• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Capacitor upgrade in crossover - You CAN'T handle the TRUTH! - Part 3

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,088
Likes
7,544
Location
Bjerringbro, Denmark.
As far as I can see, it's about acoustical measurements picturing the electromechanical behavior of speaker drivers. Like I said before: It has zero relevance to the electrical behavior of capacitors(?).
 

pogo

Major Contributor
Joined
Sep 4, 2020
Messages
1,241
Likes
382
The 3D Step Response Measurement has the following advantages to make the sound quality of the reproduction more visible (see also my post #114):

- Stimulis, which come closer to music, etc
- 3D graph, which contains more comparable interrelated components

Of course, this measuring method is not limited to the referenced cases. With a suitable signal/measuring chain and adequate resolution, it should be possible to show differences, when there are some present.

I don't know whether e.g. ARTA supports such a measurement.
 
OP
C

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,086
Location
.de, DE, DEU
The 3D Step Response Measurement has the following advantages to make the sound quality of the reproduction more visible (see also my post #114):

- Stimulis, which come closer to music, etc
- 3D graph, which contains more comparable interrelated components
You might want to mention that these statements serve to promote a commercial measurement system.

I don't know whether e.g. ARTA supports such a measurement.
Arta has no 3D representation of decay processes, but the decay process can be displayed there either time based (CSD) or oscillation periods based (burst decay).

When using the burst decay there was no difference between the high-end capacitor compared to a normal film capacitor too.


@pogo it is interesting, you try with all means to cast doubt on the results, which is okay. You should question everything and check for possible errors.

On the other hand, did you question your "belief" in the "audible sound gain by high-end capacitors" just as vehemently?
If so, then you should have no problem to provide real facts about the measurable and above all audible improvements (facts, not opinions) when using high-end capacitors.

Why doesn't any manufacturer of high-end capacitors participate and show how to make the differences in the audible range of loudspeakers visible/measurable?

As manufacturer you need to be able to measure the tonal change in the loudspeaker, otherwise you can't optimize your capacitors and have to guess what will happen if a change is made - via simple hearing tests, as we all know by now, this does not work.
 

pogo

Major Contributor
Joined
Sep 4, 2020
Messages
1,241
Likes
382
CTRL,

the burst decay measurement from ARTA uses a different stimuli and does not show the behaviour of the sound reproduction in the non swung-in state. This behaviour also seems to be very important, as previously cited. Therefore, if you want to get closer to the complete truth, you should also measure this additionally.

The discussion of what manufacturers do or not do is not useful here.
 

InvertedFlight

New Member
Joined
Sep 24, 2020
Messages
3
Likes
0
CTRL,

I purchased a pair of old Acoustic Research speakers AR9, made in late 1970s. They are being delivered as I type this and I began to read through various posts, primarily on classic speaker pages, how caps in these vintage speakers must be replaced to avoid tweeter and mid driver damage.
I cannot find exact quotes at the moment but the logic behind this argument suggested that out of spec cap would allow signal that was beyond what the driver could safely handle and effectively damage it.

Fearing the outcome, I began to read more about recapping crossovers and fortunately stumbled on your posts here.
Thank you for taking your time and energy to provide solid scientific evidence with regard to this.

With that in mind, do you see any need for me to replace these old capacitors? Is there truly a risk of damage to drivers if I don't ? One of the caps appears to be leaking, please see pictures below. I'm assuming it may not function properly but I have no tools to test it. Should it be replaced?

I would be most appreciative if you could share your thoughts on this.
Thank you for taking the time to read this reply and my apologies to everyone if I'm hijacking the thread.
 

Attachments

  • xover 1.jpeg
    xover 1.jpeg
    963.1 KB · Views: 393
  • xover2.jpeg
    xover2.jpeg
    549.6 KB · Views: 616
Last edited:
OP
C

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,086
Location
.de, DE, DEU
...how caps in these vintage speakers must be replaced to avoid tweeter and mid driver damage. I cannot find exact quotes at the moment but the logic behind this argument suggested that out of spec cap would allow signal that was beyond what the driver could safely handle and effectively damage it.
As electrolytic capacitors age, their capacitance values change (decrease) and their dielectric strength decreases.

