Fuses.
Fuses, being cheap, are virtully mandatory with commercial equipment for obvious and sensible reasons. No manufacturer wishes to deal with problems frequently caused by careless users who do stupid things like 'hot-swapping' cables which they have been warned against in the instructions, but which few read until after the event.
Customers' safety is paramount, together with the safety of other attached equipment. It is not easy to dismiss claims that "Out of the blue, your power amp just went bang and it has wiped-out my expensive speakers, so what are you going to do about this?", and in this 'compensation climate' nowadays especially, not everyone is honest. I am ashamed to hear of people who will deliberately spill red wine on a carpet to claim for a new one under the Insurance cover, but it happens!
However, with some thought it is possible to minimise fuses and retain safety margins, and for DIY, commercial pressures don't apply. This doesn't mean that I advocate recklessness in any such case, but it is also important to consider where any dangers lie from a DIY'ers viewpoint.
To be practical about this, I am told that there has never been an instance in the UK (according to Fire Service records) where a significant fire has been the result of failure of domestic audio equipment in use, and there is (usually) little within the normal metal enclosure which is very combustible.
Even if this is not entirely correct, in my many years of involvement I have never come across any such case, and there are many other far more likely perils to consider which we accept as being commonplace.
For DIY'ers it shouldn't be difficult to carry out any repairs arising from slightly less fastidious protection devices, and I would expect you to treat your equipment with respect, hopefully avoiding unnecessary accidents.
In the hope of avoiding any nit-picking 'aftermaths', when using expressions like "sounds better" I mean that there is apparently less degradation to the signal, and it is this degradation which spoils the tiny nuances we can otherwise hear and which take one closer to the original performance, making it more believable. This can happen by various means, possibly by unwantedly filtering-out these very low-level sounds or perhaps by distorting them, but whatever causes these losses, the result is a less natural and enjoyable sound.
I don't believe that with analogue audio it is ever possible to improve on the original signal generated by the 'front-end' transducer, although it is possible to manipulate this in certain ways, most of which will probably not be heard as an overall improvement.
Therefore, the best available cartridge (or laser assembly) should be used to get *all* of the recorded information from the media, and the trick then is to preserve as much of this as possible throughout the remaining chain to the speakers or headphones. When this is achieved, the results are quite stunning, but it does require a lot of care in getting to this stage, and every step counts as I have tried to suggest earlier.
One poor choice anywhere in this chain will bring down and restrict the final results, please believe me, and dealing with fuses as an issue is merely one of many areas which should be given attention if the finest results are desired.
My particular interest in fuses began in the '70s when as a close friend of the late UK audio Guru JLH, he sometimes sent me new circuits for a second opinion on their sonic attributes before publishing the designs, as he seemed to value my hearing abilities.
Around this time JLH was excited about a forthcoming power amp design of his and he been collaborating with a kit-supplier, not for any financial return, but to avoid the usual plethora of follow-ups to his articles which were difficult and very time-consuming to deal with, along the lines of "Where can I obtain part X from" etc.
I was duly provided with a prototype kit of parts which I built and listened to, and it didn't sound bad at all. However there were some 12 fuses in the chain between the wall plug and the speaker outputs, which seemed a bit OTT to me. One (mandatory in the UK) in the wall plug, one at the AC input, one at each output, and some (8 in all) on the PCBs to cover the dual PS rails - both high current and low current, and which were separated.
JLH explained the reasoning of course (along the lines mentioned above), but he readily accepted that this 'belt & braces' approach was a bit over-cautious, and partly with his encouragement I started some experiments with some of these fuses, to satisfy my curiosity over their potential effects on the sound.
Where time-lag fuses were used, I tried about 4 different makes and types of construction, and all 'sounded' slightly different to me when listening. I did also try some different makes of quick-blows where these were used, but as they are necessarily virtually identical in their method of construction (i.e. a single wire in a glass tube) I couldn't repeatedly and reliably tell them apart IIRC.
I aso tried bypassing them with film caps out of curiosity, and although this improved the results a little when used in AC locations like the mains input fuse, when bypassing the DC rail fuses it didn't have any noticeable affect, much as I had anticipated.
Wire-ended fuses were not so popular in those days, and being suspicious of the effects of the springy (magnetic-which I don't care for) open-frame fuse holders on the PCBs, I soldered wires directly from the adjacent pads to the ends of the fuses, and this was a clear step in the right direction. I had previously found problems with other magnetic components and especially clip or push-on type contacts, and had already affected some slight improvements when I soldered the PS wires to the PCBs, to replace the push-on spade contacts supplied in this kit. I would therefore go along with John's suggestion of using wire-ended fuses if you cannot do without them at all, as after all it doesn't take very long to solder a couple of fuse ends if they ever blow in use, and it will usually take a lot longer opening up the enclosure to get at even clip-in fuses should they blow. Your time is free, unlike a manufacturer or dealer who would be presented with another unwanted 'overhead' if soldered fuses are installed.
After several months of these interesting trials, I noticed that the PCB fuses had taken on a quite different appearance which could be readily seen through the glass as these were Q-B fuses in this location. Under slight magnification, the fuse wires had deteriorated from a shiny smooth surface to a very rough dull look, so I broke a couple out and had a look under a microscope, only to be appalled at what was apparent. Obvously these wires which are flexible, soft and malleable when new (I broke some new ones to check) had completely changed to being brittle and fragile, with a surface like craters on the moon, and deep transverse fissures all along their length! What this did to the sound was suddenly more easy to understand, and replacing all 8 of these fuses with new ones did effect a very noticeable change to the results when listening. I also noticed at powering-up, that these fine fuse-wires actively 'wriggled' with the thermal stress, and this being the case will have work-hardened to become very brittle as all (non-ferrous) metals do when subjected to bending or any other deformation.
In this connection many tests which I have carried-out (mostly subsequent to this fuse issue) have convinced me without any shadow of doubt that hardened metals always have a worse affect on good sonics than when the same metal parts are annealed. I won't discuss this any further now, but although I personally haven't yet made any direct comparisons between Cryo'd parts/wires and the same items not Cryo'd, this process most likely does anneal the metals involved. I have carried out some experiments with heat annealing (and I guess that the method of annealing is probably not important, so long as the result is the same) and universally this improves sonics, for example drawn wires, which if they remain hard from the drawing process, don't sound so good.
