Naim sound quality... old/new... quasi/comp... a few thoughts:
It is interesting to note that Avondale has moved from quasi to complementary o/p stages e.g. NCC200 to NCC220. Sound quality is allegedly improved. They have also "broken" the single o/p transistor rule with the NCC300. These amps have a very strong following! The target for the NCC300 is to equal or beat the sound quality of the NAP300.
Earlier in the thread some people opined that the JV era Naim is better sounding than the new stuff and that the new stuff sounds "dark". The NAP 200 seemed to be included in the "old sound" category, even though it's much later than the JV era.
Personally, I would take a NAP300 over anything I've heard from the JV era. By any standards, the NAP300 is a great amp (with its associated preamp).
The older stuff certainly has more character and in some situations can sound more pleasing. e.g. at low listening levels the "loudness" effect and the slightly "dirty" sound of old Naim can help to flesh out the music. I also believe it favours some types of music over others.
I guess , in the end, many people don't want a "straight wire with gain" as evidenced by the popularity of this thread and amps like the NHB108 (which I'm tempted to build).
It is interesting to note that Avondale has moved from quasi to complementary o/p stages e.g. NCC200 to NCC220. Sound quality is allegedly improved. They have also "broken" the single o/p transistor rule with the NCC300. These amps have a very strong following! The target for the NCC300 is to equal or beat the sound quality of the NAP300.
Earlier in the thread some people opined that the JV era Naim is better sounding than the new stuff and that the new stuff sounds "dark". The NAP 200 seemed to be included in the "old sound" category, even though it's much later than the JV era.
Personally, I would take a NAP300 over anything I've heard from the JV era. By any standards, the NAP300 is a great amp (with its associated preamp).
The older stuff certainly has more character and in some situations can sound more pleasing. e.g. at low listening levels the "loudness" effect and the slightly "dirty" sound of old Naim can help to flesh out the music. I also believe it favours some types of music over others.
I guess , in the end, many people don't want a "straight wire with gain" as evidenced by the popularity of this thread and amps like the NHB108 (which I'm tempted to build).
I like the way the seeming contradictions in the original Naim design give rise to interesting discussions.
NPN and PNP use different semi-conductor technology. In this universe we cannot use positrons in our circuits. Nature is asymmetrical. So it seems to me that quasi-complementary is the more symmetrical and more natural. Fully complementary is only more symmetrical in its schematic symbolization.
Not everyone appreciates this. Nor do crafty semiconductor manufacturers help by claiming complementary transistors in data sheets. What exactly do they mean by "complementary"? It never means electrically equivalent and reverse polarity. No, no, no. So be wary of this. Especially with MOSFETs. NPN and PNP are different technologies.
Don't get me started on Avondale.
NPN and PNP use different semi-conductor technology. In this universe we cannot use positrons in our circuits. Nature is asymmetrical. So it seems to me that quasi-complementary is the more symmetrical and more natural. Fully complementary is only more symmetrical in its schematic symbolization.
Not everyone appreciates this. Nor do crafty semiconductor manufacturers help by claiming complementary transistors in data sheets. What exactly do they mean by "complementary"? It never means electrically equivalent and reverse polarity. No, no, no. So be wary of this. Especially with MOSFETs. NPN and PNP are different technologies.
Don't get me started on Avondale.
But we are not talking here about single N transistor vs. single P transistor. We are talking about composite supertransistor and it seems to me that Sziklai supertransistor is more linear than Darlington supertransistor or at least they have dissimilar characteristics when used in push pull o/p stages, more dissimilar than fully complementary Darlingtons. That's what I heard. I have no formal education in electronics at all, so I could be wrong. But, after all, what matters is perception and I admit that quasi can sound "right" to me or some other people in spite of what appears to be shortcoming.
Member
Joined 2009
Paid Member
The gain at cross-over is asymmetric for quasi-comp - this was published by D. Self many years ago. I can't see how it's more symmetrical. And how can it be more natural - especially when the quasi-comp as used by Naim relies on the use of a p-type device. If you look at the drivers for the output transistors you see it's almost a complementary output in that regard. It seems to me that the output stage is in fact two right hand shoes forced onto a left-right pair of feet
- very natural my nelly !
