advertisement


Recapping amps: facts, myths, legends etc

Tony L

Tony L

Administrator
Is there a consensus view here on pfm regarding recapping old equipment? I've seen so many conflicting views across the net from Naim who seem to think it needs doing every decade to Sugden and Quad who will only replace stuff that is off spec/causing as fault condition to the Japanese who seem to want to replace as little as possible even on things as old as Leak tube amps. I know Rob went from a 'recap everything' mindset to finding the vast majority of capacitors pulled from even early '70s kit was still in spec.

I'll give a couple of examples: I've very recently landed a very tidy Onix OA21S. It has to be one of the very last ones made as it has the same gold logo and speaker terminals as the OA401 and OA601 power amps that came out in 1990, so I guess it dates from then. It is allegedly low use. There is absolutely no evidence of bulging or leaking caps, nor of any servicing. I reset the bias (I googled it and found a thread on pfm telling me how to, I love when that happens!), it was a bit low, and DC offset is -15/-17mV which seems quite respectable. It works perfectly with no hum, crackles, switch on/off thump or any indication it is ill in any way, but it still has to be 26-27 years old! I'm tempted to leave it be as apparently Onix used very good long-lasting caps, but what is the consensus? I actually have another very much earlier OA21 that again appears to work fine, but I'll likely move this one on now as I don't need more than one!

I've also got a lovely 1973 Sony ST-5150 tuner that is the same, it just works perfectly with no hint of bulging or leaking caps and no hum. It looks like new inside! I have to admit I'd only do this one if it went wrong as it is less risk than an amp, it can't really hurt anything else downstream. I've got the matching amp too, again in lovely condition, but again I've been too scared to use it with valuable vintage speakers (I have no junk speakers!).
 
It depends on the type of cap and the storage conditions as well as age. Electrolytic caps can dry out and change spec as well as leaking or bulging so you can't always tell visually.

Some 50s cap types are notorious for aging badly, but many types will last a long, long time.

So it's a judgement call really.
 
It would be good to see some old (10 year) removed caps from naim gear characterised, capacitance would almost certainly read good if the cap is physically intact, even after decades, but ESR, leakiness (DC, not electrolyte), etc might yield something.

My experience is that decent caps operated comfortably within their specs in a non-SMPS role can last for ages, even many decades old. It surprised me.
 
old electrolytic do have the problem of leaking ,the very small value one if no issues leave them . disturbing can have adding more issues in my experience
 
Andrew.B said:
It would be good to see some old (10 year) removed caps from naim gear characterised, capacitance would almost certainly read good if the cap is physically intact, even after decades, but ESR, leakiness (DC, not electrolyte), etc might yield something.

My experience is that decent caps operated comfortably within their specs in a non-SMPS role can last for ages, even many decades old. It surprised me.
Click to expand...

Other than bulges, leaks etc what are the symptoms to look for? Am I right in thinking hum is the best subjective hint that something is amiss?

Naim are an odd case as they expect the kit to be left powered up, I guess capacitor shelf-life aside (I realise this is a factor) ten Naim years equate to three, four or five decades of normal use from folk who turn kit off when not in use. Obviously something like an Onix may have been subjected to Naim always-on ideology, but if mine is as low use as it looks it may well have sat in its box for a good proportion of its life!
 
Tony L said:
Is there a consensus view here on pfm regarding recapping old equipment?
Click to expand...

I had my Naim cd3.5 & Nait 3 serviced a little while ago, completely recapped, both were working perfectly prior to this & as in your case, all looked fine,
I fell for the myth unfortunately, cost me a huge amount too, when they returned, the sound lacked that Naim quality I enjoy, I sold them on with the full knowledge of what had been carried out, I have since purchased both again, no service, working fine, back to that Naim quality many enjoy, never again would I do this, only if something is not working would I consider a service recap.

I would be cautious Tony, if you enjoy the sound & all is working fine, I would suggest to leave well alone.
 
Tony L said:
Other than bulges, leaks etc what are the symptoms to look for? Am I right in thinking hum is the best subjective hint that something is amiss?
Click to expand...

Yes, hum is an indication of poor power supply caps, and that is where you find the large electrolytics.
 
I have just recapped a Yamaha C4 preamp (all bloody 80 of them). The unit is around 40 years old and still had its original caps, non of them showed any signs of bulging or leaking. There was also no audiable signs of the caps deteriorating, in fact in sounded great.

However the thought of ancient electrolytics was beginning to bug me so I hooked them all out and replaced them with Nichicon Fine Golds and Muse bipolars, reset the rail voltage and offset.

And the difference?

Not a lot, the Fine Golds make a small difference in the sonic signature, have a slight lift in the lower mids. But all other aspects of the sound remain the same.

So with hindsight it didn't need doing but hopefully it will be good for another 40 years. Now I have the dilemma of what to do with its brother, the M4 power amp, stick or twist?.
 
There's a huge difference in rated lifetime between some PCB mount snap in capacitors @ 2000 hours and high quality screw terminal types @ 200,000 hours.

Also although a capacitor can work for decades, past a certain point there is a degradation in sound quality, so if you recap, say a NAP250 at 10 year intervals, you will certainly hear the (usually huge) improvement.

Having said that I managed to reform some 1947 vintage electrolytic capacitors (about as old as they come) in a pair of Brook 12A 2A3 amplifiers (USA). Although I chose to use new capacitors in them while I owned them.

Tantalum capacitors don't have a wear out mechanism, but they do sound "tired" after some years. There are also large differences between small can radial capacitors in their expected lifetime.
 
Quality of the capacitor and the usage conditions are key.
Agree with Julf, heat is the killer both from conditions within the case of audio equipment and within the cap caused by high ripple current

Some years ago I moved from a position of shotgun recapping to selective recapping after testing.