Where any 'unsupported wire' Q-B type of fuses go (I cannot personally vouch for time-lag fuses which often have a central core or a supporting spring arrangement, but I wouldn't take any chances here if I still used fuses) I strongly recommend that these be replaced frequently, whether they have blown or not.
Indeed, whilst in conversation recently with the manufacturer of some audio equipment I had been asked to repair, he advised me to change the internal fuses for this very purpose of 'restoring' the sonic performance, while it was open on the bench.
By that time it was quite clear to me that fuses had to go as these effects were not acceptable to me, so I 'hard-wired' out all these 12 fuses temporarily, and the result was so clearly an improvement that this could have been another amp design altogether. JLH heard this in due course, and was rather surprised himself, but I was aware of the risks involved in not having any protection at all, and returned to a safer situation whilst I investigated alternative methods of achieving this.
Already this post is longer than I had hoped for, so I will start again soon with more details of what I found to be an acceptable way of resolving this problem, and recommend some suitable thermal (not magnetic!) circuit breakers which I have used universally for many years since, and found to be very satisfactory from their sonic aspects.
Lastly, for any prospective 'wit' who has become bored with this posting, just remember it has taken me a lot longer to write this than for you to read it, and I am not doing it for my own benefit as I already know these answers. I am simply trying in the ways I know best (as an old man) to help others who might be interested. 🙂
Regards,
Fuses, being cheap, are virtully mandatory with commercial equipment for obvious and sensible reasons. No manufacturer wishes to deal with problems frequently caused by careless users who do stupid things like 'hot-swapping' cables which they have been warned against in the instructions, but which few read until after the event.
Customers' safety is paramount, together with the safety of other attached equipment. It is not easy to dismiss claims that "Out of the blue, your power amp just went bang and it has wiped-out my expensive speakers, so what are you going to do about this?", and in this 'compensation climate' nowadays especially, not everyone is honest. I am ashamed to hear of people who will deliberately spill red wine on a carpet to claim for a new one under the Insurance cover, but it happens!
However, with some thought it is possible to minimise fuses and retain safety margins, and for DIY, commercial pressures don't apply. This doesn't mean that I advocate recklessness in any such case, but it is also important to consider where any dangers lie from a DIY'ers viewpoint.
To be practical about this, I am told that there has never been an instance in the UK (according to Fire Service records) where a significant fire has been the result of failure of domestic audio equipment in use, and there is (usually) little within the normal metal enclosure which is very combustible.
Even if this is not entirely correct, in my many years of involvement I have never come across any such case, and there are many other far more likely perils to consider which we accept as being commonplace.
For DIY'ers it shouldn't be difficult to carry out any repairs arising from slightly less fastidious protection devices, and I would expect you to treat your equipment with respect, hopefully avoiding unnecessary accidents.
In the hope of avoiding any nit-picking 'aftermaths', when using expressions like "sounds better" I mean that there is apparently less degradation to the signal, and it is this degradation which spoils the tiny nuances we can otherwise hear and which take one closer to the original performance, making it more believable. This can happen by various means, possibly by unwantedly filtering-out these very low-level sounds or perhaps by distorting them, but whatever causes these losses, the result is a less natural and enjoyable sound.
I don't believe that with analogue audio it is ever possible to improve on the original signal generated by the 'front-end' transducer, although it is possible to manipulate this in certain ways, most of which will probably not be heard as an overall improvement.
Therefore, the best available cartridge (or laser assembly) should be used to get *all* of the recorded information from the media, and the trick then is to preserve as much of this as possible throughout the remaining chain to the speakers or headphones. When this is achieved, the results are quite stunning, but it does require a lot of care in getting to this stage, and every step counts as I have tried to suggest earlier.
One poor choice anywhere in this chain will bring down and restrict the final results, please believe me, and dealing with fuses as an issue is merely one of many areas which should be given attention if the finest results are desired.
My particular interest in fuses began in the '70s when as a close friend of the late UK audio Guru JLH, he sometimes sent me new circuits for a second opinion on their sonic attributes before publishing the designs, as he seemed to value my hearing abilities.
Around this time JLH was excited about a forthcoming power amp design of his and he been collaborating with a kit-supplier, not for any financial return, but to avoid the usual plethora of follow-ups to his articles which were difficult and very time-consuming to deal with, along the lines of "Where can I obtain part X from" etc.
I was duly provided with a prototype kit of parts which I built and listened to, and it didn't sound bad at all. However there were some 12 fuses in the chain between the wall plug and the speaker outputs, which seemed a bit OTT to me. One (mandatory in the UK) in the wall plug, one at the AC input, one at each output, and some (8 in all) on the PCBs to cover the dual PS rails - both high current and low current, and which were separated.
JLH explained the reasoning of course (along the lines mentioned above), but he readily accepted that this 'belt & braces' approach was a bit over-cautious, and partly with his encouragement I started some experiments with some of these fuses, to satisfy my curiosity over their potential effects on the sound.
Where time-lag fuses were used, I tried about 4 different makes and types of construction, and all 'sounded' slightly different to me when listening. I did also try some different makes of quick-blows where these were used, but as they are necessarily virtually identical in their method of construction (i.e. a single wire in a glass tube) I couldn't repeatedly and reliably tell them apart IIRC.
I aso tried bypassing them with film caps out of curiosity, and although this improved the results a little when used in AC locations like the mains input fuse, when bypassing the DC rail fuses it didn't have any noticeable affect, much as I had anticipated.
Wire-ended fuses were not so popular in those days, and being suspicious of the effects of the springy (magnetic-which I don't care for) open-frame fuse holders on the PCBs, I soldered wires directly from the adjacent pads to the ends of the fuses, and this was a clear step in the right direction. I had previously found problems with other magnetic components and especially clip or push-on type contacts, and had already affected some slight improvements when I soldered the PS wires to the PCBs, to replace the push-on spade contacts supplied in this kit. I would therefore go along with John's suggestion of using wire-ended fuses if you cannot do without them at all, as after all it doesn't take very long to solder a couple of fuse ends if they ever blow in use, and it will usually take a lot longer opening up the enclosure to get at even clip-in fuses should they blow. Your time is free, unlike a manufacturer or dealer who would be presented with another unwanted 'overhead' if soldered fuses are installed.
