why a recap makes no difference

yotems

Well-Known Member
ok now that you’re here and reading this the real question is why may a complete electrolytic recap make no difference in sound? what other hardware ages and is responsible for deteriorated sound? ive heard recaps make subtle differences but ive seen where they have made no changes whatsoever to an amp of age where you would expect SOME difference.

so what all parts have the seasoned vets noticed to contribute to sound deterioration with age?
 
None of the old caps were bad could be one reason there was no change in sound. I've removed old capacitors only to find they measure better than a new one. Another thing I've experienced with several customers, the less technical/theory knowledge the more the sound changed. They spent money therefore it must be better.

Recently had an virgin early Marantz 2270 in the shop, replaced one capacitor, the protection relay was chattering. Went thru the alignments and put it through it's paces on a distortion analyzer, passed all of the distortion/frequency measurements with flying colors with plenty of room to spare.

Craig
 
I have never believed in "total recaps", because as you say sometimes doesn't really make a difference in the sound.
First I will start with the differences in "Failed Capacitors", vs out of tolerence, or high ESR Capacitors.
Failed Capacitors :
An open main power supply cap can produce a hum and or very weak bass and possible distortion.
Power supply caps on regulator boards can fail many times due to the heat associated to those regulator supply boards.
Shorted capacitors usually get pretty hot, can blow out, and many times cause another failure such as an open resistor feeding DC to the circuit with the shorted capacitor.
An open signal passing cap (Coupling) can cause loss of signal, or a very low and tinny signal

Out of tolerance capacitors:
Many circuits in the control amp, tone control, volume control, equalizer, and initial amplifier circuits use low value electrolitics or tantalum caps. These can drift way out of tolerance usually on the high side, and can develop high ESR, (equivilant series resistance), or become leaky. The rated values are usually 0.47, 1, 2.2, 4.7, on up to 10uf. These are the caps that pass signal, and not power supply filters. They many times are referred to as "coupling Capacitors".
I usually test a number of them on a board using a peak atlas ESR and cap tester. If I find differences in value (Greater than 20%) or ESR (Greater than 1.5 ohm) from left or right side, I almost always change them with high end type like Nichicon fine gold, so that I know the signal level, and tone level should be equal. (I test all caps before installing to make sure I can match them as best as possible. ).
A leaky coupling capacitor can cause a noisy control even after its been deox'd, due to the DC voltage appearing on the control as a resuly for the leaky cap. Leaky caps in tone and filter circuits can cause loud popping when actuating switches, or other controls.

Capacitors that I usually leave in place.
Most DC filter caps on the tone, control, equalizer, amplifier, and other boards are usually in the 50uf to 100uf range, and almost always read fine, well within the 20% tolerance, and have sometimes lower ESR than new replacements. I of course spot check some, but very seldom need to change them.

So in a nutshell, Bad capacitors are usually noticable one way or the other, Older DC filters don't usually fail so you may not notice any change whatsoever if you replace them. But Signal passing (coupling Capacitors) do make a difference many times. You just need to test some, to see if there are noticable differences you want to correct.
 
It depends on the circuit too. I've done tons of recaps and sometimes experienced minor changes after, sometimes huge differences. Ie: with cap-coupled output stages - replacing those in amplifiers has made a huge difference in sound when I have done it. Or input coupling caps as well.
 
except for the power caps, the total cost of a recap is oft in the $10 range if you buy in bulk from mouser (i.e. get the 25+ price not the 1-9 price). $10 for parts that are worlds better than the ones replacing. This is so much a no brainer. its sorta like rebuilding your muscle car engine but leaving the rings in because they look good and aint in pieces.

I know this topic comes up every 3-6 months and gets beaten like an expired equine, but a refresh, refurb or restore of any machine dictates juice filled aluminum be yanked and replaced. anything less is just laziness.
 
Other than computer grade filter capacitors the cost IS minimal it's the time people don't want to pay for. Just got in two hacked McIntosh MC240s the other day and the customer brought in some used twist-lock capacitors from 1966 to install so the amps would be original again. These amps are NEVER going to be original again and I talked him into using new modern capacitors as I didn't feel comfortable doing all that work with 50+ year old capacitors.