Last edited:
I've been reading and listening to arguments for quasi technology since the time it was the only way and comparably good PNP power transistors were rare unobtanium. Japan led the way to low distortion complementary design and the established manufacturers in the west soon felt the pressure to either follow suit, re-badge the cheaper and better products or quit audio manufacturing completely.
Whether PNP transistors are inferior components because of their different technology is, for better or worse, history because few, if any large scale audio manufacturers have used quasi designs for 40 years. The reason, primarily, is to remain competitive in the low distortion race and if you want minimum distortion, complementary is demonstrably the way to go. That is why arguments for the superiority of quasi-comp. tech will continue to fall on deaf ears.
'Quasi' gives us higher distortion because as the 2 halves of the output stage are now abruptly dissimilar in the crossover region, the harmonics generated are larger and tend to be even. This a has a better outcome than it might seem since if we're going to have higher harmonic distortion, we'd naturally prefer it to be euphonic.
Make no mistake about it, many of us do prefer sound that is coloured with euphonic distortion products and we get more from it in personal listening, relaxing with our favourite music where we focus on the recording's detail and nuances of difference between the electronics, speakers etc in our audio systems.
The increasing numbers of people that flock to JLH, Naim and AKSA's threads tell you how bored people are with ultimate high performance designs that are more an engineering puzzle to discuss in simulation than a simple, affordable project that more than satisfies the average DIY. That's why I like the real-deal Naim sound in preference to my meticulous blameless style designs which are clean as a whistle but boring as hell
Whether PNP transistors are inferior components because of their different technology is, for better or worse, history because few, if any large scale audio manufacturers have used quasi designs for 40 years. The reason, primarily, is to remain competitive in the low distortion race and if you want minimum distortion, complementary is demonstrably the way to go. That is why arguments for the superiority of quasi-comp. tech will continue to fall on deaf ears.
'Quasi' gives us higher distortion because as the 2 halves of the output stage are now abruptly dissimilar in the crossover region, the harmonics generated are larger and tend to be even. This a has a better outcome than it might seem since if we're going to have higher harmonic distortion, we'd naturally prefer it to be euphonic.
Make no mistake about it, many of us do prefer sound that is coloured with euphonic distortion products and we get more from it in personal listening, relaxing with our favourite music where we focus on the recording's detail and nuances of difference between the electronics, speakers etc in our audio systems.
The increasing numbers of people that flock to JLH, Naim and AKSA's threads tell you how bored people are with ultimate high performance designs that are more an engineering puzzle to discuss in simulation than a simple, affordable project that more than satisfies the average DIY. That's why I like the real-deal Naim sound in preference to my meticulous blameless style designs which are clean as a whistle but boring as hell
It is better that it favors some type of music over others than sounding boring with ALL types of music, isn't it?
I also think that when company decides to produce ultra linear (=boring) products, tone control preamplifier is a must! That is the only way to give some character to sound and make it tasteful. But most companies that produce such amps don't give the user that chance. Therefore user is left helpless and the only thing he can do is sell the product and choose some other product. Only Japanese producers understand that and that's the reason Japanese producers fit tone control on amps of spectacular quality (like Accuphase, Luxman, Yamaha). At least you have a chance to make them sound good if you feel that flat position is not sounding good.
Last edited:
The gain at cross-over is asymmetric for quasi-comp - this was published by D. Self many years ago. I can't see how it's more symmetrical. And how can it be more natural - especially when the quasi-comp as used by Naim relies on the use of a p-type device. If you look at the drivers for the output transistors you see it's almost a complementary output in that regard. It seems to me that the output stage is in fact two right hand shoes forced onto a left-right pair of feet
Yes. Quasi complimentary outputs have different gain, impedance and frequency response on the positive and negative halves of the output voltage swing. The negative side requires a RC load to stabilize the driver-output feedback but this is wasted on the NPN+NPN follower. A good analogy is having two right hands, two left feet etc. ie not symmetric.
Last edited:
Not if competently designed.
You might as well say Donald Trump published it. Don't get me started on Self.
Nature was born with only right feet, electrically. The Naim quasi is not perfect, as you say, because a pnp is required but mitigating the imbalance in drivers is easier than in power transistors.I can't see how it's more symmetrical. And how can it be more natural - especially when the quasi-comp as used by Naim relies on the use of a p-type device. If you look at the drivers for the output transistors you see it's almost a complementary output in that regard. It seems to me that the output stage is in fact two right hand shoes forced onto a left-right pair of feet
Tube amps avoid this issue as they are only available in one type. Because of Mother Nature.
