You should only use a music signal to break them in. If you use 60 Hz they'll create hum ever after. As the twig is bent, so is the tree inclined...
Well,
now it gets *really* esoteric
Then, if I understand you right, If I break them in with pure black metal, my amp will never be capable of playing classical music right because of its metal-poisoned coupling caps?
I think I need a beer.
Cheers,
Andreas
PS. I like this forum, but if I would take everything written here seriously, I would need a therapist soon...
Should be fine for use in Europe, then. Aah... The benefits of 50 Hz mains.
Then, if I understand you right, If I break them in with pure black metal, my amp will never be capable of playing classical music right because of its metal-poisoned coupling caps?
Yeah. Tough sh*t if you have a varied taste in music and like to listen to black death metal one day and the Brandenburg Concertos the next. Cryin' shame, I tell ya...
Send one my way, would you?!
~Tom
Send one my way, would you?!
~Tom
On its way! But remember to give it a break-in time before consuming!
Rundmaus
Great little experiment for the diyAudio crew if they feel so inclined. Pick an amp circuit that is relatively simple (low parts count, tube of course) that meets the criteria of being worthy enough of 'seasoned' capacitors. Pick a set of capacitors that 'age' well. Select and match two sets of capacitors. Build the circuit using a two position multi-pole switch so that you could switch between one set of capacitors that have been 'run in' for hundreds of hours, and another set on the other position that retains the capacitors voltage virginity until the test is ready to be carried out.
Do a blind listening test with the burnt in set and then the factory fresh set. Just for giggles do it again after giving the fresh set 10 hours of their own burn in time. Repeat the blind test.
Wonder what the outcome would be.
Do a blind listening test with the burnt in set and then the factory fresh set. Just for giggles do it again after giving the fresh set 10 hours of their own burn in time. Repeat the blind test.
Wonder what the outcome would be.
How about two identical amplifiers, working into the same load, one that has run for 1000+ hours, and one "hot off the presses" so to speak. Measure distortion at a set band of frequencies, and chart the tables up so we can see? I personally find it a little hard to believe anything other than old electrolytics would benefit from a break-in period myself, but the ritual of breaking in an amp is well practiced in this household
Last edited:
So let me get this straight........A Black Gate takes a thousand hours to "break in", lets assume it's used in the PS.
I've got a new product to sell ya!! I run an RF signal into your precious Black Gate & break it in in Six milliseconds!
________________________________________________Rick............
I've got a new product to sell ya!! I run an RF signal into your precious Black Gate & break it in in Six milliseconds!
________________________________________________Rick............
your expectations was....?
Your original q had a number of statements made as fact when they are actually strongly held opinion.
Given this, it is difficult to suggest a process especially when the parameters (what caps? Used where in the circuit?) were not defined.
Here's a clue though. This is post 29 and no-one has yet stated their method for breaking in a capacitor - even those who seem to be believers have given at best vague outlines of what may be required....
Your original q had a number of statements made as fact when they are actually strongly held opinion.
Given this, it is difficult to suggest a process especially when the parameters (what caps? Used where in the circuit?) were not defined.
Here's a clue though. This is post 29 and no-one has yet stated their method for breaking in a capacitor - even those who seem to be believers have given at best vague outlines of what may be required....
Doanworrybouddit. This business of "break in" is pure audiophoolery. You have the Blackgates, use 'em; be happy.
Any time I see claims like this, I grab my wallet and hold on tight. The time for some overpriced bit of folderol is that time it takes you to get used to the new (and likely inferior) sound.
The only component that may need some amount of break-in are new VTs. When these go into service, it might take a few hours for the bias to settle down. It may take awhile for new speeks to loosen up, but that's about it.
Are we talking just power supply filtering electrolytics here? Or is there some connection with using electrolytics as coupling caps - of which the only connection I know of is for cathode bias bypass usage
Knowing the application would provide a bound on the scientific assessment being hauled in to this topic.
Electrolytics inherently have huge tolerances in all their equivalent model parameters, and that tolerance extends to ageing (failure as a result of out-of-tolerance spec). So the natural variation with time of those parameters is bound to be large, and is accentuated by their naturally short age to get to a failure condition. Note that the 'large' and 'huge' descriptions relate to all other amplifier parts (excluding valves and speakers) that people tend to want to specify down to sub 1% tolerance over life.
That said, electrolytic capacitors have more analgous behaviour to a battery than say an NPO or silvered mica. They do inherently have time dependant characteristics, both at very short ms level, and at hours level. I would not disconnect the observed change in audible behaviour - because it is somewhat similar to peoples perceptions of tubes warming up and ageing, and speaker suspension systems 'bedding in' and materials ageing over years and with temperature. But I think these all relate to distortion mechanisms, and whether someone likes a particular distortion mechanism over another, rather than the hi-fi drive to eliminate distortions mechanisms (ie. don't use black gates, just use a bucket of poly).
