here there is no talk of preferential trails taken by electrons, but of well-traveled signal and with a few selected excellent components
I think
and when in a final I start to see input cap, signal-mass cap in the feedback loop, various compensations in a circuit already of its simplest, well, the circuit does not look like much performante
these Naim are simple, almost basic, can be cloned without problems spending very little because the transistors are always replaceable, but say they sound good, in short, I would say no
I too cloned the 140, and ......
I think
and when in a final I start to see input cap, signal-mass cap in the feedback loop, various compensations in a circuit already of its simplest, well, the circuit does not look like much performante
these Naim are simple, almost basic, can be cloned without problems spending very little because the transistors are always replaceable, but say they sound good, in short, I would say no
I too cloned the 140, and ......
Last edited:
Member
Joined 2009
Paid Member
I think you are missing the point if you feel the schematic is too basic or flawed. It is the very flaws that that are important. If you try to improve on them, you no longer have what the designer intended. You might not like what the designer intended, that's fine too, just move on to something else - as somebody else on this forum said once, there's for sure an amplifier available for each persons tastes.
I tried a higher voltage on my clone. Interestingly, I don't think it was for the better. And I had a similar experience with my AKSA clone. I don't have them set up at present so I can't re-check - we're relying on my memory. Nevertheless, my perception is that these amps sound sweeter in the range +/-36V to +/-42V than at anything higher. I have other amps that are happy at higher voltages.
I tried a higher voltage on my clone. Interestingly, I don't think it was for the better. And I had a similar experience with my AKSA clone. I don't have them set up at present so I can't re-check - we're relying on my memory. Nevertheless, my perception is that these amps sound sweeter in the range +/-36V to +/-42V than at anything higher. I have other amps that are happy at higher voltages.
Last edited:
It's not like many others haven't already tinkered and tried everything possible with NAP designs either. Popularity and recognition of the Naim brand is a strong magnet for audiophiles, curious folk and other manufacturers - all for good reasons. If our own efforts are to be meaningful though, we need to avoid the usual DIY stumbling blocks of guesswork, observer bias, poor test equipment and lacking any reference or even a standard amplifier to make proper comparisons.
I would spend some time reading the Naim and DIY forums at pink fish media before guessing about possible modifications that would probably suit mass produced products rather than the tightly specified Naim products. This useful reference doc. for old Naim models also originates at PFM: Modifying Naim Audio power amplifiers
I would spend some time reading the Naim and DIY forums at pink fish media before guessing about possible modifications that would probably suit mass produced products rather than the tightly specified Naim products. This useful reference doc. for old Naim models also originates at PFM: Modifying Naim Audio power amplifiers
If you look at the ratings of the ZTX653/753 VAS transistors, you'll notice that the rail-rail voltage limit is around 80V with a safety factor. There are few suitable transistors other than these and their SMD versions, than can match these types for the VAS transistor and current sink, where a high Cob of 30pF is clearly wanted. So, for practicality, it's ~40V max. rails or move on to something else, as Bigun suggests.
The NAP design is based on a generic type anyway and the simplest version is a fully worked example seen in Douglas Self's books and on his website, in an article examining distortion in amplifiers. Check it out if you like to tinker with more possibilities. Distortion In Power Amplifiers - Home Warranty Appliances
Most people don't have good ears. Those who have, mostly don't understand what causing the difference in sound. Almost any topology can sound special if it meets certain conditions. Knowing this conditions is not something easy to do. It is not as simple as it looks.
If you haven't heard the original 140, don't be sure it will sound the same with your clone. For sure it will not be the same sound. Tell me something, do you know the hfe of the original Naim output transistor? Do you know the hfe of your clone? Is bottom NPN has the same hfe with the top one?
Say you have heard the original 140. If you have good ears, you should be able to hear where the amp excels and where it is lacking. It is then up to you if you think the amp is not up to your taste.
