Sansui big capacitor replacement - disappointment

[Morden2004]

More info on ESR

As a public service :yes: here is a page from my ESR meter manual. I think it helps see where I have been with my concerns for ESR readings.

However, as you read that last paragraph you will note that this is not an exact science. Only through experience have ESR values been observed in devices where problems were evident and were eliminated by replacing those electrolytics.

Read on:

ESR Manual

Paul

 

[bluesky]

Thanks

I have always wondered about caps. Thank you for your most valuable information. It is so very much appreiciated!!!!!!!!!!!!!!

 

[EchoWars]

For myself........I don't usually even bother to "turn-on" a unit that is undergoing restoration. I totally re-cap it first.......then troubleshoot any "bugs" that remain afterward. It's simply more efficient, and it allows me to "debug" a unit ONCE before its finished.

Things like doing the amp calibration.......are often tricky "unrestored" and will have to be done once its recapped anyway, so I simply eliminate the caps as a problem area before doing it. In a related vein for instance....the fuseable resistors on the driver pcbs of most Sansui amps have drifted way out by now in unrestored units. We replace all of them immediately during a restoration...

I've pretty much adopted the same procedure after chasing my tail on a few units that ended up being 99% cured with a capacitor swap...not the usual scenario for sure, but made enough of an impression on me and wasted enough of my time that I recap and replace fuse resistors first, and troubleshoot later.

I'm amazed by the amazement of others in the thread who seem to be discovering for the first time (despite all the info that has been posted over the history of AK, and dozens of other web sites) that old electrolytic caps are problematic. The information is there, the curious need only look.

As far as the significance of ESR, the small measured value should not be ignored. This ESR value cannot be properly compared to a linear value, such as a simple resistance, because it does not behave that way within the audio frequency spectrum. The electrolytic cap is inherently non-linear device due to its construction (series resistance, current leakage, inductance, dielectric absorption...all varying with frequency), and yet is used as if it were a linear device, which entails compromises. All the designer (and the restorer) can do is to minimise the nonlinearities by choosing the best cap he can to replace aging caps where these nonlinearities have become troublesome to the point of audibility or circuit failure.

 

[Kjell-B]

That old electronic components may be problematic is well known.

The problem is that if all old components are replaced by modern production, the amplifier is no longer a vintage amplifier, and the customer may well go shopping for a modern hifi amplifier instead.

If I were to deliver my amplifier (or tuner, etc.) to a service shop, my main concern would be the preservation of the original sound. That would require the identification of INDIVIDUAL defective components, not a total change of all components.

I am glad this practice of restoration has been revealed and discussed; then the potential customer knows the options.

 

[EchoWars]

I don't know anyone who is advocating the 'total change of all components', only the components that are worn-out and contributing to failures and undesired sonics.

Again, the topology of an amp contributes to 99.9% of its sonic character, as long as the components used in that topology are appropriately chosen for the task at hand. If this were not so, there'd be one amplifier for sale, and we'd all be using it. If you doubt this premise, no one here is going to be able to convince you differently. Head over to diyaudio.com, and ask guys like Nelson Pass and John Curl, who make their living designing amps and preamps. Their response (and others) should be enlightening for you.

 

[Crazy]

But you --> Echowars & others are "advocating the 'total change of all electrolytic caps" as preventive.?????

Correct?????

And one thing I just don't "get" is why if electrolytic caps have such a life that you find so many "stock" vintage Hi-Fi's still cranking along just fine.

It would seem logical that all "stock" vintage stuff over say 25 years at most would be non-functional according to conventional wisdom.

I mean if it's used it's way past the "hours of life"..........if not used it's worse and should be dead.

That's the point that keeps sticking in my craw.

As a "non-expert", I tend to take the simple approach of 1+1=2.

hmmmmmmmmmmmmmmmmmmmmm.....................

 

[EchoWars]

But you --> Echowars & others are "advocating the 'total change of all electrolytic caps" as preventive.?????

Correct?????

'Total change of all electrolytic caps' is a far cry from 'total change of all components'.

And one thing I just don't "get" is why if electrolytic caps have such a life that you find so many "stock" vintage Hi-Fi's still cranking along just fine.