Electrolytics have had a substantial level of scientific investigation, and that is still continuing due to their use in switchmodes and equipment reliability. There are many IEEE examples to peruse if you have access to that collection.
Ciao, Tim
Knowing the application would provide a bound on the scientific assessment being hauled in to this topic.
Electrolytics inherently have huge tolerances in all their equivalent model parameters, and that tolerance extends to ageing (failure as a result of out-of-tolerance spec). So the natural variation with time of those parameters is bound to be large, and is accentuated by their naturally short age to get to a failure condition. Note that the 'large' and 'huge' descriptions relate to all other amplifier parts (excluding valves and speakers) that people tend to want to specify down to sub 1% tolerance over life
.That said, electrolytic capacitors have more analgous behaviour to a battery than say an NPO or silvered mica. They do inherently have time dependant characteristics, both at very short ms level, and at hours level. I would not disconnect the observed change in audible behaviour - because it is somewhat similar to peoples perceptions of tubes warming up and ageing, and speaker suspension systems 'bedding in' and materials ageing over years and with temperature. But I think these all relate to distortion mechanisms, and whether someone likes a particular distortion mechanism over another, rather than the hi-fi drive to eliminate distortions mechanisms (ie. don't use black gates, just use a bucket of poly).
Electrolytics have had a substantial level of scientific investigation, and that is still continuing due to their use in switchmodes and equipment reliability. There are many IEEE examples to peruse if you have access to that collection.
Ciao, Tim
This is not exactly what you are looking for, but are you familiar with "reforming" electrolytic capacitors?
Phrarod, can I throw in an olive branch to continue this thread. I sit on the scientific side, and can appreciate that those without such training will state certain things as a 'given', whereas that is a red rag to a bull for someone on the other side of the fence.
From an engineering perspective, if a cap in a cathode bias location was varying then I'd put in another type - and if it had lower parasitics and distortion mechanisms then great. Eg. lots of parallel little 1uF 50V metalised poly caps. But I do guitar amps more than hi-fi, so if someone enjoys a particular distorting component or mechanism then heh it has its place.
If you wanted to 'age' an electrolytic then I think it is valid to excercise it in the same manner as its usage. There may be a time advantage to trying to put in place an accelerated test mechanism, but that opens the test up to producing other non-intended affects which may be difficult to appreciate.
My initial impression would be that the aging comes about from application of the range of bias voltage levels seen in the application, and the passing of current levels as seen in the application. If they were deemed to be 'large-scale' effects, then I can't see the benefit per se in adding small scale effects. Ie. the test circuit could be as simple as a full-wave rectified mains signal, with peak voltage equal to the peak level likely in the cathode bias application, and a loading resistance to discharge the capacitor between pulses to a level of say idle (eg. as in a PP cathode bias stage).
Does that sound reasonable (pun intended).
Ciao, Tim
From an engineering perspective, if a cap in a cathode bias location was varying then I'd put in another type - and if it had lower parasitics and distortion mechanisms then great. Eg. lots of parallel little 1uF 50V metalised poly caps. But I do guitar amps more than hi-fi, so if someone enjoys a particular distorting component or mechanism then heh it has its place.
If you wanted to 'age' an electrolytic then I think it is valid to excercise it in the same manner as its usage. There may be a time advantage to trying to put in place an accelerated test mechanism, but that opens the test up to producing other non-intended affects which may be difficult to appreciate.
My initial impression would be that the aging comes about from application of the range of bias voltage levels seen in the application, and the passing of current levels as seen in the application. If they were deemed to be 'large-scale' effects, then I can't see the benefit per se in adding small scale effects. Ie. the test circuit could be as simple as a full-wave rectified mains signal, with peak voltage equal to the peak level likely in the cathode bias application, and a loading resistance to discharge the capacitor between pulses to a level of say idle (eg. as in a PP cathode bias stage).
Does that sound reasonable (pun intended).
Ciao, Tim
Just for interest here is someones opinion.
Each person can have their own thoughts, however I think it presents what the OP is thinking. Please correct me if I am mistaken!
SoundStage! Network Ultra Audio -- Archived Article
Regards
M. Gregg
An electrolytic used as a cathode bypass cap probably will 'burn in'. All you need to do is apply the expected bias voltage. There is a simple mechanism for this.
In any electrolytic there is a permanent battle between the oxide layer (which is the dielectric) degrading due to age, and being reformed by electrolytic action (this is the useful outcome of the leakage current). The oxide layer will tend to reach equilibrium when it is thick enough to prevent very much current from flowing. When you build an amp you will normally use caps with a voltage rating somewhat higher than the expected operating voltage. The oxide layer thickness will actually correspond to a slightly higher voltage than this, to allow some leeway and some ageing while the cap sits on a shelf somewhere.