After several months of these interesting trials, I noticed that the PCB fuses had taken on a quite different appearance which could be readily seen through the glass as these were Q-B fuses in this location. Under slight magnification, the fuse wires had deteriorated from a shiny smooth surface to a very rough dull look, so I broke a couple out and had a look under a microscope, only to be appalled at what was apparent. Obvously these wires which are flexible, soft and malleable when new (I broke some new ones to check) had completely changed to being brittle and fragile, with a surface like craters on the moon, and deep transverse fissures all along their length! What this did to the sound was suddenly more easy to understand, and replacing all 8 of these fuses with new ones did effect a very noticeable change to the results when listening. I also noticed at powering-up, that these fine fuse-wires actively 'wriggled' with the thermal stress, and this being the case will have work-hardened to become very brittle as all (non-ferrous) metals do when subjected to bending or any other deformation.
In this connection many tests which I have carried-out (mostly subsequent to this fuse issue) have convinced me without any shadow of doubt that hardened metals always have a worse affect on good sonics than when the same metal parts are annealed. I won't discuss this any further now, but although I personally haven't yet made any direct comparisons between Cryo'd parts/wires and the same items not Cryo'd, this process most likely does anneal the metals involved. I have carried out some experiments with heat annealing (and I guess that the method of annealing is probably not important, so long as the result is the same) and universally this improves sonics, for example drawn wires, which if they remain hard from the drawing process, don't sound so good.
Where any 'unsupported wire' Q-B type of fuses go (I cannot personally vouch for time-lag fuses which often have a central core or a supporting spring arrangement, but I wouldn't take any chances here if I still used fuses) I strongly recommend that these be replaced frequently, whether they have blown or not.
Indeed, whilst in conversation recently with the manufacturer of some audio equipment I had been asked to repair, he advised me to change the internal fuses for this very purpose of 'restoring' the sonic performance, while it was open on the bench.
By that time it was quite clear to me that fuses had to go as these effects were not acceptable to me, so I 'hard-wired' out all these 12 fuses temporarily, and the result was so clearly an improvement that this could have been another amp design altogether. JLH heard this in due course, and was rather surprised himself, but I was aware of the risks involved in not having any protection at all, and returned to a safer situation whilst I investigated alternative methods of achieving this.
Already this post is longer than I had hoped for, so I will start again soon with more details of what I found to be an acceptable way of resolving this problem, and recommend some suitable thermal (not magnetic!) circuit breakers which I have used universally for many years since, and found to be very satisfactory from their sonic aspects.
Lastly, for any prospective 'wit' who has become bored with this posting, just remember it has taken me a lot longer to write this than for you to read it, and I am not doing it for my own benefit as I already know these answers. I am simply trying in the ways I know best (as an old man) to help others who might be interested. 🙂
Regards,
courage said:
You are very welcome, Franklin, and I hope that it is of some value.
Last year an audio commission prevented us from having any holidays away from home, but we made up for it this year and returned about a month ago after 10 very enjoyable days in your country.
Regards,
KBK said:
Hi KBK,
No, its not what you seem to be thinking at all, but thank you for your thoughts here.
I also design and build my own speakers and I am quite satisfied with their performance.
My 'problem' (minor though it is) is that I have always been an extremely sensitive person physically since childhood - touch, sight, and hearing etc. - which has enabled me to do many things in life which others are unable to. However, there always seem to be some downsides to unusual 'benefits' and with advancing years I have become incredibly fussy about most matters lately, and much less tolerant overall to physical stimuli.
The example I mentioned was simply a desire for perfection, and only affects a handful of recordings I have. There is nothing objectionable about the sounds to me at either setting (and I don't listen at 'elevated' levels anyway) and a few years ago I wouldn't have noticed it.
"Give a mouse a cookie - and he'll want a glass of milk" is quite apt here I guess, and with accurate 1dB setting steps which I have been using at home for a while, I guess I have become a bit spoiled.
I should be ashamed of myself!
Regards,
Sigurd Ruschkow said:
Hi Sigurd,
I subsequently corrected my figures as you may have seen, but it is still only 13 resistors per channel actually, and when I previously counted the resistor pads on a spare PCB at hand, I must have had a 'senior moment' and forgot that this covers 2 channels. So there are 13 different values to be bespoke made, but this is still quite costly unless the manufacturer I did this design for wishes to help out.
I will try them soon, one way or another, but it is not a priority for me right now.
You are right about it being a never-ending quest, but isn't that part of the fun in life? Without such an impetus I wouldn't have learned what I now know, and I am still discovering new things most days of my life. System improvements are an iterative process as far as I am concerned, and improvements in one area often (always?) highlight shortcomings or the weakest parts in other areas.
With passing years I have found identifying these weaker areas easier (note- NOT easy, nor necessarily always successfully!) probably due to 35-40 yrs of familiarity/experience with such matters, but as one's test system improves and the resolution goes up, it is much easier to hear precisely what is going on and one can get more 'guesses' right first time. This is also encouraging as less time is wasted going down time-consuming (but usually illuminating) blind-alleys, but inevitably this will still happen far more times than one might wish.
Regards,
Don't most bi-stable relays then use permanent magnets to "remember" and hold the last state?janneman said:
Is the concern a magnetic field or an electromagnetic field with residual AC noise in the coil current? It seems a relevant question as the residual AC in the coil can be made arbitrarily low.
I had problems with pots, I clearly dion't have the access to resources that John does. The Blowtorch is certainly famous enough, if John says you can make a pot work, them I'm sure you can.
Still other's do use relays, Conrad Johnson in their high end pre-amps, Placette, and a host of others I can't think of this instant.
An earlier Levinson used an all analog multiplier and someone must have used a multiplying DAC by now.