It would seem logical that all "stock" vintage stuff over say 25 years at most would be non-functional according to conventional wisdom.

I mean if it's used it's way past the "hours of life"..........if not used it's worse and should be dead.

That's the point that keeps sticking in my craw.

Not everything ages or even fails at the same rate, it's as simple as that...too many manufacturing, construction, environmental, and usage variables. Take some electrical engineering classes on failure statistics and analysis...the scope of the subject is beyond anything but cursory examination here.

 

[pustelniakr]

If I may, I would like to contribute a bit of boiler-plate from my DIY threads:

Re-Capping: Motivations, Considerations, Results

--------------------------------------------------------------------------------

My experience: Recapping generally yields only an incremental improvement in the way a unit sounds, but should significantly increase the life-expectancy of a unit.

Here are your incentives/motivations, so balance them against cost and/or effort:

I have measured a large number of caps, replaced during a recap. The majority are at or below the low end of their specs, unless they are in a location where they get thermally stressed.


When 'lytic caps go, as the odds of doing so rises rapidly once their rated life expectancy of around 20 years has been reached or exceeded, they go any one of 3 ways, drift out of tolerance, fail open, or fail short. They all drift, and statistically, failing open or short is about a 50/50 mix.


If a cap fails open, you may notice, you may not, depending on where it is in the circuit. Generally the fault is easily repaired, with minimal peripheral damage.


The main risk: If a cap fails short, it can, and generally does, take out significant associated circuitry (ie. semiconductors made out of unobtainium).


Recapping is cheap, part-wise, averaging $0.35 per cap, but labor is high. A good tech can be counted to be able to replace around 10 caps per hour (including unit disassembly, cap removal, pad cleanup, cap installation, and flux cleanup). Even if the tech only charges $10 per hour, in a unit with 150 caps to replace (not uncommon on gear I work on), the labor really mounts up when it takes 15 hours to do the job. Typical modern labor costs are in excess of $60 per hour. AK restoration artists are not in it for the money (but money is nice), so we generally do not charge anywhere near that rate, but our labor is worth that rate.


The MOST IMPORTANT aspect of a recap, is the opportunity to refresh all circuit board solder joints. Flow-soldered connections have only about a 30-year life expectancy. In my experience, I have solved more problems caused by old, oxidized, fractured solder joints than by failed/drifted caps. Here is where the real bang for the buck is realized, and MUST NOT be underestimated.

With all of these aspects considered, recapping is really only a viable procedure for upper end gear, unless the gear owner is an avid collector, and wants all of a series in excellent condition, etc.

Emphasis:
Recapping is not the panacea for modern man. Recapping is the last thing I do to a unit, other than cosmetic work and final/performance testing. I won't recap a unit, until I have determined that it is fully functional. Recapping is more for an attempt at adding longevity to the mix. The last thing you want is for a 30-year-old 'lytic to fail short (they also fail open, and drift out of spec). Recapping is not a good "repair" technique (falls in the realm of 'shotgun repair', which I hate).

If you have the product knowledge to know that your time is better spent recapping a piece or model before repair/restoration, more power to you, since that is not shotgun repair. It is the product of experience and knowledge. Otherwise, fix wht is broke before recapping. That way you will know if you break something during a recap, which is a possibility, unless you are perfect.

My contribution is not intended to fuel, or add to an argument. I offer only my experience and knowledge. Folks can disregard or not. I will suffer no loss, either way. I do know this: recapping is normally a royal pain in the neck. I would prefer not to have to recap, but my conscience will not let me sell gear that has not been recapped. If I restore customer gear, I offer the recap as an option, and an expensive one (for myself and the customer).

If you like the sound of your vintage gear, with its old caps, and are willing to accept watever the risk is to not recap, that is you own choice. It is my hope that you will make an informed decision, and not a platitudinal or knee-jerk decision.

I have recapped many units. Sonic improvement has been a peripheral result in every case so far. I have had no complaints. I have never heard a complaint about the EWs work either, and, I think we are standing in the same area of the map on this topic. Your results may vary.

I do encourage you not to desparage folks who are purists, and wish to keep old gear as such. Likewise, please do not cast aspersions on folks who assess the risks and recommend minimizing them as much as possible.

