Dejan, no hard and fast rules, or technique - what I've done over the years has varied depending upon my knowledge at the time, or interest in learning something new; and to be perfectly honest, I'm a lazy bugger - sometimes close enough is good enough, I wear out quickly these days,...
Everything I do is based upon what feedback I get from listening to the amp playing tracks, I don't do measurements, or guesstimate an approach from looking at the internals - my attitude is that of a troubleshooter: I listen for what the amplifier is doing wrong, pick the subjectively most offensive misbehaviour and proceed to do what I believe will be most effective in improving things ... in one sense asking me how I go about things is like asking a mechanic how he fixes a car - in most cases I reckon the answer he'll give is, "Well, depends on what's wrong with it ...!!",
Okay, I listen for the naturalness of the sound it produces, and how loud it can go before it loses its composure ... "naturalness"? This is a subjective measure of how much of a smell it has of that typical "hifi" sound, the artificial quality which makes vocals, for example, sound nothing like a real human voice. The ability to go loud is pretty obvious, which has to track along with the ability to remain "natural".
What I then do depends totally on what I hear, I don't bother fiddling with anything unless I believe it will have a direct impact on the problems I heard from the listening.
So sorry, this is probably not what you were after ... I don't work from a menu of "things to do" - I listen, and then let the results of that guide me ...
No, quite all right Frank.
My methods are a bit different. I look first at the PSU section because I assume that's where the biggest "savings" will happen. When I see a say nominally 50W/8 Ohms amp with two weedy little 4,700 uF I don't have to guess or think much, replace them with 10,000 uF which I always have a stock of. Can't go wrong.
Next, the regulators. Look 'em over look up the specs for the transistors used because I often find they have been rather stretched in an otherwise resonable circuit, so swapping them them for something bigger and stronger is a fairly safe bet.
The solution of usinf fuses as protection for the output stage, much as I hate it as I find it to be unreliable, is best because it frees me from having to understand and recalculate the protection circuits if I want to change the usually very cost-conscious output pairs. Often, these are 130 or 150W devices, and I usually swap them for ON Semi's version of 200W devices, the MJL 3281/1302 Like all Motoroloa/ON Semi power devices, they seem to be indestructable unless you bang them with a hammer, and besides, more power reserve can never hurt.
And so forth. I usually don't touch the preamp except to install bigger capacitors in its PSU lines
I agree that the PSU area is key, except that my approach would be to go either completely OTT, say 10 times the energy storage, or re-engineer parts of how the supplies work. What I'm after here is for the unit to be capable of going right up to clipping without noticeable tonal changes - we agree on this, of course - with the Philips HT setup cheapie I used the re-engineering variation, and this was highly successful, in part because the woofer duties were passed over to a separate amplifier and PS, in the original unit.
It always comes down to, "Am I happy with the sound, on all the recordings I listen to?" If not, then there is a problem somewhere, which can be fixed - this is an approach that has always, always worked ... if it ain't broke ...
It always comes down to, "Am I happy with the sound, on all the recordings I listen to?" If not, then there is a problem somewhere, which can be fixed - this is an approach that has always, always worked ... if it ain't broke ...
Not so Nige. You are still with us, and when you think about it, Frank and I are not really different. We're hands on people, even if we do use simulators.
I agree that the PSU area is key, except that my approach would be to go either completely OTT, say 10 times the energy storage, or re-engineer parts of how the supplies work. What I'm after here is for the unit to be capable of going right up to clipping without noticeable tonal changes - we agree on this, of course - with the Philips HT setup cheapie I used the re-engineering variation, and this was highly successful, in part because the woofer duties were passed over to a separate amplifier and PS, in the original unit.
It always comes down to, "Am I happy with the sound, on all the recordings I listen to?" If not, then there is a problem somewhere, which can be fixed - this is an approach that has always, always worked ... if it ain't broke ...
10 times the energy storage is nice and fine, Frank, but it's often not possible to implement for lack of space. In the Marantz 1152 DC integrated amp, I had to lay the two 22,000 uF caps horizontally, they wouldn't fit otherwise.