For me getting up to change the volume is not preferred, but it is tolerable.
Bobken said:
Yes, in a binary configuration it is possible to get away with one relay and a pair of resistors per each binary step, so for 1 dB step and 64 steps one needs 12 resistors + 1 for load = 13. I did design this kind of volume control in 2001, using a custom resistor array - essentially a single resistor with 6 taps and a set of 7 trimmed to 0.1% resistors for 128 steps . Unfortunately, it never reached production.
Cheers
Alex
Attachments
This fuse thing, disturbs me. It is my experience that we need protection from shorts and major breakdowns in audio equipment. UL was started as a fire underwriter, and found more than a century ago that fuses were necessary to prevent fires.
Parasound has had instances where only the fuses prevented SERIOUS fires, and even then, we came darn close. Two instances happened when a customer put one of my power amps in a closed wooden cabinet and ran it night and day. This overheated the caps (mainland Chinese manufacture, and this is important in this case) and the caps went into thermal runaway and spewed conductive solvent all over the inside of the case, creating serious and multiple shorts. The accident certainly scorched the cabinet, and without fuses, who knows?
Now, I am not a 'safety wonk' someone who jumps up and down IF some ground is bypassed for better performance, but my life experience shows me that we can go too far. It would be like taking the rev limiter out of autos. They would go faster, but I would probably blow the engine, the way I rev. That is why the rev limiter is there. Someone with an automatic transmission may never reach max revs in their lifetime, then the rev limiter seems an almost useless add on.
For decades we have noticed that fuses often cause sonic degradation. It should not be too surprising. When you look at what a fuse is, how it is usually made seriously nonlinear with temperature, and that it takes self heating to make it blow, you find a resistor in series with the line that changes with the music signal dynamics.
However, there are viable alternatives. Just like some kid who found that taking the air cleaner on his first car made his car run slightly faster, leaving off the air cleaner is not a good idea. Maybe an improved air cleaner will work well enough, and offer the protection necessary.
It is the same with fuses. Now there are expensive alternative fuses, made of pure silver, that are more linear, yet seem to meet safety specs.
There are also circuit breakers, both thermal and electromagnetic. They can sound pretty good too, IF they are used properly.
In any case, I cannot disagree with someone who finds that fuses virtually always change the sound in their highly tweaked system, but we cannot recommend this to enthusiastic amateurs, or even professionals. It is playing with fire.
Parasound has had instances where only the fuses prevented SERIOUS fires, and even then, we came darn close. Two instances happened when a customer put one of my power amps in a closed wooden cabinet and ran it night and day. This overheated the caps (mainland Chinese manufacture, and this is important in this case) and the caps went into thermal runaway and spewed conductive solvent all over the inside of the case, creating serious and multiple shorts. The accident certainly scorched the cabinet, and without fuses, who knows?
Now, I am not a 'safety wonk' someone who jumps up and down IF some ground is bypassed for better performance, but my life experience shows me that we can go too far. It would be like taking the rev limiter out of autos. They would go faster, but I would probably blow the engine, the way I rev. That is why the rev limiter is there. Someone with an automatic transmission may never reach max revs in their lifetime, then the rev limiter seems an almost useless add on.
For decades we have noticed that fuses often cause sonic degradation. It should not be too surprising. When you look at what a fuse is, how it is usually made seriously nonlinear with temperature, and that it takes self heating to make it blow, you find a resistor in series with the line that changes with the music signal dynamics.
However, there are viable alternatives. Just like some kid who found that taking the air cleaner on his first car made his car run slightly faster, leaving off the air cleaner is not a good idea. Maybe an improved air cleaner will work well enough, and offer the protection necessary.
It is the same with fuses. Now there are expensive alternative fuses, made of pure silver, that are more linear, yet seem to meet safety specs.
There are also circuit breakers, both thermal and electromagnetic. They can sound pretty good too, IF they are used properly.
In any case, I cannot disagree with someone who finds that fuses virtually always change the sound in their highly tweaked system, but we cannot recommend this to enthusiastic amateurs, or even professionals. It is playing with fire.
I'm suspicious of Silver fuses. Getting the wire to melt predictably isn't easy and silver has a much better thermal conductivity that the special alloys so It would need to be thinner or otherwise specially constructed. Getting UL approval would require a lot of testing and at $$$ each a lot of bucks. Do they have a recycle program to reclaim the silver after the fuse blows?
Having been involved with Surge protectors for the last 10 years i have seen the importance of through safety counciousness in the design, even when it may compromise the performance. The about pof power available on the mains is very considerable. We test surge protectors with a 220V 500A service to make sure they disconnect properly. (That's a 110KW service, much higher than any domestic service.) Its required because that the potential fault current if the line is shorted. And that much power can make many things burn, not to mention turning organic things (like people) into carbon.
Speakers catching on fire is actually too common- think about a coil of fine wire glued to paper, fed a lot of power in a wood box- and even with fusing can happen. Even in the UK.
Having been involved with Surge protectors for the last 10 years i have seen the importance of through safety counciousness in the design, even when it may compromise the performance. The about pof power available on the mains is very considerable. We test surge protectors with a 220V 500A service to make sure they disconnect properly. (That's a 110KW service, much higher than any domestic service.) Its required because that the potential fault current if the line is shorted. And that much power can make many things burn, not to mention turning organic things (like people) into carbon.
Speakers catching on fire is actually too common- think about a coil of fine wire glued to paper, fed a lot of power in a wood box- and even with fusing can happen. Even in the UK.
Demian, I suspect that the silver fuses are German rated for safety. Better than UL, in my opinion.
Fuses (or lack of) part 2.
Having decided that conventional fuses were just not good enough for very HQ audio applications, the obvious alternative is a circuit-breaker.
Unfortunately, these are rather costly, so I thought about (and discussed with JLH) the real requirements for these protection devices like fuses, together with the consequences if they were dispensed with.
Going from the wall-socket, the fuses in plugs in the UK are (not very good) ceramic rather slow-reacting things anyway (but doubtless excellent non-linear resistors!) and with always retaining lower-rated protection at the other end of the mains lead concerned, as I wouldn't dispense with this, all that this plug fuse will protect is the lead itself.