IMO, what you need is to improve the distortion (without changing what it is already good at). With this basic circuit it can be expensive. And you know what? Think about this:
Loudspeaker distortion is often much greater than amplifier distortion. Some amps, especially with special drive like Naims, can easily excite speaker cone distortion. Some other amps, like P3A, will not do that. Take a raw high distortion speaker driver and compare the 140 with P3A. The result is like heaven and earth, the P3A will be so much distortionless. Naturally, you will prefer the P3A right? Of course. Now what if you have a GREAT low distortion speaker? What you think is great with P3A might not great anymore because now you are okay with 140 + your speaker distortion.
the 140 can be interesting for its simplicity and economy of construction having also dissipates little heat
the 140 can always be replicated, clear, no clone will ever be the same original that even the same Naim, today, not could repeat faithfully
apart from the basic scheme, the Naim sound is also defined by the tantalum capacitors used in the finals and pre-English of the time, in addition to the obvious filter capacitors that are nowadays unobtainable.
Who ever decides to clone the 140 will know with absolute certainty that it will never be able to replicate the sound and that even the original 140 will sound, today, very different from the new original.
That said, the 140 seems excellent to be cloned, within today's limits, spending little.
example, Spectral DMA 50, another world compared to 140, it is almost impossible to clone faithfully, but you can redo
the 140 can always be replicated, clear, no clone will ever be the same original that even the same Naim, today, not could repeat faithfully
apart from the basic scheme, the Naim sound is also defined by the tantalum capacitors used in the finals and pre-English of the time, in addition to the obvious filter capacitors that are nowadays unobtainable.
Who ever decides to clone the 140 will know with absolute certainty that it will never be able to replicate the sound and that even the original 140 will sound, today, very different from the new original.
That said, the 140 seems excellent to be cloned, within today's limits, spending little.
example, Spectral DMA 50, another world compared to 140, it is almost impossible to clone faithfully, but you can redo
Last edited:
ask if anyone has tried to feed the clone 140 to 50 vcc, mean to ask what changes have been made (I know what to do, but I ask if others have already walked the way).
Obviously you have to intervene on the transistors and change them with, I do not know, but I would say 2n 5551 and 5401 that seem optimal and final like the 1302 A and complementary or similar, since the 1302 are out of production
Obviously you have to intervene on the transistors and change them with, I do not know, but I would say 2n 5551 and 5401 that seem optimal and final like the 1302 A and complementary or similar, since the 1302 are out of production
today redoing a final amp 20 years ago is often impossible because there are no more important transistors (see i jfet 2n 5566)
So a diy is forced to repay on simple schemes and projects and to do what you can with what is around from serious dealers such as Mouser, Distrelec, Farnell, RS:
others are almost always cheats
it happens that on E BAY many sites in the Far East, to sell, falsify the codes of j fet and transistor
So a diy is forced to repay on simple schemes and projects and to do what you can with what is around from serious dealers such as Mouser, Distrelec, Farnell, RS:
others are almost always cheats
it happens that on E BAY many sites in the Far East, to sell, falsify the codes of j fet and transistor
Last edited:
I think you got the picture. Thus, it is important to know how things can sound differently and why. It doesn't matter if the newer Naim is using completely different topology. What matters is they know what they are trying to achieve and how. Take for example the uncommon compensation method they are using. Such method has benefit over others which tend to affect sound negatively. But even so, many people don't understand how it works and how to use it. In the newer Naims, the topology is different but they still use the same technique. Also the same is their choice of the VAS transistor. It is a unique amplifier in itself.
If you make such drastic change, even tho the topology is the same, it is your own design. The 140 is a good amp built over a poor topology. It took time to get to the good result.
Why redo an old amp if it is not a better design? It's true that current products from China have poor sound quality. Many are fakes that measured out of spec. Some others comply with basic measurement specs, often better than original, but when you listen to it,
So for your own music system, purchase now the high quality parts while you can (and those from US are fine, btw). Even one set can last forever if you're lucky
And choose good amp based on good topology.I can't to discuss Naim's technical choices, but I not share them because I see very basic schemes .