Enjoy,
Rich P

 

[Kjell-B]

Well, I can only repeat that vintage electrolytic capacitors often are in very good shape unless stressed by heat or high ripple (close to the rectifier). Also, the output capacitor in AC coupled amps may have a more limited life span.

When I remove caps from 40 years (or older) equipment, they are almost always above specs re. capacitance (and leak). In the few cases I find defective caps, they show total failure.

I believe most vintage equipment can be used for 100 years or (much ?) longer with only very little maintenance or repair. And absolutely no total recap. Some brands may show more problems than other. In my hands US caps seem to have more limited life span than early European and Japanese.

If I buy an amplifier (etc.) and find there are many defective components I buy one more and use one of them for delivering parts. It has alway worked well for me - I am using many types of vintage equipment daily and never experience any problems - I have never repaired any item more than the initial service when purchased. The reliability is excellent !

Also my extensive tube equipment was built using mostly 40-60 years old radio parts (and a few new parts) - I have never repaired anything later. The old components (incl. electrolytics) hold very well. This is one reason I have engaged myself in this debate.

I think the main reason for the need of repairs is the selection of equipment on eBay and similar places: People tend to sell their old equipment when there are some symptoms of partial failure in the electronics. This may lead to the impression that old equipment often needs repairs. Which is wrong.

 

[Kjell-B]

Head over to diyaudio.com, and ask guys like Nelson Pass and John Curl, who make their living designing amps and preamps. Their response (and others) should be enlightening for you.

Just a little off record, John Curl tells that one of his most precious audio gear is the old Telefunken Bajazzo portable radio from early 1960's. I have myself the same radio. It is with germanium transistors only, transformer-coupled audio stage and is literally filled with electrolytic capacitors. I am sure the radios were never recapped !

 

[EchoWars]

This may lead to the impression that old equipment often needs repairs. Which is wrong.

Huh?:scratch2: You might re-examine that statement.

Just a little off record, John Curl tells that one of his most precious audio gear is the old Telefunken Bajazzo portable radio from early 1960's. I have myself the same radio. It is with germanium transistors only, transformer-coupled audio stage and is literally filled with electrolytic capacitors. I am sure the radios were never recapped !

And this has to do with.....what? Did you ask the Diyaudio.com members about recapping old gear?

Again, this is all going nowhere. You and other are more than entitled to your opinion, but I ask that you tread cautiously when suggesting that I and others such as QuadBob are either taking advantage of those who we have 'cajoled' into thinking that old caps much be replaced, or that we are destroying the value of vintage electronic gear. One thing you may not know is that I (and Bob also, I think) spend the vast majority of our efforts working on mid-70's to early-80's era Japanese gear...hardly the thing that people are collecting to resell for stratospheric profits in their golden years (although the values of some of these units are getting to be rather steep of late).

Funny that this morning I just opened up a Hitachi HMA-8300 power amp. A nice amp, and kind of a rare bird. Was given to me for a 'tune-up', and as far as the owner knew, was operating fine. On the bench, offset looked good, and the bias was in the ballpark, so I connected dummy loads and a sig-gen, and scoped the output while I raised the signal level.

The amp refused to manufacture more than about 60WPC without the protection engaging (this on an amp rated at 200WPC). Monitoring the high-current supply voltages while increasing the signal showed that the positive rail was dropping from a quiescent +95V to about +70V under the mild load of 22VRMS into 8 ohms. I had some large power supply caps here that would suffice as replacements, and once installed, the amp promptly made 325WPC into 8 ohms. The faulty power supply cap was NOT leaking, and showed no signs of venting or bulging. It even tests good, with acceptably low ESR. But...it is nevertheless a bad cap.

The owner never even knew there was a problem, but I see this sort of thing every day!

I'll send you the cap, if you want. :yes:

I'm done.:zoom:

 

[Crazy]

Hold on...........

I really hope nobody is under the impression that I ----->

" am suggesting that Echowars and others such as QuadBob are either taking advantage of those who they have 'cajoled' into thinking that old caps much be replaced, or that they are destroying the value of vintage electronic gear."

Because I am not and am not even close to any thinking like that. I also don't read that into any other posts.