Craig
 
Its cheap, and while you have it apart, why not? I have found enough old caps with corroded leads and even corroded to the point where the corrosion has eaten through the leads. Usually due to the glues they used to secure them. For the most part I only work on my own stuff and if I have to replace a few I just replace them all while im at it for peace of mind. If its something I'm going to keep I don't want to have to open it up later down the road. To each his own I guess. Nothing wrong with just replaceing failed componants either.
 
Other than computer grade filter capacitors the cost IS minimal it's the time people don't want to pay for. Just got in two hacked McIntosh MC240s the other day and the customer brought in some used twist-lock capacitors from 1966 to install so the amps would be original again. These amps are NEVER going to be original again and I talked him into using new modern capacitors as I didn't feel comfortable doing all that work with 50+ year old capacitors.

Craig
With older tube equipment I always replace all the electeolytics. Some film caps too. Some scary things can happen if they fail in a tube amp. Magic smoke, blown power and output transformers and lighting in a bottle.
 
Other than computer grade filter capacitors the cost IS minimal it's the time people don't want to pay for. Just got in two hacked McIntosh MC240s the other day and the customer brought in some used twist-lock capacitors from 1966 to install so the amps would be original again. These amps are NEVER going to be original again and I talked him into using new modern capacitors as I didn't feel comfortable doing all that work with 50+ year old capacitors.

Craig

What I do with those old twist-lock caps is gut them out and very carefully "stuff" the new electrolytics inside them, or if there's room, mount them in the chassis and simply cut the leads going to those old twist-lock caps. That way, the gear still looks original.

On a side note, I just hit 1000 posts with this post and became a "Super Member"! Yay for me! LOL
 
As others said, sometimes none of the old capacitors are actually bad, or if they are, they're failing in ways which don't really affect the sound.

For the past few weeks I've been listening to a Realistic Stereo 40A amplifier, from 1961, and all of its capacitors are original.. I re-formed and checked them all, and it sounds just fantastic! I would anticipate a re-cap to be very anticlimactic with this piece.

Usually when I do a re-cap, it's because I encounter one or more failed components, and understand the other similar parts cannot be far behind.

As for hardware that ages and is responsible for deteriorated sound?

-capacitors can get leaky or change in value
-tubes can get weak, go noisy, or develop heater to cathode shorts
-resistors can go noisy or drift, especially carbon composition ones
-plugs and sockets can corrode, and make poor contact or lose spring tensiion
-speakers can deteriorate in terms of weakened suspensions, changed resonant frequencies, dried out glues, etc.
-transistors can develop tin whiskers, or can go noisy
-controls can become noisy or dirty,
-switch contacts can wear out
-belts can stretch, idler wheels can flat spot or go hard
-anything with moving parts can suffer from old lubricants getting sticky or drying out
-old ears can lose high frequency response and sensitivity!
 
Last edited:
As others said, sometimes none of the old capacitors are actually bad, or if they are, they're failing in ways which don't really affect the sound.

For the past few weeks I've been listening to a Realistic Stereo 40A amplifier, from 1961, and all of its capacitors are original.. I re-formed and checked them all, and it sounds just fantastic! I would anticipate a re-cap to be very anticlimactic with this piece.

Usually when I do a re-cap, it's because I encounter one or more failed components, and understand the other similar parts cannot be far behind.

As for hardware that ages and is responsible for deteriorated sound?

-capacitors can get leaky or change in value
-tubes can get weak, go noisy, or develop heater to cathode shorts
-resistors can go noisy or drift, especially carbon composition ones
-plugs and sockets can corrode, and make poor contact or lose spring tensiion
-speakers can deteriorate in terms of weakened suspensions, changed resonant frequencies, dried out glues, etc.
-transistors can develop tin whiskers, or can go noisy
-controls can become noisy or dirty,
-switch contacts can wear out
-belts can stretch, idler wheels can flat spot or go hard
-anything with moving parts can suffer from old lubricants getting sticky or drying out
-old ears can lose high frequency response and sensitivity!
I like this, and believe the "Old Ears" comment holds the most weight. Especially old ears like mine who used to work on Aircraft, and has always loved the music LOUD!!!
 