I have never had a dog which would chew through such mains leads (and I doubt it would have saved him from 'frying' if the plug fuse was still installed) nor does my wife mow the carpets with our outdoors lawn-mower, so I concluded that there was no great risk in dispensing with this wretched plug fuse, so for my own domestic use I no longer install fuses in wall plugs. No doubt I will quickly be corrected if I am wrong, but I think that the UK is (almost?) unique in this plug fuse requirement anyway, so I reckoned I wouldn't be any less safe than any other Continentals, Americans, Canadians, Japanese etc. etc., but for all others in the UK reading this, you must make your own decisions about this choice and live with the consequences.
Leaving the equipment's input protection for now, I have built many other designs, some of which even have fuses in the transformer's secondaries, but they have never blown in use and when linked out, the sonic performance is always enhanced. Much the same with PS rail fuses on the PCBs which have never given me any problems in over 40 yrs. of my using such equipment, and as JLH agreed, with a fused dual PS it is quite likely that if one fuse blows, the opposite polarity fuse may well not do so. In many (most?) such instances this will very likely give the attached speaker a 'dose' of full rail voltage from the remaining PS, and I know of few other drivers apart from ones I use (ATC) which will take very kindly to this abuse, unless (as in many cases) there is an intervening series capacitor which should prevent any DC problems from causing damage, whichever choice is adopted.
Generally I don't care much for the obvious alternative here, being a protection scheme possibly with a relay to break the circuit if there are output irregularities, but this is a very much better choice than any non-linear fuse which is playing havoc with the delicate audio signals when it is first installed, and it is anyone's guess what will happen after some months in use.
I don't believe that any protection scheme will ever 'enhance' sonics in any way, and being something of a minimalist by nature I prefer to avoid any additional circuitry unless it is vital, and as I use better connections (less harmful) between amps and speakers than relays of any kind, I also prefer not to compromise this integrity.
Considering a new amp design currently which is DC coupled (of course!) and which will be used in part of an active speaker system utilising tweeters retailing in the region of £500 a pair which will be directly coupled to the amp's output, I may depart from this, at least until I am happy with the stability of the amp concerned!
For those who like the protection afforded by a relay in the amp's output, Amplimo make a very cleverly-designed relay specifically for this purpose, with one butch contact which always leads in operation (and more-significantly breaks last, as I mentioned in the relay discussion) and with another contact constructed of higher-quality materials acting as a backup, which shouldn't deteriorate over time. I have some of these Amplimo relays for testing, and compared with a straight wire during listening trials the losses are very slight, but I haven't yet used them in any design, nor for any extended period. Should I ever feel the need to use an output relay, these would certainly be my first choice here.
Returning to the unit's mains input protection, which I believe should be retained whatever (unfortunately), especially if most other protection is dispensed with, there are many suitable circuit breakers, but a long while ago I hit upon the excellent German made ETA types. These are very versatile in current ratings, apearance choices, alternative tag arrangements etc., and ETA will make these up to customer's specs quite quickly. If they trip, you won't ever be short of the right type of fuse as all that is needed is to close the switch again!
The range I probably use most is 3120, and a lot of distributors keep these in quite a variety of ratings (Farnell, Rapid & R.S. in Europe, for example) and although they cost a lot more than fuse arrangements they are excellent sonically, unlike fuses. These are a combined switch and breaker in a single unit, and this avoids the need for a separate on-off switch too, and can be had in single and dual pole versions, and most 'popular' values can probably be found 'off the shelves' if you hunt around a bit. The last time I had some specially made by ETA in the UK, it took only a few days and the cost (even in relatively low quantities) was not much different from distributor's advertised prices, so you can get an idea of costs from a search on this.
I have also listened to the sonic effects of a lot of different rocker switches (this is another tale!) which also vary quite a bit surprisingly, but the ETA breakers are about as benign as any other switch I ever listened to, and as one usually requires an on-off switch anyway, why not combine the two?
I have used this range universally for myself with very little if any sonic detereoration over many years, and recently in several commercial designs, and in amps, DACs, buffers, and all kinds of other audio equipment.
For lower current devices, there are several other ETA breakers which I guess will perform similarly from the sonic aspect, and I have also used the cheaper push-button breakers ETA make, also with success, but, if you use these, choose the then-necessary additional on-off switch carefully!
Probably other makes of similar breakers will be O.K. too, and much superior to fuses, but I will only speak from experience and being quite satisfied with ETA's offerings for so many years now, I don't feel the need to experiment further. I haven't used anything other in any design since their 'discovery', but if a need arose which ETA cannot satisfy (unlikely) I would try alternatives but never go back to fuses.
There are a couple of points to watch out for, especially if like me you are hung-up on keeping mains impedance low for power amps, particularly those which don't draw current at a continuous rate. Below around 3A rating the contact resistance does go up a bit, but above this value most other mains switch resistances will be very little different anyway (and many deteriorate alarmingly with only a few dozen operations!) and with the switch-type breakers you avoid this extra switch. Mostly, the lower-rated types will be used for preamps DACs or whatever probably with a more constant current requirement, and I don't have so much concern about the mains impedance in these cases. Indeed, in some designs deliberately added impedance in the PS swamps this anyway.
The other snag is their DC voltage rating, which is usually very much lower than the AC rating, with the 3120 series being OK for 240v AC, but merely 50v DC.
I haven't yet found this to be a problem and there are higher DC rated products, but recently a manufacturer asked me to recommend some component changes which might improve the sonics of a powerful amp module he was working on. Coming up with a list of possible passive device changes to try out as potential alternatives was easy, as was a more suitable transformer, but these modules had two 6A fuses in the dual PS, and the rail voltages were in the region of +/- 90v DC.
I spent nearly an entire morning searching for suitable breakers to replace these fuses, and still only came up with one likely candidate, and the difficulties in mounting this woud have made it un-economic in my opinion.