I can only speak for myself, naturally.............but my entire focus has been on me as a newbie to this stuff trying to figure out what the correct moves are in maintaining hi-fi stuff. What needs to be done, when to do it and why.

This whole thing has not been any type of attack but just a simple debate.

So far it looks like E-caps should "ALL" fail after 20 years max according to "science". Yet many are found to be alive and kicking at 30 years +. According to known scientific analysis...all "stock" 70's Hi-Fi's should be long dead.....yet many, many are not. ( This IS a clue I think)

It looks like a total recap is fine & dandy if you are a tech talented enough to do it..........but it seems like maybe hiring out that job might not be cost effective and taking a wait & see stance might be the smarter move if you are just the average DIYer or an average vintage hi-fi lover.

And like everything else it "appears" that science has many variables and predictions based on science are not always as predictable as they seem. Or maybe all the variables have just not been brought to light yet????:scratch2:

For myself...........I don't have a tech anywhere close to me, number 1.
Number 2, even if I did, I don't think I would spend hundreds or a thousand on recapping a 1970's $50 to $100 receiver. EVEN THOUGH I REALLY ENJOY THEM! :yes:

But that is MY TAKE. TO EACH THEIR OWN. NO HARM-NO FOUL.

What I think I will do though personally is------------->

Just keep playing my "perfectly working" ( or so it seems) 8080DB.........see what happens over the next 1 day to 20 years, lol. It was $150 bucks to buy..........I'll gamble it cranks on for at least a couple more years. Call it an experiment if you will.
My "spare" 881 will get solder practice and such performed on it. Then my functioning 881 with the new power caps will get a total recap performed by me ( hell, I've broke it once and it recovered.let's have another go! LOL). I shall go slow and "try" to do my homework prior to surgery.:thmbsp:
My other stuff will get used just enough to keep it from rotting until I become good at dealing with testing and recapping. I also have lots of and can get lots of cheap crap like old $5 Kenwoods and such to dissect and experiment on.

You see guys/gals and any aliens that might also monitor this forum............

THE REASON I am such a royal pain in the a$$ is because I'm here to learn..........and humanoids like me wanting to learn are known to be trying on the nerves.

AND--------> I THANK ALL who take the time to help. ( FROM THE BOTTOM OF MY HEART)

Ya never know...........I may just learn enough to help some other pilgrim down the road..............

AIN'T THAT WHAT "AUDIO KARMA" is all about???????????????????:scratch2:

 

[Crazy]

'Total change of all electrolytic caps' is a far cry from 'total change of all components'.

I realize this and was just attempting to clarify the point :yes: :thmbsp:

Not everything ages or even fails at the same rate, it's as simple as that...too many manufacturing, construction, environmental, and usage variables. Take some electrical engineering classes on failure statistics and analysis...the scope of the subject is beyond anything but cursory examination here.

Exactly........and this "variable" is a major sticking point with me. Guess I need to hone up my "testing talents" and equipment to be able to see this all more clearly.

PEACE BABY........it's just a debate & I thank you for your responses.:yes:

 

[Crazy]

Well, I can only repeat that vintage electrolytic capacitors often are in very good shape unless stressed by heat or high ripple (close to the rectifier). Also, the output capacitor in AC coupled amps may have a more limited life span.

When I remove caps from 40 years (or older) equipment, they are almost always above specs re. capacitance (and leak). In the few cases I find defective caps, they show total failure.

I believe most vintage equipment can be used for 100 years or (much ?) longer with only very little maintenance or repair. And absolutely no total recap. Some brands may show more problems than other. In my hands US caps seem to have more limited life span than early European and Japanese.

If I buy an amplifier (etc.) and find there are many defective components I buy one more and use one of them for delivering parts. It has alway worked well for me - I am using many types of vintage equipment daily and never experience any problems - I have never repaired any item more than the initial service when purchased. The reliability is excellent !

Also my extensive tube equipment was built using mostly 40-60 years old radio parts (and a few new parts) - I have never repaired anything later. The old components (incl. electrolytics) hold very well. This is one reason I have engaged myself in this debate.

I think the main reason for the need of repairs is the selection of equipment on eBay and similar places: People tend to sell their old equipment when there are some symptoms of partial failure in the electronics. This may lead to the impression that old equipment often needs repairs. Which is wrong.