Worked on a 20 year old unit at work. Replaced the Ecaps. Checked them for leakage at rated voltage. One cap had 5X the normal leakage for the cap. Did not want the unit to come back as a RTM.
 
ive had full electrolytic ( except a few tuner boards) recaps on 15 amps/receivers all 40-55 years old on 3 of them there was a night and day difference of the rest half pretty much sounded the same but had a lot more punch the other half sounded the same to me. None of them sounded worse. i dont know how many were bad but i dont trust 40 year old caps and i feel better having it done.
 
ok, what about an amp that received a total recap due to having poor sonics and the total recap not making much if any difference to the person who did the recap/owns the amp while expecting there to be a difference? the amp has single digit mv dc offset readings and runs cool and healthy it seems, but i cant help but feel it didnt sound like this when it was new. nothing wrong with it either as it doesnt sound out of the ordinary for its age compared to similar gear of the same age, i guess maybe i just expected much more from it based on its internet reputation..? maybe especially because of the total recap. its possible it did make a difference but it wasnt what i was expecting.. i was expecting it to be as sonically crisp as some 90’s gear i had for a bit that i attributed its crispness to its age (electrolytics in particular)...

im starting to think at the end of the day the quality of the amp and its design is the biggest contributing factor to how good an amp sounds no matter the age
 
Last edited:
im starting to think at the end of the day the quality of the amp and its design is the biggest contributing factor to how good and amp sounds no matter the age

I totally agree with this.

Did you replace any of the transistors in the unit in question? l would argue that sometimes new low noise transistors might have more impact on the sound than new caps (depending on original cap condition of course). I don't have data to back this up though and l also don't have the time to experiment on a unit whereby replacing e caps then testing, then going back in and replacing transistors etc and retesting.

Is your unit from the early to mid 70's? If so it likely won't have that crisp sound of your 90's units.
 
I totally agree with this.

Did you replace any of the transistors in the unit in question? l would argue that sometimes new low noise transistors might have more impact on the sound than new caps (depending on original cap condition of course). I don't have data to back this up though and l also don't have the time to experiment on a unit whereby replacing e caps then testing, then going back in and replacing transistors etc and retesting.

Is your unit from the early to mid 70's? If so it likely won't have that crisp sound of your 90's units.

yeah i believe it was first conceived and sold in 1976. dual mono with “batwing” blue and black nec outputs

its a shame too because the crispiness is all it seems to lack. its stayed with me longer than some of those crispy 90’s units mainly because it seems like it COULD be VERY good. it makes my speakers move more than many amps ive had rated twice the watts per channel it just doesnt do it with the precision necessary to really creep up your spine

i only replaced all e caps
 
Last edited:
ive gone through a lot of different gear from 70’s to late 90’s in the past 2 years and have ran them all head to head with whatever gear i have acquired to wittle down whats necessary and whats not. upon purchasing a piece of rotel gear from the late 90’s, i exploited my vintage gears age, but i had a hard time finding anything with a deeper/more extended frequency response from the mid-late 90’s that was affordable. i ended up snagging some early 90’s rotel separates thinking they’d have more low end than my smaller late 90’s integrated but the differences were negligible.

enter, the nec a-910 integrated amplifier. it smoked the rotel equipment (my mcs 3285 also smoked the rotel equipment but still isnt as “throaty” as id like). i bought an ra-985bx as one last benefit of the doubt for rotel and it too was slain with ease.

now is where it gets interesting and more in line with the general topic of age related deterioration. keep in mind the a-910 is from 1987-1990

enter, the 1987-1990 pioneer a-717. just wow. its bass is so fluid, natural, and reverberating. you can hear some of the most delicate sounds with utter precision. i would have to imagine its close to what people like about tube amplifiers. the a-910 barely scratches the surface of the low notes that the pioneer can just flood the room with from sheer woofer movement. the pioneer made two bose 201’s send heavier low frequency tremors through my floor than the nec can with both bose 201’s and mcs 3895 bass reflex speakers. its almost as if the nec is stationed above ground trying to lift the lower frequencies out whereas the pioneer was already underneath the ground pushing the low frequencies out from underneath themselves..