It also greatly depemds on how the output stage was initially built. For example, H/K 6550 integrated was equipped with two 13,000 uF caps. Changing these for bigger did exactly nothing for it because there was no problem to fix in the first place because it was done very competently. I swapped them anyway for two 15,000uF caps because that's how much space i had.
Nige, it is a sad fact that most people buy without trying out first. And, as you rightly pointed out, they will sooner buy what they and their friends see as a desirable model rather that another which is not so desirable but sounds better. Fell free to blame the mags and many Internet sites for this sad state of affairs. Also, I find far too many people think about the asking price in terms of second hand price they can get when selling, which means they are not serious, they are playing a rather expensive game trying to outbid themselves.
A friend was selling a pair of speakers he made himself. They sounded truly excellent, built around Morel drivers. A guy came and said: "Yes, they sound better than my Dynaudio speakers, but what can I get for them when I want to sell them?" Cretin. He's buying now only to sell later in search of elusivfe audio nirvana and when you ffer him a piece of it, he says he will be selling them.
A friend was selling a pair of speakers he made himself. They sounded truly excellent, built around Morel drivers. A guy came and said: "Yes, they sound better than my Dynaudio speakers, but what can I get for them when I want to sell them?" Cretin. He's buying now only to sell later in search of elusivfe audio nirvana and when you ffer him a piece of it, he says he will be selling them.
Yes, I have found simulators to be very powerful tools for understanding "what's really going on", and that's often quite different from what the standard textbooks say, or indicate is important - I've been through 3 separate software packages over the years, and have learnt much from each one.
As an example, my own gainclone design was really all about getting the best power supply happening, and this was done totally in the simulator: I worked out the optimum component for each position, then bought the parts and assembled the beast. It worked first go, and I didn't change a thing, or experiment with the configuration - I was happy with the sound, in the first instance. Its only "weakness" was that it was still influenced by poor quality mains - but I was still learning about how to best deal with such, at the time ...
As an example, my own gainclone design was really all about getting the best power supply happening, and this was done totally in the simulator: I worked out the optimum component for each position, then bought the parts and assembled the beast. It worked first go, and I didn't change a thing, or experiment with the configuration - I was happy with the sound, in the first instance. Its only "weakness" was that it was still influenced by poor quality mains - but I was still learning about how to best deal with such, at the time ...
Yes, space can be a killer - you would laugh your head off if you saw my Perreaux, it's an unsavoury mess of extra capacitance - a 3D sculpture of bits, that I was playing with, step at a time
.The gainclone I was fully in charge of, there was plenty of space to breathe, no prob's here.
And, yes, one needs to know, as in being able to hear, that something is good enough as is - work on the bits that matter, that really alter the sound for the better - and learn, if necessary, to recognise when one is going in the wrong direction - be prepared to rip something out, discard it if it doesn't work ...
BC1's
That's very kind and you ( John C also ) . I consider myself as a torch holder like in the Olympics.
My friend Martyn is a part time sound engineer and John a real one. We went the collect some broken Spendor BC1's and had a lads day out on Jan 1st ( steak pie in 3 tons Dedington ). My back is out so I was along for the ride. Leamington is 40 miles North. The BC1's left in a porch for us. £279 broken. On the way back I speculated a broken pigtail. Now why I should think only that I have no idea. Sure enough that was it. I had 5 goes to resolder it to the voice coil stub of about 0.5 mm hard up to the cone. Martyn took it home to post to an ex Spendor engineer. No need, 6 th time lucky. He rolled th3 solder down the iron as we all know to be wrong. As when Spendor silicon rubber to make it strong. The pigtail was not dry jointed. It had broken in the best possible place. The pigtail is generous so no problems. It has been run loud since the repair. S no 10203 ( broken ) 10204.
OMG they are good. My SMGa are better and very similar. 90% of the time it is possible to imagine it is an OB speaker. Sadly 10 % of the time you can't. That 10% hits very hard as it comes from nowhere. Bass is both excellent and very pooor. I have to think the very poor is recording defects as films seem very OK ( ironic) . I do notice the same with other speakers but not to the same extent. Upper mid is slightly constricted. That is a surprise as these are genuine monitor speakers. Some suggest the money that went into them was more like NASA. BBC LS5/8 ( ? ) with STC/ITT supertweeter ( AKG microphone , 24 R coil resistance ).