But you are saying you find 1970's Japanese caps to be "good ones"/ high quality also????

And this is from personal experience like stated with your older stuff???

 

[Kjell-B]

Funny that this morning I just opened up a Hitachi HMA-8300 power amp. A nice amp, and kind of a rare bird. Was given to me for a 'tune-up', and as far as the owner knew, was operating fine. On the bench, offset looked good, and the bias was in the ballpark, so I connected dummy loads and a sig-gen, and scoped the output while I raised the signal level.

The amp refused to manufacture more than about 60WPC without the protection engaging (this on an amp rated at 200WPC). Monitoring the high-current supply voltages while increasing the signal showed that the positive rail was dropping from a quiescent +95V to about +70V under the mild load of 22VRMS into 8 ohms. I had some large power supply caps here that would suffice as replacements, and once installed, the amp promptly made 325WPC into 8 ohms. The faulty power supply cap was NOT leaking, and showed no signs of venting or bulging. It even tests good, with acceptably low ESR. But...it is nevertheless a bad cap.

The owner never even knew there was a problem, but I see this sort of thing every day!

Thank you for a very interesting and important information.
And I agree completely.

If we are speaking of large mammoth amplifiers 1975-85, a total recap of the power section/power amp section may be just the right thing. The risk of catastrophic failure overrides any other concern as audio quality of components, etc. Also, the quality of capacitors declined around 1980 and into the 90’s.

Obviously our disagreements in this thread is partially caused by lack of specifying what equipment we are talking about. Also my fault.

But vintage comprise much more than large power amps.

Vintage for me are tuners, tape recorders, equalizers, processors, preamps, low power integrated amps (max 2x30 watts). Here the risk of destroying anything is very low, and most old capacitors may stay where they are.

Electrolytic caps which should be replaced are caps stressed by high temperature, caps closest to the rectifier in power supply, and output caps in AC-coupled amps. But these are relatively few caps. All the others caps in a vintage product should in my opinion not be changed unless really defective.

 

[targeteye]

(this on an amp rated at 200WPC). .....I had some large power supply caps here that would suffice as replacements, and once installed, the amp promptly made 325WPC into 8 ohms. .:zoom:

Did you mean 225WPC? Otherwise this may be the king of under rated amps!

Steve

 

[riverrat]

Fascinating thread, one of the best ever here. A fair amount of it is going over my head, but I'm still learning alot.

I also appreciate the respectful tone, which can be hard to maintain due to the impersonal nature of web-based communication. Significant disagreements, or perhaps not so significant, informed viewpoints being put forth, lots of quality info. Really nice to see!

 

[EchoWars]

Did you mean 225WPC? Otherwise this may be the king of under rated amps!

Steve

No typo. 325WPC.

 

[targeteye]

Electrolytic caps which should be replaced are caps stressed by high temperature, caps closest to the rectifier in power supply, and output caps in AC-coupled amps. But these are relatively few caps. All the others caps in a vintage product should in my opinion not be changed unless really defective.

I tend to agree with that statement if indeed to goal is to maintain to greatest extent possible the authenticy of the peice AND folks take the time to check the Leakage and Capacitance of every Electrolytic and Paper cap in the particular device and if your really trying to preserve the SOUND the ESR as well. In addition special care must be given to check periodically the power supply caps elswise a you may need your spare chassis for its power tranny.

Most people do not have the time, desire, or resources to do what you describe. My hats off to you.

edit: missed the part about caps closest to the rectifier, so my statement about power supply caps is invalid

Steve

 

[wsjoe]

Specs are one thing and listening tests are another.
Electrolytic capacitors may measure within specs but when replaced with new ones, the soundstage completely changes. The highs and low frequency content just comes alive. CDfixer and I have experienced this many many times. We thought we were wasting time and money by replacing larger capacitors inside 8080xx and 9090xx Sansui gear from the 70's, but one day, CDfixer replaced the large 12,000 uf filter capacitors in the power stage. Boy it made a huge difference. It was like listening to FM stereo compared to AM. If you don't hear this change, you might need a hearing aid to pickup the high frequency content.

Just from my experience.