Personally, I would definitely replace very old electrolytic capacitors. In the tweeter branch of the crossover I would use foil capacitors (if electrolytic capacitors are also used there). An LCR meter (hardware or software) can be helpful.

Damaged capacitors must of course be replaced immediately and care must be taken with old electrolytic capacitors, as they contain PCBs in the electrolyte.
 

InvertedFlight

New Member
Joined
Sep 24, 2020
Messages
3
Likes
0
Very much appreciate your reply !

One more question if I may, and forgive my complete ignorance.
Can I replace an original 80mFd cap in the crossover with a 82mFd or will that negatively affect the performance ? I'm having difficulty finding such a capacitor.
 
Last edited:

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
Very much appreciate your reply !

One more question if I may, and forgive my complete ignorance.
Can I replace an original 80mFd cap in the crossover with a 82mFd or will that negatively affect the performance ? I'm having difficulty finding such a capacitor.
That’s well within tolerances.
 

Stomper

New Member
Joined
Oct 21, 2020
Messages
3
Likes
0
Like InvertedFlight, I'm looking for some advice related to upgrading capacitors and whether they're worth it. I have a pair of Technics SB-6000 two-way speakers, circa 1977, and would like to do a recap. I like the way they sound, so am not looking to make them sound 'better'. The factory caps are Panasonic Y Series (radial, bi-polar, electrolytic): 3.3, 10, 22, 33 and 68 uF. It appears these caps are (not surprisingly) no longer available. Questions:

1. The Panasonic documentation for Y Series states they are "designed specially for crossover networks", as does the spec sheet for possible replacements - Nichicon UDB Series. Is this just marketing hooey, or are there specific design/construction characteristics of a crossover capacitor (vs an amp) I need to be aware of when choosing a replacement? Or, will any 'audio' capacitor do?

2. Should I should stick with electrolytic to not risk changing the sound beyond what aging might've already done?

3. Do you have any recommendations for replacement caps given the market value of the speakers is likely not more than $300 to $400?
 

beagleman

Major Contributor
Joined
Sep 24, 2020
Messages
1,156
Likes
1,576
Location
Pittsburgh Pa
If a listener is able to easily perceive a sound difference, it follows they should be able to demonstrate they can identify that sound just as easily in a comparison.
Someone mentioned that sound is subjective, but a blind test is objective.

I do not agree with that. A blind test uses the listeners SAME subjective impressions, but puts doubt into their brain, as their set in stone biases and expectations, are being put to the test by this method.

If the listeners subjective impressions are highly colored by expectation and bias, they will not be able to demonstrate which is which reliably.

IF their impressions are not biased or based on expectations, it follows, they should easily be able to tell which is which.

I see it as less of a hearing or component test, but a test of a listeners biases and expectations.
 
OP
C

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,086
Location
.de, DE, DEU
I like the way they sound, so am not looking to make them sound 'better'.
If you like the sound and no capacitor is obviously damaged, everything is fine, why change this?

The factory caps are Panasonic Y Series (radial, bi-polar, electrolytic): 3.3, 10, 22, 33 and 68 uF.
Guess the 3.3µF capacitor is in series to the tweeter, if this is the case it should be replaced by a (cheap) film capacitor (usually +-3-5% tolerance).
I would replace all other bipolar electrolytic capacitors (usually +-10% tolerance) with such again (as long as they are not in series with the tweeter).