The box is very light plywood with bitumin pads and low grade foam ( rubber ? ) . The drive unit looks like no other. Cast chassis in zinc alloy at a guess. Alnico magnet I am told. If so it is very large and painted in blue hammerite. Red spot sensetivity on this one. The fixing is 5 point. The use 4 mm post which is handy for testing. The crossover is very high grade but not for the sake of looking good. The back screwed on and looks slightly wrong. Wrong for a reason I am told.
The mid top is Celestion HF 1300 marked 16 ohms and measures 12.6 R DC ( Bass 7.9 ). The Celestion was a selected grade.
It is very hard to hear the box. When you do it is sudden and unwelcolmed. The Dynaco is more obviously a box which is often more OK. All I know is the impossible has almost happened here. The stereo image is not possible. The weird bit is voices could be better. This is almost certainly the source as it is true of all speakers. I would say the Spendor is slightly worse. That is because 90% of the time I use OB speakers.
Martyn was going through a very bad time. He sold his BC1's as a sort of life laundry. I was so angry as I should have bought them. Not least the stupid £350 he asked. I told him they should be worth £2000 if the shoe box LS5/5A worth > £1000. I more so beleive this now.
Martyn is 101 % able to do the work I did . One small difference is I have rebuilt a few BC1's. This pair I suspect is better than factory. I took a lot of time. However putting a light bulb in is harder. They are built by people who take no shortcuts not show off. The reason I have the honours is he has to negotiate them coming back home I suspect? The strangest part is the guy who owned them had one upside down and has scratched that side. Some wood oil has made it almost good as new. The grille was upside down also. I gently gave it a heat gun stretch.
Now the possitives. These are speakers better than most will ever know. They must be biased towards the Quad ESL 57 in sound balance. They out class the 57 and my SMGa in polar responce although not to the extent I thought. Stereo is spectacular. Organ music very real except in dynamics. The biggest surprize is how they sound great at low volume. This might be ideal for people who are not allowed to have realistic levels. MP3 is excellent with them. That was a surprise. They seem to almost love it. They make the detail as good as supporters of MP3 think it to be. At times the box is heard, bass dissapears, and slightly strident. Were it not for that I should sell what I have and have some BC1's.
In Linn Naim speak they don't have PRaT. I think I know why. They are a monitor and do not enhance. The Dynaco does. As odd as it sounds if I had to choose I would choose the Dynaco. The reason is the SMGa is better and I don't need a speaker 85% as good as the ones I own. The A25's get thorwn in the car and I never care how they look when they come out.
The BC1 reminds me of the girls my mum said were beautiful. If she hadn't I would have worked it out for myself and would have imagined the extra 15 %. It's not hype. Just that no product can ever live up the the BC1 reputation. All I can say is no way can a speaker that looks this tame do what it does. 4 mm terminal spacing is about 40 mm and not 19mm ( 3/4). For a professional product that is odd. Even Naim use 19 mm. That is because it's real designer was from Pro Audio.
How I got to do Spendor is they had just been reviewd against Tangent RS4. The Spendor had been panned. We sold Tangent and to my astonishment my boss and friend Julian Mason thought that a good reason to try some Spendor. From the word go it was a great product for us. The same boss then spotted a review of Spendor against KEF 104AB which liked the Spendor more. Same boss and same result. I love 104AB and 104 also. 104 is raw but very good. My boss had hit on an idea. Do out of fashion products. Oxford was right for that. All there dance to a different drum in Oxford.
BBC Radio 3 has a new acoustic recording at 10.30 in just under 1 hour. I will listen on Quad FM3 and record it MP3 on hard-drive as it is easier. My MP3 is very high grade. In colouration the MP3 is better.
We tried Harbeth that look almost identical. Dudley was an odd fish. He was at the BBC same time as Spencer. Similar use of names to give a product name. Martyn and Amanda Miles. Can anyone make up a nice product name as Martyn has made some LS3/5A clones. He calls his little non existant company MGM Audio. I can not think he will get away with that. Martyn had these Harbeths. What a mistake.BC1's are to us better. I sold my pair of Harbeths off for less than trade. I am told the now version £4000. I have heard the top of the range Hartbeth and suspect it better than BC1. For £279 they are not.