so does that mean the super healthy, rotel slaying, a-910 needs a recap? or does it mean the a-717 is just a higher fidelity amplifier than the 910 and MUCH more than the rotels and EXPONENTIALLY more than anything less than the rotels that led me on this quest to better conquer my existing gear??

the build quality of each amp is equal if not the nec even being of mostly higher quality components(short of the solid aluminum knobs, casted xformers, and honeycomb chassis) than the pioneer too for anyone that cares

i still listen to the nec more because it has a remote and doesnt seem as song dependent on eq setting changes, for anyone that cares
 
Last edited:
ive gone through a lot of different gear from 70’s to late 90’s in the past 2 years and have ran them all head to head with whatever gear i have acquired to wittle down whats necessary and whats not. upon purchasing a piece of rotel gear from the late 90’s, i exploited my vintage gears age, but i had a hard time finding anything with a deeper/more extended frequency response from the mid-late 90’s that was affordable. i ended up snagging some early 90’s rotel separates thinking they’d have more low end than my smaller late 90’s integrated but the differences were negligible.

enter, the nec a-910 integrated amplifier. it smoked the rotel equipment (my mcs 3285 also smoked the rotel equipment but still isnt as “throaty” as id like). i bought an ra-985bx as one last benefit of the doubt for rotel and it too was slain with ease.

now is where it gets interesting and more in line with the general topic of age related deterioration. keep in mind the a-910 is from 1987-1990

enter, the 1987-1990 pioneer a-717. just wow. its bass is so fluid, natural, and reverberating. you can hear some of the most delicate sounds with utter precision. i would have to imagine its close to what people like about tube amplifiers. the a-910 barely scratches the surface of the low notes that the pioneer can just flood the room with from sheer woofer movement. the pioneer made two bose 201’s send heavier low frequency tremors through my floor than the nec can with both bose 201’s and mcs 3895 bass reflex speakers. its almost as if the nec is stationed above ground trying to lift the lower frequencies out whereas the pioneer was already underneath the ground pushing the low frequencies out from underneath themselves..

so does that mean the super healthy, rotel slaying, a-910 needs a recap? or does it mean the a-717 is just a higher fidelity amplifier than the 910 and MUCH more than the rotels and EXPONENTIALLY more than anything less than the rotels that led me on this quest to better conquer my existing gear??

the build quality of each amp is equal if not the nec even being of mostly higher quality components(short of the solid aluminum knobs, casted xformers, and honeycomb chassis) than the pioneer too for anyone that cares

i still listen to the nec more because it has a remote and doesnt seem as song dependent on eq setting changes, for anyone that cares
As you have found, different units can behave and sound different than others. That Pioneer is a tough one to compare against others, especially due to the fact its frequency response is 1HZ to 150KHZ. Can your Speakers produce or your ears hear the 1HZ to 150KHZ??? Most speakers struggle to produce sound below 20hz. but with a low frequency response like the pioneer, I can imagine how it gets your floors shaking and windows rattling.
 
As you have found, different units can behave and sound different than others. That Pioneer is a tough one to compare against others, especially due to the fact its frequency response is 1HZ to 150KHZ. Can your Speakers produce or your ears hear the 1HZ to 150KHZ??? Most speakers struggle to produce sound below 20hz. but with a low frequency response like the pioneer, I can imagine how it gets your floors shaking and windows rattling.

while the pioneers frequency response makes sense vs the a-910 10hz-40khz, the super healthy single digit dc offset ra-985bx i briefly had boasted a frequency response of 4hz-100khz but it didnt have anywhere near the low end, mids, or highs that the a-910 has. the 985 was detailed but thin, bright, harsh, fatiguing, and simply had no bass extension nor mid bass warmth.

ive seen a lot of amps rated at the standard 20hz-20khz that i have to imagine will do exactly what that pioneer a-717 can. from now on i cant say i wont be thinking a bit higher of an amp for being rated as close to 1hz as possible though
 
Last edited:
Back
Top Bottom