Last edited:
Mosfets are different... they are basically a variable resistor and don't care which terminal is which, therefore they are (possibly) more suited to Quasi operation.
Good question, and only the beholder can decide.
I can certainly understand the attraction of amps llike my NCC200 clone, but it's not my favourite amp overall. Neither is my MyRef with its loads of zeroes distortion figures.
Then there's the question of how much of the real Naim sound do we achieve? IME the power supply makes a huge difference to the sound.
Then yet another question as to what is the Naim sound? Some of their amps sound very different to others... just change the PS for the preamp slightly and the sound can change dramatically.
Traderbam doesn't like the Avondale version, but there are many 10s of people who rate them very highly and certainly above Naim equivalents. I've only read a couple of examples of people who prefer Naim, or perhaps say it's a swings and roundabouts thing.
I have owned NCC200s twice and I'm one of the few in the swings and roundabouts camp.
But they use a totally different technique to work around this i.e. a transformer.
My aluminium bars have arrived. Now there is one thing which is still not clear to me: Can I use ONLY grease between the heat sink bar and the transistors? Is there no insulation needed? I also have these rubber/silicone pads. But if I use these, then I shouldn’t use grease (that’s what I read here). I don’t have mica washers. So: Only grease or only rubber pads? The grease is HY883. Thermal conductivity >6.5W/m-k.
It isn't clear in the datasheet but the metal area on the back of a transistor is usually connected to its collector. You could try confirming this with a multimeter. If so, you definitely need to electrically insulate every transistor from the heatsink bar and therefore use rubber without grease.
Thanks!! Yes, the the back is indeed connected to the collector, I just checked this with my multimeter. Good to know
Attachments
in fact, it depends on the case of the transistor.Thanks!! Yes, the the back is indeed connected to the collector, I just checked this with my multimeter. Good to know
some need a "tube" insulation and other no but most have the sole
connected to the collector
OK, so now ALMOST everything seems to work perfectly and it sounds very very good, BUT: I have also built this NAC-42 preamp clone (Chinese kit as well, same seller). It gets its 24V right from the NAP-200 (4-pin connector). When the preamp is connected I always have some hum. The hum gets amplified as I turn the preamp volume pot, but is also present with the volume on 0. When I disconnect the preamp boards completely and connect the source directly to the NAP-200 then there is no hum at all.
The enclosure of the preamp is connected to signal ground/24V ground. The NAP-200 enclosure is connected to signal ground via 2 diodes in reverse and a 10 ohms/5W resistor. But it makes no difference at all if I connect it directly to signal ground instead. Only if I don’t connect it to signal ground at all (but still to IEC socket earth of course) then it’s also humming (even louder), no matter if the preamp is connected or not.
So now I don’t know if the hum comes from the 24V supply in the NAP-200 or from somewhere in the preamp?
The enclosure of the preamp is connected to signal ground/24V ground. The NAP-200 enclosure is connected to signal ground via 2 diodes in reverse and a 10 ohms/5W resistor. But it makes no difference at all if I connect it directly to signal ground instead. Only if I don’t connect it to signal ground at all (but still to IEC socket earth of course) then it’s also humming (even louder), no matter if the preamp is connected or not.
So now I don’t know if the hum comes from the 24V supply in the NAP-200 or from somewhere in the preamp?
What is quasi distortion profile? How does it differ to complementary distortion profile?
ivanlukic said:
What distortion? There is reason why second order distortion is liked by people. There is reason why monotonic declining distortion could be a good thing (for this, you will never find the answer in books or mentioned on the net).
You can get similar Naim's distortion with other topologies. Distortion is Naim's weakness, not it's strength.
ivanlukic said:
Explain scientifically how high linearity equates to boring sound?
There is no problem with linearity. Problem is, it's the only thing designers think responsible with sound quality. It's ironical. They believe that very low distortion is inaudible yet they chase ultra low distortion like crazy, do not understand what they have compromised.
your nac 42.5 must be powered in 18 vac if you have kept the diode bridge
in place on the pcb.
otherwise, and if it's like NAIT, you need to have a small diode bridge on the NCC pcb coming out of 24vdc as shown in this picture