That's very kind and you ( John C also ) . I consider myself as a torch holder like in the Olympics.
My friend Martyn is a part time sound engineer and John a real one. We went the collect some broken Spendor BC1's and had a lads day out on Jan 1st ( steak pie in 3 tons Dedington ). My back is out so I was along for the ride. Leamington is 40 miles North. The BC1's left in a porch for us. £279 broken. On the way back I speculated a broken pigtail. Now why I should think only that I have no idea. Sure enough that was it. I had 5 goes to resolder it to the voice coil stub of about 0.5 mm hard up to the cone. Martyn took it home to post to an ex Spendor engineer. No need, 6 th time lucky. He rolled th3 solder down the iron as we all know to be wrong. As when Spendor silicon rubber to make it strong. The pigtail was not dry jointed. It had broken in the best possible place. The pigtail is generous so no problems. It has been run loud since the repair. S no 10203 ( broken ) 10204.
OMG they are good. My SMGa are better and very similar. 90% of the time it is possible to imagine it is an OB speaker. Sadly 10 % of the time you can't. That 10% hits very hard as it comes from nowhere. Bass is both excellent and very pooor. I have to think the very poor is recording defects as films seem very OK ( ironic) . I do notice the same with other speakers but not to the same extent. Upper mid is slightly constricted. That is a surprise as these are genuine monitor speakers. Some suggest the money that went into them was more like NASA. BBC LS5/8 ( ? ) with STC/ITT supertweeter ( AKG microphone , 24 R coil resistance ).
The box is very light plywood with bitumin pads and low grade foam ( rubber ? ) . The drive unit looks like no other. Cast chassis in zinc alloy at a guess. Alnico magnet I am told. If so it is very large and painted in blue hammerite. Red spot sensetivity on this one. The fixing is 5 point. The use 4 mm post which is handy for testing. The crossover is very high grade but not for the sake of looking good. The back screwed on and looks slightly wrong. Wrong for a reason I am told.
The mid top is Celestion HF 1300 marked 16 ohms and measures 12.6 R DC ( Bass 7.9 ). The Celestion was a selected grade.
It is very hard to hear the box. When you do it is sudden and unwelcolmed. The Dynaco is more obviously a box which is often more OK. All I know is the impossible has almost happened here. The stereo image is not possible. The weird bit is voices could be better. This is almost certainly the source as it is true of all speakers. I would say the Spendor is slightly worse. That is because 90% of the time I use OB speakers.
Martyn was going through a very bad time. He sold his BC1's as a sort of life laundry. I was so angry as I should have bought them. Not least the stupid £350 he asked. I told him they should be worth £2000 if the shoe box LS5/5A worth > £1000. I more so beleive this now.
Martyn is 101 % able to do the work I did . One small difference is I have rebuilt a few BC1's. This pair I suspect is better than factory. I took a lot of time. However putting a light bulb in is harder. They are built by people who take no shortcuts not show off. The reason I have the honours is he has to negotiate them coming back home I suspect? The strangest part is the guy who owned them had one upside down and has scratched that side. Some wood oil has made it almost good as new. The grille was upside down also. I gently gave it a heat gun stretch.
Now the possitives. These are speakers better than most will ever know. They must be biased towards the Quad ESL 57 in sound balance. They out class the 57 and my SMGa in polar responce although not to the extent I thought. Stereo is spectacular. Organ music very real except in dynamics. The biggest surprize is how they sound great at low volume. This might be ideal for people who are not allowed to have realistic levels. MP3 is excellent with them. That was a surprise. They seem to almost love it. They make the detail as good as supporters of MP3 think it to be. At times the box is heard, bass dissapears, and slightly strident. Were it not for that I should sell what I have and have some BC1's.
In Linn Naim speak they don't have PRaT. I think I know why. They are a monitor and do not enhance. The Dynaco does. As odd as it sounds if I had to choose I would choose the Dynaco. The reason is the SMGa is better and I don't need a speaker 85% as good as the ones I own. The A25's get thorwn in the car and I never care how they look when they come out.
