic will be fine.
1b) Standard poly type will be well good enough.
1c) Sometimes. It doesn't hurt.
1d) No and no, though 1uF is cheap anyway that you do it.
2) A polar electrolytic bypassed by a much smaller similar or poly.
(In fact replace all electrolytics in the amplifier, doubling the values
of all supply, coupling and decoupling capacitors, new ones will fit.)
3) Make up the parallel total any way you like, don't be too precious.
The Leaks don't need flash capacitors. Just ones that work properly.
New bipolar electrolytics will make far more difference than upgrading
those to basic or flashy film types paralleled or not.
My friend has an original pair. Not worth boutique components IMHO.
rgds, sreten.
http://www.falconacoustics.co.uk/
Last edited:
Sounds like you require a refresher yourself. If the components in shunts don't make any difference, then notches, zobels and such wouldn't be used.
It all matters.
Later,
Wolf
One thing that is often overlooked when thinking about replacing electrolytics with film caps is the fact that the ESR of the caps will likely be quite different (with electros generally having much higher ESR).
Depending on the crossover, the esr in the electros may in fact be adding some damping that won't be there with the film caps, leading potentially to some peaking when film caps are substituted. The other place where electros vs films seems to make a big difference is in notch filters, with my experience being that electros of the same value as films in the same notch give quite different results.
In a lot of cases it shouldn't matter, but it doesn't mean that in all cases it won't matter
Tony.
Depending on the crossover, the esr in the electros may in fact be adding some damping that won't be there with the film caps, leading potentially to some peaking when film caps are substituted. The other place where electros vs films seems to make a big difference is in notch filters, with my experience being that electros of the same value as films in the same notch give quite different results.
In a lot of cases it shouldn't matter, but it doesn't mean that in all cases it won't matter
Tony.
Thanks for the relays and advice.
It seams that in the amp I can just put in and bypass any good quality cap for the speaker coupling. how much smaller should the bypass cap be in the speaker coupling.
in the speakers I it seams that I will gain by just refreshing the caps. But I think I'll experiment and try some different combinations. But the best option seams like a good poly to the tweeter. Good electrolytic over the bass.
thanks
Tom
It seams that in the amp I can just put in and bypass any good quality cap for the speaker coupling. how much smaller should the bypass cap be in the speaker coupling.
in the speakers I it seams that I will gain by just refreshing the caps. But I think I'll experiment and try some different combinations. But the best option seams like a good poly to the tweeter. Good electrolytic over the bass.
thanks
Tom
And yet using a statement that is untrue to define something doesn't make it accurate either. Generally 'in the signalpath' means in series to most users of the phrase. Since this is not DC and current flows both ways, all components are in the signal path. Acceptance of a term is one thing, accuracy of a description is another.
I'm not nitpicking nor trying to split hairs, just stating the facts. Due to the current definition of 'in the signal path' being misconstrued, it should be just stated as 'in series', even though parallel component quality still matters.
Now back to capacitors...
Wolf
It is in the circuit's signal path; whether it's out-of-bandwidth or in-bandwidth is another issue.
Would I think an NPE wrong for an LCR for out of bandwidth- no.
(It really sounds like you think I don't know/understand what I'm doing, and it's not the first time you've tried correcting me.)
Just stating the facts....
Wolf
Would I think an NPE wrong for an LCR for out of bandwidth- no.
(It really sounds like you think I don't know/understand what I'm doing, and it's not the first time you've tried correcting me.)
Just stating the facts....
Wolf
Wolf :
Hi,
I'm not trying to correct you, whilst the simple facts are correct they
don't tell the whole truth, as you well know, as said above it is IMO
nitpicking given the use of the term by the OP was pretty clear.
Of course I don't think you don't know / understand what you are doing.
I didn't address my last post to anyone in particular, whilst series and
parallel is a clear fact, "in the signalpath" or not, is much more debatable,
hence my example, and the implication you need to understand that.
In no way was I implying that you don't know that, but some people don't.
rgds, sreten.
Hi,
I'm not trying to correct you, whilst the simple facts are correct they
don't tell the whole truth, as you well know, as said above it is IMO
nitpicking given the use of the term by the OP was pretty clear.
Of course I don't think you don't know / understand what you are doing.
I didn't address my last post to anyone in particular, whilst series and
parallel is a clear fact, "in the signalpath" or not, is much more debatable,
hence my example, and the implication you need to understand that.
In no way was I implying that you don't know that, but some people don't.
rgds, sreten.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.