As an example, when I recapped a small 1972 vintage Pioneer amplifier recently all of the 20 or so ELNA and Nippon Chemicon electrolytic caps tested good - not just good but better than new in some cases. The Pioneer runs cool and no caps are run to within a few volts of maximum rating.

OTOH I've seen countless 10-20 year old units with visibly leaking caps, usually from lesser known brands. The common use of SMPS has made this situation worse as these can place tougher endurance demands on caps.

So my advice is to check for visible signs of distress, listen for signs of degradation (hum, poor response) and spot check with an ESR meter.

Note there are some known bad caps - those which routinely degrade and fail prematurely and if a unit contains those you might want to swap them as a precaution. Small value (<10uf) Sanyo sky blue caps coming to mind.

Another note: Not all bad caps are bad :)
Consider for example an inter-stage coupling cap of say 100uf. A 10uf cap would likely still perform the same function and have insignificant effect on the roll-off frequency. So if your 100uf cap drifts in value a little (which they tend to do), you will see no effect on performance because the precise capacitance value is not critical within wide limits and a moderately high ESR is of no consequence in a high impedance circuit. OTOH you'd probably not want to use this capacitor in a power supply...
 
I was convinced my Rogers A100 amp must need a recap even though it sonded great but there was an occasional crackle
I stripped it down and could see no signs of leaking caps
I removed the 4 large electrolytics in the power supply but they all tested great and low ESR so I put them back in
I lubed all pots and switches whilst it was appart and the crackle has disappeared and it sound even better.
On the other hand I had a Cambridge AV amp witch suffered for quite a few leaky caps which was only about 5 years old

None of my DIY amps have needed any caps yet but I do always use decent quality caps where ever I can and where appropriate of course.

on my Reel 2 Reel players there are certain caps that have known to leak dry out and worst of all fail short
On the Studer and Revoxes its the gold Frako caps which were great caps in their day but can take out ICs when they go. The little blue phillips are also unreliable so I change all of them. The Rifa X2 film caps crack and fail sometimes short and smoke an sting the place out and sometimes open so they ineffective
Solid tants seem to go very high ESR so I allways change them as well sometimes for electrolytic's. The orange solid aluminium caps and wima film caps never seem to fail
On the Panasonic RS1500 tape decks the purple elecrolytics age similar to the Frakos so best to change them all out now these are lovely tape decks but I do worry about if the caps fail so I have bought a full set of them, just need to find time to do the work

There is always arguments to only change what fails or to recap some say don't worry if its working leave well alone and some say ticking time bo*b

Alan
 
I have to admit the apparent low-quality/low lifespan of many modern electrolytic caps is another reason for leaving well alone! So many stories of kit with nasty Chinese caps that swell up within a few years.

One thing I’ve taken to doing now I’ve got more kit floating around than systems to use it in is making sure I power everything ‘spare’ up at least once a year for a day or so with the view of reforming caps etc. I’m aware caps have a finite shelf life just sitting unused and get the (maybe false) impression this may lead to longer life.
 
I have recaped several 250s hicaps my Nait1 twice the Rotel in the garage system all with massive improvements in the sound, once you know how an amp that needs servicing sounds it’s very easy to spot.
So I am a firm believer in servicing.

Pete
 
In general I'll leave well alone if working. I've had a few components that have had every electrolytic replaced in the hope that they'll run trouble free for the next few decades. In general though I've had more issues with modern equipment, though I certainly don't think modern caps from the known brands are to be avoided - just the cheap Chinese stuff. I don't own any Naim gear and think a 10 year service is excessive. My Japanese equipment (marantz, Nakamichi and Technics) is filled with high-quality components including branded capacitors and I fully expect it to operate for decades.
 
37298021670_f822131b22_b.jpg


FWIW here's a pic (taken by the previous owner) of the OA21S I mentioned. As I understand it the big blue caps PSU were made specially for Onix by BHC and are rather hard to replace like for like. As I say I can see or hear no issues and whilst the DC offset is a little higher than I'd ideally like (-16mV/-18mV) it is still within perfectly acceptable levels as I understand it. I reset the bias as per instructions in this thread and it really does sound good!
 
From what I can gather, a capacitor may physically leak it's electrolyte, but may also, over time gradually start to leak DC, which I suppose is what causes the hum. A capacitor as it ages may begin to increase in resistance; possibly remaining within spec when measured for capacitance, whilst altering the circuit as it gains in unwanted resistance.

I wonder if it's worth desoldering one side of a couple of capacitors and measuring the resistance across it. If you chose the ones likely to be most exposed to heat, then you'd have the worst case example; so if they're fine then I'd imagine the others would be too.

understand, I'm perfectly happy to be corrected on this, I'm far from an expert. This is merely the hypothesis of a novice.

I should also add that for a bit of kit that looks so nice and tidy inside, sounds good and has been successfully re-biased; I, personally would leave well alone.
 
The problem with this kind of amp (and many others) is most of the effort is in disassembling the thing, and if you get that far its quicker just to do it and replace the lot. The price of the caps is peanuts in the grand scheme of things, the issue is the considerable faff and the risk of doing damage to something (in this case the very fragile and irreplaceable input selector switch tops the list). It is just not designed for quick field service the way say a Quad 303 or whatever is!
 
it's nice when undertaking this kind of thing do to just a couple at a time, then have a quick listen to make sure an attempted fix hasn't made things worse. If you do the whole thing at once then something doesn't sound right, it's hard to know what specifically is causing the problem. If, like you say it's some effort to take to bits and get back together again then that would be another reason for me not to do anything with it... add to that irreplaceable parts, and that would confirm things for me.
 
You must log in or register to reply here.


advertisement


Back
Top