Once in the circuit, the oxide layer will gradually adjust itself to the applied voltage. In most cases this will mean it gets a bit thinner. It is known that if an electrolytic has been used at a particular voltage for a long period (well within its stated rating) then you can't simply apply the rated voltage without first reforming to the higher voltage again. DIYers re-using components from old projects may need to be aware of this. However, in many cases this reforming happens automatically in the new circuit (especially with small lower voltage caps).
A thinner oxide should mean a higher capacitance (I don't know if anyone has measured this). So a bypass electrolytic may gradually increase its value over the first few hundred hours of operation. Now here is the bit people might not like: in a well-engineered circuit you will not notice the change as the bypass cap will not set the LF rolloff, but a poor circuit may make this change apparent. So, as I have said before, a circuit which is 'discriminatory' is actually a poor circuit. If you notice electrolytic burn-in, then your circuit needs redesigning - at the very least you probably need bigger electrolytics or smaller coupling caps.
In any electrolytic there is a permanent battle between the oxide layer (which is the dielectric) degrading due to age, and being reformed by electrolytic action (this is the useful outcome of the leakage current). The oxide layer will tend to reach equilibrium when it is thick enough to prevent very much current from flowing. When you build an amp you will normally use caps with a voltage rating somewhat higher than the expected operating voltage. The oxide layer thickness will actually correspond to a slightly higher voltage than this, to allow some leeway and some ageing while the cap sits on a shelf somewhere.
Once in the circuit, the oxide layer will gradually adjust itself to the applied voltage. In most cases this will mean it gets a bit thinner. It is known that if an electrolytic has been used at a particular voltage for a long period (well within its stated rating) then you can't simply apply the rated voltage without first reforming to the higher voltage again. DIYers re-using components from old projects may need to be aware of this. However, in many cases this reforming happens automatically in the new circuit (especially with small lower voltage caps).
A thinner oxide should mean a higher capacitance (I don't know if anyone has measured this). So a bypass electrolytic may gradually increase its value over the first few hundred hours of operation. Now here is the bit people might not like: in a well-engineered circuit you will not notice the change as the bypass cap will not set the LF rolloff, but a poor circuit may make this change apparent. So, as I have said before, a circuit which is 'discriminatory' is actually a poor circuit. If you notice electrolytic burn-in, then your circuit needs redesigning - at the very least you probably need bigger electrolytics or smaller coupling caps.
Andreas, just for the record, I was being sarcastic!
Kenneth
Noticed
Then there's hope for this forum
The papers I've seen indicate that the the CV product effectively remains constant in the first few 1000 hrs operation, and that the adjustment of dielectric thickness to suit the applied voltage caused a slight decrease in C if the applied voltage is higher (ie. counter to DF96's proposition). However the technical papers are all about identifying the change in C and ESR over service life, so we are talking about tests conducted for many 1000's hrs and typically at quite high temperatures approaching the specified 105C of the caps.
This I think is different from any very early change in parameters, say in the first 100's of hrs of operation, possibly as a lack of manufacturer time spent on forming the dielectric, or possibly because the chemistry of that particular capacitor causes quite a change in parameters when operated at a different condition than that during manufacturer forming.
Wrt the parameters that may change, I doubt that a varying leakage current is an issue in a cathode bias circuit unless the level of leakage is gross compared to a normal electrolytic, and this would probably be related to application of a relatively high bias voltage. I doubt the level of a varying ESR is an issue, unless the ESR pole is in the audio range. I guess that leaves capacitance variation, and any time-delay electrochemical process that probably impacts effective capacitance. But then the influence of capacitance per se (LF roll off) becomes the focus, and we get back to DF96's last point .
If it is a quirk of the capacitor type being used, then applying a bias voltage profile similar to the application seems the only valid approach to take for burning in. Phrarod may be able to identify that profile for us - is it a preamp cathode bypass, or a more onerous PP stage application? Is the normally applied voltage substantially lower than the voltage rating of the capacitor?
This I think is different from any very early change in parameters, say in the first 100's of hrs of operation, possibly as a lack of manufacturer time spent on forming the dielectric, or possibly because the chemistry of that particular capacitor causes quite a change in parameters when operated at a different condition than that during manufacturer forming.
Wrt the parameters that may change, I doubt that a varying leakage current is an issue in a cathode bias circuit unless the level of leakage is gross compared to a normal electrolytic, and this would probably be related to application of a relatively high bias voltage. I doubt the level of a varying ESR is an issue, unless the ESR pole is in the audio range. I guess that leaves capacitance variation, and any time-delay electrochemical process that probably impacts effective capacitance. But then the influence of capacitance per se (LF roll off) becomes the focus, and we get back to DF96's last point .
If it is a quirk of the capacitor type being used, then applying a bias voltage profile similar to the application seems the only valid approach to take for burning in. Phrarod may be able to identify that profile for us - is it a preamp cathode bypass, or a more onerous PP stage application? Is the normally applied voltage substantially lower than the voltage rating of the capacitor?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.