At least as DIY'ers there are no worries about the cost/benefit ratio for any of your mods! 🙂
The special fuses John refers to are very costly, and I guess that they will be an improvement over any standard fuse available, but I don't think for a moment that they will be up to the contact breakers 'performance' especially over time, and in most instances I think the CBs will be a cheaper option, albeit not a simple clip-in installation. I also believe that unless one has unreliable and/or unstable equipment which I would steer well clear of personally, provided that a sensibly-chosen breaker is installed in the audio equipment (including power amps) at their mains input, that this will be a safe enough option from the fire aspect, and it is up to individuals to decide on the risks which could apply with the rest of their equipment. Indeed, the vast majority of any gear other than power amps that I have looked at over the years, only have a single mains input fuse installed by their makers, anyway.
Finally, I will not build any equipment and haven't done so for 25 years without adding a Transzorb (or one of the later clones) across the mains input, but *after* the breakers as they fail shorted. These are quite benign in unwanted characteristics unlike all other caps, VDRs, or whatever else I have seen recommended over the years, and all of which adversely affect the sound in an extremely high-resolution system, whether we like it or not. Transzorbs are quite remarkable devices at dealing with surges in the supply, and they actually enhance the listening experience in every piece of equipment I have tried them, and that is a lot.
Regards,
Having decided that conventional fuses were just not good enough for very HQ audio applications, the obvious alternative is a circuit-breaker.
Unfortunately, these are rather costly, so I thought about (and discussed with JLH) the real requirements for these protection devices like fuses, together with the consequences if they were dispensed with.
Going from the wall-socket, the fuses in plugs in the UK are (not very good) ceramic rather slow-reacting things anyway (but doubtless excellent non-linear resistors!) and with always retaining lower-rated protection at the other end of the mains lead concerned, as I wouldn't dispense with this, all that this plug fuse will protect is the lead itself.
I have never had a dog which would chew through such mains leads (and I doubt it would have saved him from 'frying' if the plug fuse was still installed) nor does my wife mow the carpets with our outdoors lawn-mower, so I concluded that there was no great risk in dispensing with this wretched plug fuse, so for my own domestic use I no longer install fuses in wall plugs. No doubt I will quickly be corrected if I am wrong, but I think that the UK is (almost?) unique in this plug fuse requirement anyway, so I reckoned I wouldn't be any less safe than any other Continentals, Americans, Canadians, Japanese etc. etc., but for all others in the UK reading this, you must make your own decisions about this choice and live with the consequences.
Leaving the equipment's input protection for now, I have built many other designs, some of which even have fuses in the transformer's secondaries, but they have never blown in use and when linked out, the sonic performance is always enhanced. Much the same with PS rail fuses on the PCBs which have never given me any problems in over 40 yrs. of my using such equipment, and as JLH agreed, with a fused dual PS it is quite likely that if one fuse blows, the opposite polarity fuse may well not do so. In many (most?) such instances this will very likely give the attached speaker a 'dose' of full rail voltage from the remaining PS, and I know of few other drivers apart from ones I use (ATC) which will take very kindly to this abuse, unless (as in many cases) there is an intervening series capacitor which should prevent any DC problems from causing damage, whichever choice is adopted.
Generally I don't care much for the obvious alternative here, being a protection scheme possibly with a relay to break the circuit if there are output irregularities, but this is a very much better choice than any non-linear fuse which is playing havoc with the delicate audio signals when it is first installed, and it is anyone's guess what will happen after some months in use.
I don't believe that any protection scheme will ever 'enhance' sonics in any way, and being something of a minimalist by nature I prefer to avoid any additional circuitry unless it is vital, and as I use better connections (less harmful) between amps and speakers than relays of any kind, I also prefer not to compromise this integrity.
Considering a new amp design currently which is DC coupled (of course!) and which will be used in part of an active speaker system utilising tweeters retailing in the region of £500 a pair which will be directly coupled to the amp's output, I may depart from this, at least until I am happy with the stability of the amp concerned!
For those who like the protection afforded by a relay in the amp's output, Amplimo make a very cleverly-designed relay specifically for this purpose, with one butch contact which always leads in operation (and more-significantly breaks last, as I mentioned in the relay discussion) and with another contact constructed of higher-quality materials acting as a backup, which shouldn't deteriorate over time. I have some of these Amplimo relays for testing, and compared with a straight wire during listening trials the losses are very slight, but I haven't yet used them in any design, nor for any extended period. Should I ever feel the need to use an output relay, these would certainly be my first choice here.
Returning to the unit's mains input protection, which I believe should be retained whatever (unfortunately), especially if most other protection is dispensed with, there are many suitable circuit breakers, but a long while ago I hit upon the excellent German made ETA types. These are very versatile in current ratings, apearance choices, alternative tag arrangements etc., and ETA will make these up to customer's specs quite quickly. If they trip, you won't ever be short of the right type of fuse as all that is needed is to close the switch again!
The range I probably use most is 3120, and a lot of distributors keep these in quite a variety of ratings (Farnell, Rapid & R.S. in Europe, for example) and although they cost a lot more than fuse arrangements they are excellent sonically, unlike fuses. These are a combined switch and breaker in a single unit, and this avoids the need for a separate on-off switch too, and can be had in single and dual pole versions, and most 'popular' values can probably be found 'off the shelves' if you hunt around a bit. The last time I had some specially made by ETA in the UK, it took only a few days and the cost (even in relatively low quantities) was not much different from distributor's advertised prices, so you can get an idea of costs from a search on this.
I have also listened to the sonic effects of a lot of different rocker switches (this is another tale!) which also vary quite a bit surprisingly, but the ETA breakers are about as benign as any other switch I ever listened to, and as one usually requires an on-off switch anyway, why not combine the two?
I have used this range universally for myself with very little if any sonic detereoration over many years, and recently in several commercial designs, and in amps, DACs, buffers, and all kinds of other audio equipment.
For lower current devices, there are several other ETA breakers which I guess will perform similarly from the sonic aspect, and I have also used the cheaper push-button breakers ETA make, also with success, but, if you use these, choose the then-necessary additional on-off switch carefully!