If the capacitors are replaced without measuring the capacitors (old and new) with an LCR meter, the sound may change slightly and the pair deviation of the speakers may increase.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
Like InvertedFlight, I'm looking for some advice related to upgrading capacitors and whether they're worth it. I have a pair of Technics SB-6000 two-way speakers, circa 1977, and would like to do a recap. I like the way they sound, so am not looking to make them sound 'better'. The factory caps are Panasonic Y Series (radial, bi-polar, electrolytic): 3.3, 10, 22, 33 and 68 uF. It appears these caps are (not surprisingly) no longer available. Questions:

1. The Panasonic documentation for Y Series states they are "designed specially for crossover networks", as does the spec sheet for possible replacements - Nichicon UDB Series. Is this just marketing hooey, or are there specific design/construction characteristics of a crossover capacitor (vs an amp) I need to be aware of when choosing a replacement? Or, will any 'audio' capacitor do?

2. Should I should stick with electrolytic to not risk changing the sound beyond what aging might've already done?

3. Do you have any recommendations for replacement caps given the market value of the speakers is likely not more than $300 to $400?
Use decent brand (not audiophile!) bipolar electrolytics, and if their ESR is lower than the originals, put in a small series resistance to make up for that.
 

Stomper

New Member
Joined
Oct 21, 2020
Messages
3
Likes
0
If you like the sound and no capacitor is obviously damaged, everything is fine, why change this?

I'm so happy you asked. I've been led to believe (in a few forums) that any 40+ year old electrolytic capacitor is too old/degraded to use, and might even be dangerous, so should be replaced. Although I was thinking it would be fun to test my skills in a recap, I'm not necessarily looking for extra work. :)

Guess the 3.3µF capacitor is in series to the tweeter, if this is the case it should be replaced by a (cheap) film capacitor (usually +-3-5% tolerance).
I would replace all other bipolar electrolytic capacitors (usually +-10% tolerance) with such again (as long as they are not in series with the tweeter).

Dumb question: How would I know if they are "in series with the tweeter"?

If the capacitors are replaced without measuring the capacitors (old and new) with an LCR meter, the sound may change slightly and the pair deviation of the speakers may increase.
 

Stomper

New Member
Joined
Oct 21, 2020
Messages
3
Likes
0
Use decent brand (not audiophile!) bipolar electrolytics, and if their ESR is lower than the originals, put in a small series resistance to make up for that.

I'm finding it a challenge to locate radial, bipolar caps of any brand with the capacitance and voltage (50) required. The only ones I've found so far are Nichicon.
 
Last edited:

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
Why do they need to be radial?

What value and voltage are you looking for?

edit: NM, you said what the values are. Digikey appears to have all of them.
 
Last edited:
OP
C

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,086
Location
.de, DE, DEU
I'm finding it a challenge to locate radial caps of any brand with the capacitance and voltage (50) required. The only ones I've found so far are Nichicon.
A higher dielectric strength (e.g. 100V, 250V) is no problem.
All capacitance values belong to the E6 series and should be easy to find.
If only the axial design is available, this can be installed like radial with some shrink sleeve and extension of a connecting wire.
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
A little bit of knowledge is so limiting. o_O

It is good that knowledgeable ASR members are willing to be helpful.
 

dennnic

Member
Joined
Jan 13, 2021
Messages
86
Likes
9
Location
Serbia
Hello to all folks around, I registered just to follow this thread.

Replacing one part with another of identical specs and getting the same result makes sense. However, you have to agree that there are many similar threads, posts and experiences from different users, who all claim they hear some difference.

Agreed, some difference may come from age-altered specs of old capacitors that were replaced with a brand new ones. However, if expensive capacitors do not make any change in the sound, why are they used by speakers manufacturers at all? If they all sound the same, why don't we just see a basic $3 crossovers in speakers, especially in ones under $1000-$1500 price?
I seriously doubt that a manufacturer would bother to put better caps in a relatively cheap speaker, if they don't do anything at all. Most of buyers (especially buyers of more affordable equipment) doesn't have a clue about that stuff and the rest of them won't really care.
 
Last edited:
Top Bottom