The BC1 reminds me of the girls my mum said were beautiful. If she hadn't I would have worked it out for myself and would have imagined the extra 15 %. It's not hype. Just that no product can ever live up the the BC1 reputation. All I can say is no way can a speaker that looks this tame do what it does. 4 mm terminal spacing is about 40 mm and not 19mm ( 3/4). For a professional product that is odd. Even Naim use 19 mm. That is because it's real designer was from Pro Audio.
How I got to do Spendor is they had just been reviewd against Tangent RS4. The Spendor had been panned. We sold Tangent and to my astonishment my boss and friend Julian Mason thought that a good reason to try some Spendor. From the word go it was a great product for us. The same boss then spotted a review of Spendor against KEF 104AB which liked the Spendor more. Same boss and same result. I love 104AB and 104 also. 104 is raw but very good. My boss had hit on an idea. Do out of fashion products. Oxford was right for that. All there dance to a different drum in Oxford.
BBC Radio 3 has a new acoustic recording at 10.30 in just under 1 hour. I will listen on Quad FM3 and record it MP3 on hard-drive as it is easier. My MP3 is very high grade. In colouration the MP3 is better.
We tried Harbeth that look almost identical. Dudley was an odd fish. He was at the BBC same time as Spencer. Similar use of names to give a product name. Martyn and Amanda Miles. Can anyone make up a nice product name as Martyn has made some LS3/5A clones. He calls his little non existant company MGM Audio. I can not think he will get away with that. Martyn had these Harbeths. What a mistake.BC1's are to us better. I sold my pair of Harbeths off for less than trade. I am told the now version £4000. I have heard the top of the range Hartbeth and suspect it better than BC1. For £279 they are not.
My friend Martyn is a part time sound engineer and John a real one. We went the collect some broken Spendor BC1's and had a lads day out on Jan 1st ( steak pie in 3 tons Dedington ). My back is out so I was along for the ride. Leamington is 40 miles North. The BC1's left in a porch for us. £279 broken. On the way back I speculated a broken pigtail. Now why I should think only that I have no idea. Sure enough that was it. I had 5 goes to resolder it to the voice coil stub of about 0.5 mm hard up to the cone. Martyn took it home to post to an ex Spendor engineer. No need said:
Is that the thick paste with a distinctly silver colur? You are supposed to speard it around conatc points (e.g. RCA Cunch lkacks, male and female) to get an even better contact?
If so, that's just what I threw away the other day. Nobody told me what happens when the paste dries up, that it then creates a semi hard layer around what it was spread on and you're not going to get it off nohow. Rubbish, stay away from that gizmo.
Dejan, I have been an enthusiastic user of silver glues and greases for many years now, IME it's the best solution if you can't, or won't hardwire a connection. Like all processes, it has to be done correctly - sloppy implementation is worse than no treatment, one has to be as precise as making a fine soldering connection.
Main rules are: scrupulous cleaning of the surfaces before applying; absolute minimum of the paste to be applied, get rid of excess at the time of application; and once the first - no second go's! - connection is made, DO NOT DISTURB! If for whatever reason the surface is disturbed, consider the connection to be bad - go through the whole procedure again.
Frank, I use nothing but the highest silver content soldering wire I can find, currently a UK product and DAMN good too,allegedly with 9% of silver in it, but this is a paste you apply over your contact jacks. You do not solder it, just rub it on. And eventually it dries, creating a hard compact film on what it was applied to. An entirely different matter.
Leaving you with a problem of how to clean it up. Not a small problem, I might add. I say it again, beware of that solution. In addition, when applied, I did not hear any improvements anywhere.
As for cleaning the surfaced to be soldered, I usually rub them with some very fine sandpaper to remove any oxydised coponents, then rub it wit a cotton swab dipped in medicinal alcohol and finally wipe again with dry cottom swab. So far, not a one solder point has given up the ghost with me, which I take to be the payback for taking all that trouble. Shiny, clean transistor pins are easy to solder on well.
So far, NOBODY has answered my question what happens when it dries up and hardens?
And remember, I am the guy who keeps harping on about pure silver cables here. I do not need any convincing about the qualities of silver as a conductor.
Last edited:
- Status
- Not open for further replies.