Probably other makes of similar breakers will be O.K. too, and much superior to fuses, but I will only speak from experience and being quite satisfied with ETA's offerings for so many years now, I don't feel the need to experiment further. I haven't used anything other in any design since their 'discovery', but if a need arose which ETA cannot satisfy (unlikely) I would try alternatives but never go back to fuses.
There are a couple of points to watch out for, especially if like me you are hung-up on keeping mains impedance low for power amps, particularly those which don't draw current at a continuous rate. Below around 3A rating the contact resistance does go up a bit, but above this value most other mains switch resistances will be very little different anyway (and many deteriorate alarmingly with only a few dozen operations!) and with the switch-type breakers you avoid this extra switch. Mostly, the lower-rated types will be used for preamps DACs or whatever probably with a more constant current requirement, and I don't have so much concern about the mains impedance in these cases. Indeed, in some designs deliberately added impedance in the PS swamps this anyway.
The other snag is their DC voltage rating, which is usually very much lower than the AC rating, with the 3120 series being OK for 240v AC, but merely 50v DC.
I haven't yet found this to be a problem and there are higher DC rated products, but recently a manufacturer asked me to recommend some component changes which might improve the sonics of a powerful amp module he was working on. Coming up with a list of possible passive device changes to try out as potential alternatives was easy, as was a more suitable transformer, but these modules had two 6A fuses in the dual PS, and the rail voltages were in the region of +/- 90v DC.
I spent nearly an entire morning searching for suitable breakers to replace these fuses, and still only came up with one likely candidate, and the difficulties in mounting this woud have made it un-economic in my opinion.
At least as DIY'ers there are no worries about the cost/benefit ratio for any of your mods! 🙂
The special fuses John refers to are very costly, and I guess that they will be an improvement over any standard fuse available, but I don't think for a moment that they will be up to the contact breakers 'performance' especially over time, and in most instances I think the CBs will be a cheaper option, albeit not a simple clip-in installation. I also believe that unless one has unreliable and/or unstable equipment which I would steer well clear of personally, provided that a sensibly-chosen breaker is installed in the audio equipment (including power amps) at their mains input, that this will be a safe enough option from the fire aspect, and it is up to individuals to decide on the risks which could apply with the rest of their equipment. Indeed, the vast majority of any gear other than power amps that I have looked at over the years, only have a single mains input fuse installed by their makers, anyway.
Finally, I will not build any equipment and haven't done so for 25 years without adding a Transzorb (or one of the later clones) across the mains input, but *after* the breakers as they fail shorted. These are quite benign in unwanted characteristics unlike all other caps, VDRs, or whatever else I have seen recommended over the years, and all of which adversely affect the sound in an extremely high-resolution system, whether we like it or not. Transzorbs are quite remarkable devices at dealing with surges in the supply, and they actually enhance the listening experience in every piece of equipment I have tried them, and that is a lot.
Regards,
I've been building my own fuses as of late. 😉 😉
Do you not have 'balanced' AC in the UK? I can't imagine running 220-240AC on one line.
In the US and Canada, at the 120AC, we're restricted to a single active line, so this device you speak of, what is it? (Transzorb)
ok. I see. transient voltage suppressors.
I'm running balanced 120Vac off of a 100kva transformer (about 500lbs or so), so I'm probably a bit lower in AC line noise than most that are in NA. (North America)
Although...every time we have a storm here..I wonder what it would be like to get a direct hit on that transformer.
I do have about 4000W of solar cells and have been meaning to to put together a solar powered quad of 100V power rails for amplification.
Could be fun.
Do you not have 'balanced' AC in the UK? I can't imagine running 220-240AC on one line.
In the US and Canada, at the 120AC, we're restricted to a single active line, so this device you speak of, what is it? (Transzorb)
ok. I see. transient voltage suppressors.
I'm running balanced 120Vac off of a 100kva transformer (about 500lbs or so), so I'm probably a bit lower in AC line noise than most that are in NA. (North America)
Although...every time we have a storm here..I wonder what it would be like to get a direct hit on that transformer.
I do have about 4000W of solar cells and have been meaning to to put together a solar powered quad of 100V power rails for amplification.
Could be fun.
john curl said:
Personally I never use any fuses in my own built audio amplifiers.
I can do this
.. as I am the only user and take whatever responsibility
for the consequency.
In my mind fuses, in amp power supply / loudspeaker output are un-desired.
We must realize that such fuses will be a part the performance.
However small part it is in reality.
There are possibilities other than fuses.
For protection.
Current sensing coils windings around power supply wires
+ very low resistance circuit breaking relays is one hi-fi option.
I never practice any protection for my own homebuilt gears.
I realize all can not keep designs as clean as this.
Lineup
brag
I do not understand what you suddenly have against me, lineup
Because this wouldnt be a fair description of my post.
If you re-read it with postive glasses on.
All I was saying is that I can get away without any fuses.
And that current sensing of wires + relays is one good option to use.
Regards.
I do not understand what you suddenly have against me, lineup
Because this wouldnt be a fair description of my post.
If you re-read it with postive glasses on.
All I was saying is that I can get away without any fuses.
And that current sensing of wires + relays is one good option to use.
Regards.
1audio said:
Hi Demian,
I can assure you that I think about everything, much more so than anyone else I have yet come across in my lifetime, but reality has to come into the picture somewhere, too. This does not mean I consider myself omniscient, and I do still sometimes make mistakes or overlook matters, but I have yet to learn of anything specific which would change my present views on this issue.
Instead of accepting what (almost) everyone else is apparently content with, I know that attention to all such details is what counts for the finest results. Hence my extensive investigations into these areas which are ignored by so many, but which bring substantial rewards by improving the listening experience.
We are in quite enough trouble over on this side of the pond without you picking on us 🙂 but unlike some fellow countrymen, I don't share the arrogant attitude that we in the UK are any more important in the whole scheme of things than the rest of the world. Accordingly I don't accept that we deserve any *greater protection* through fuses in wall plugs (for what they are worth) than you Americans, or whoever.
Quite apart from my own long-term experiences and apparently the Fire Service records here, Nelson has been selling (rather high-current!) amplifiers for a long while now, and having looked at a lot of his schematics, I don't recall ever seeing anything other than a single fuse being used at the mains input. If I am wrong here it won't be more than a few minutes before I am corrected, but if I am right even in not all instances, I wonder how he also gets away with this implied reckless attitude to safety?