But basics schemes are also for the result to listen, because if the basic gave me great performances then it would be different
I look at the 140 cloned for the reasons repeatedly exposed, not to redo the Naim sound that I don't like much
Last edited:
true, it would become my scheme, but I would have a few more watts, better if doubling the power pair from 1 pair to 2
I delete the input capacitor or I use a Mundorf type, I see if with a dual BJT input and a trimmer on the emitter I can better control the offset (I just do not understand why placing the 2 transistors of the IN differential one here and there there else, not even in thermal contact)
I would also make it stabilized in the entry stages, drivers and finals
I could get a good result with a few €
Why redo an old amp if it is not a better design? It's true that current products from China have poor sound quality. Many are fakes that measured out of spec. Some others comply with basic measurement specs, often better than original, but when you listen to it,
So for your own music system, purchase now the high quality parts while you can (and those from US are fine, btw). Even one set can last forever if you're lucky
I look at the 140 because it would cost little, it is easy to copy, but especially because I just can not find a few 2N 5566
I had found them I would have cloned the Spectral DMA 50 or Norma 8.7 B mono
The acoustica.org doc is a great resource to sketch out the various areas that need to be thought about. However, I'd caution that some of the conclusions in it appear to suffer from the stumbling blocks mentioned above.
I see a lot of folks attracted to the Naim design for its simplicity and cost-effectiveness. I won't comment on the latter because I don't pay much attention to that, but I think it would be a grave mistake to classify the Naim design as simple.
While the basic topology is simple-ish, there's nothing simple about the compensation schemes chosen, or the effects of the unbalanced LTP and VAS requirements. The feedback in the Sziklai pair is tight, but even that can get you into trouble. I can't recall another popular DIY design that's on such a stability knife-edge.
As a case in point, take the common preference for a MJ15003 output stage. Is that because people actually like the sound of a slow output stage better (JV certainly didn't), or because they like the sound of a non-oscillating amp better?
If you want a simple starter project look to Pass, JLH, etc. If you're a bit higher up the curve and want to play around with a similar topology, the P3A is probably a better starting point. If you want to experiment with the somewhat esoteric design decisions in the Naim circuit, or take a trip down memory lane, then by all means have at it -- just don't go in thinking it's simple.
While the basic topology is simple-ish, there's nothing simple about the compensation schemes chosen, or the effects of the unbalanced LTP and VAS requirements. The feedback in the Sziklai pair is tight, but even that can get you into trouble. I can't recall another popular DIY design that's on such a stability knife-edge.
As a case in point, take the common preference for a MJ15003 output stage. Is that because people actually like the sound of a slow output stage better (JV certainly didn't), or because they like the sound of a non-oscillating amp better?
If you want a simple starter project look to Pass, JLH, etc. If you're a bit higher up the curve and want to play around with a similar topology, the P3A is probably a better starting point. If you want to experiment with the somewhat esoteric design decisions in the Naim circuit, or take a trip down memory lane, then by all means have at it -- just don't go in thinking it's simple.
Yes, the Naim design is sophisticated. There is more to it than the schematic and parts list. You can't truly clone one without cloning the part selection knowledge too, which is not in the public domain. I have never seen a credible clone.
My experience is that a simple copy makes a bit of a mid-fi circuit. Among the sound quality of a NAD or Arcam or Cambridge Audio, to name a few. But the NAP is attractive for DIY because it has relatively few parts and runs cool.
My experience is that a simple copy makes a bit of a mid-fi circuit. Among the sound quality of a NAD or Arcam or Cambridge Audio, to name a few. But the NAP is attractive for DIY because it has relatively few parts and runs cool.
the design scheme Naim does not seem sophisticated, it seems rather, as indeed it is, taken from a Philips application notes of half a century ago
Then Naim climbed to make it stable, and the scattered compensations confirm it, not succeeding well: in fact, with power cables not Naim the final wavered
little or no attention to stabilize the offset if not sowing condensers everywhere
Spectral has a sophisticated and studied scheme, Acoustat Twin 200 as well, but they are irreproducible
Then Naim climbed to make it stable, and the scattered compensations confirm it, not succeeding well: in fact, with power cables not Naim the final wavered
little or no attention to stabilize the offset if not sowing condensers everywhere
Spectral has a sophisticated and studied scheme, Acoustat Twin 200 as well, but they are irreproducible
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.