I recommend exactly the same approach, but suggest that the sound will be improved (especially with passage of time when fuse-deterioration inevitably occurs) if the fuses are replaced with a generally more accurate and more reliable means of protection, through the use of circuit breakers. IF I still favoured fuses, I can also assure you that by now I would have carefully investigated their breaking performance after some time in use, as following on from my earlier investigations I would put money on the fact that this will change, although I won't guess at which way it might go.
I do appreciate that you were not aware of my own personal recommendations for protection when you commented as above, and I am not looking for an argument with anyone, as I said up front. However, I still feel it is important to stick to reality in such situations, and even unwittingly intimidating anyone into avoiding improving their own DIY set-ups with vague suggestions of certain disaster being around the corner are not very helpful, if I may observe. Also, although this relates to a comment from John, since the time when the Underwriters Laboratory was inaugurated, quality and especially reliablity of electronics in general has improved immensely in my experience.
Voice coils on *paper* formers, where on earth have you seen these? 😉
Regards,
KBK said:
Hi KBK,
No, our supply is nominally 230vAC (now), and it is not balanced for domestic users so I guess we must like living dangerously!
I cannot speak too highly of Transzorbs and what they do by way of protection/enhancement to the mains supply. I have the original maker's (Gen. Semi. Ind. Inc.) data book dating back maybe 25 yrs. or more, and if you can get hold of a copy somehow, you will find this very valuable.
It is over 300 pages crammed full of extremely useful suppression/protection-related info, data, applications advice, and a few light-hearted snippets.
The one which sticks in my mind most is "A lightning strike can contain currents in excess of 200,000 amperes, with a total collective energy to lift the QE 2 (the Liner) two feet in the water."
Many times I have recommended (also on this Forum) the use of these devices or their generic clones nowadays, which don't appear any different from some brief trials I undertook. I still use GSI versions because I needed to have them imported back then, and a large order gave me a good price-break.
Interestingly, and not surprisingly, all of the subsequent feedback has been positive since making these suggestions to others.
I feel I should give this thread a break now before John gets fed-up with my intrusions.
Regards,
Re: Fuses (or lack of) part 2.
Bob -
thanks for sharing your views on the fuse problem!
I do say problem as fuses are components that we actually do not want in our high end gear but are an evil necessity. Maybe I should write over current protection instead as there are many ways to "fuse protect" an apparatus. Circuit brakers maybe being the best solution.
A power-on switch is something I have stopped using for low power stuff. Even my class AB power amps are on all the time. But with class A power amps, the situation is a bit different as even in cold Sweden my class A power amps heats up the room too much when they are left on 24h/day. Here a circuit braker will be used for power on/off.
Also, I will use thermal circuit brakers mounted on the fan cooled heatsinks. These brakers will cut of AC mains when the HS temp is over 70 deg C.
So, I try to use a minimum of over-current protection but will never ever stop using some form of "fuse" on the AC mains.
Regarding DC protection on the output of power amps, I am kinda ambivalent. I sometimes use them, sometimes not. Not having found the "right" relay makes the decision harder. Also, the added complexity adds to the drawback side.
I would like to add that many households here in Sweden have not only the standard ceramic fuses in the electricity mains central but also ground current sensing brakers. Ie if there is more than a few mA in the AC mains GND wire in the central, the AC mains is shut down for the hole house.
BTW,
we have had mice chew up our AC mains cables in the outdoor electricity mains AC central. They chew on the copper, too, so, after a while, there were not much copper left and finally the cable burned up as it was too thin!
Some fires are started this way. Or they start as the copper becomes too "oxidised" and the cables get too hot and voila we have a fire in the cable inside a wall.
Sigurd
Bob -
thanks for sharing your views on the fuse problem!
I do say problem as fuses are components that we actually do not want in our high end gear but are an evil necessity. Maybe I should write over current protection instead as there are many ways to "fuse protect" an apparatus. Circuit brakers maybe being the best solution.
A power-on switch is something I have stopped using for low power stuff. Even my class AB power amps are on all the time. But with class A power amps, the situation is a bit different as even in cold Sweden my class A power amps heats up the room too much when they are left on 24h/day. Here a circuit braker will be used for power on/off.
Also, I will use thermal circuit brakers mounted on the fan cooled heatsinks. These brakers will cut of AC mains when the HS temp is over 70 deg C.
So, I try to use a minimum of over-current protection but will never ever stop using some form of "fuse" on the AC mains.
Regarding DC protection on the output of power amps, I am kinda ambivalent. I sometimes use them, sometimes not. Not having found the "right" relay makes the decision harder. Also, the added complexity adds to the drawback side.
I would like to add that many households here in Sweden have not only the standard ceramic fuses in the electricity mains central but also ground current sensing brakers. Ie if there is more than a few mA in the AC mains GND wire in the central, the AC mains is shut down for the hole house.
BTW,
we have had mice chew up our AC mains cables in the outdoor electricity mains AC central. They chew on the copper, too, so, after a while, there were not much copper left and finally the cable burned up as it was too thin!
Some fires are started this way. Or they start as the copper becomes too "oxidised" and the cables get too hot and voila we have a fire in the cable inside a wall.
Sigurd
Bobken said:
SY said:
The moving armature is in the magnetic path and the PM supplies the "spring" force to hold the relay armature in position. There is an air gap in this magnetic circuit, and there is 50/60Hz ac flux from the power transformer inside the case in a one-box unit. I don't know how much of this ac line flux gets into the relay armature air gap, since it decreases with the cube of distance in the near field. But it's conceivable that some ac line flux variation could get into the audio signal via the relay armature magnetic circuit.
Vibration (from a nearby loudspeaker for instance) could also induce very small dynamic changes in the armature air gap. Every mechanical system has a resonance frequency and music covers quite a large range of vibration frequencies.
Either of these could produce an unintended change in the motion of the armature, and introduce ac flux to the audio signal.
- Status
- Not open for further replies.