Member
Joined 2009
Paid Member
It has been my experience that Cdom has a disproportionate impact on the perceived sound of an amplifier and offers at least one way to 'voice' an amplifier, staying within stability requirements. It follows that the VAS transistor choice is equally important. For this TGM10 design I'll be using the same VAS device and Cdom value as used by Naim (as far as I can tell).
Tantallum capacitors have their own distortion behaviour and I feel it is important to use them where Naim used them. We can of course do better, but that's not the objective here.
It is really very easy to produce an amplifier with very low distortion these days whether using BJT's, FET's or both. There are many examples on this forum alone. I don't give credence to the idea that minimal distortion as a goal would lead anybody to choose either the BJT or the FET as a preferred device. Both have their pro's and con's and I would suggest a designer should pick which they are most comfortable being able to implement well. We have equally seen that an AB amplifier can be made with such low distortion as to make interest in Class A 'almost' moot. I say 'almost' only because there are things about audio amplifier satisfaction that we do not yet have a firm grip on in terms of understanding why each of us likes a particular sound.
Most of these so-called feed-forward amplifiers are all dominated by negative feedback. This includes the Quad amplifiers. The issue is that these amplifiers have to have low impedance outputs so they must use feedback to obtain their 'voltage drive' for the speaker. You can not listen to one of these amplifiers and conclude you are hearing feed-forward, you are hearing a complex amplifier but one with gobs of negative feedback for the most part.
My TGM8 amplifier uses FETs as 'current providers' operating in Class C, with BJTs in Class AB, all in a negative feedback loop. I favour this approach over the current-dumping topology.
Most people, when they hear several amps blindfolded, they will say "all sound very good" or "they sound the same". For me its different. Each amp sounds different. But this could be confusing, knowing that there could be infinite numbers of possibilities with the sound. But once those differences can be related with the design then it became clearer. We're not anymore playing with uncertainties.
Just like in design, when we hear amps, we hear what we like in each amp. In amp A we like x, in amp B we like y, and so on. So it is not like liking amp A as a whole. Naim for example, I have only heard from youtube but I know enough that I dont like it as a whole. But if I hear it directly may be I will find what I like. May be it is something that im already familiar with. And may be I already know what make it sound that way.
So collecting all the goodies and put them in one amp, thats the perfect amp im after. But this is almost impossible as the things controlling thess are trade off of each other in a design. The best thing I can do is to create threzholds for each of the variable, including THD, in which many amps with this Naim topology failed. But the trick with other topology that allows ppm distortion is, I can trade a lot of distortion for having the other goodies!
Problem is, I believe that transistor is critical. We cannot just pick any arbitrary transistor and expect miracle. If we dont want compensation caps, we have to plan the stability with choosing the right/matching transistor at each stage. I think one transistor is already very complex inside that sometimes they seem to have their own sound.
Yes. But I have built some clones. Naim is I think commercially not popular worldwide.
And please don't say you cannot hear anything over the lousy audio performance of youtube videos, unless you do have good ears. If you do have good ears, I believe you will not just copy those comments we already know from amateurs on the net, because you know the possibilities.
Maybe I didn't understand your statement here... You must be kidding, right!?
It's like these commercials that show you how much better some new TV set picture is compared to yours, while watching it on your TV 🙂
No good ears will help you with smartphone mic recording in unknown room acoustic. Boombox in a bathroom is my definition for some of them ...
I think that Naim is one of the most commercially successful audio companies in the recent years. They run real business, research, production unlike some other audio "boutique" one-man-show companies. I personally find them actually too commercial to my liking (read deceptive), but there is no doubt that they are very well known on the so called hi-end market.
Last edited:
No.
Never seen that. The idea is that a little information can be useful for some and a lot of information can be useless for some. What if you had an experience watching TV A and B and then watch them through TV C. Then you watch TV D and E through TV C. Will you have any information to tell how good D or E is? May be (I don't have such experience).
It's not just about good ears. It's also about understanding. Those who has no experience with room acoustic, presented with a sound, will have no clue. But those who knows, will be able to sort out the irrelevances.
It is also important to know what to hear in a sound. If we're only trying to hear how smooth the trebles, how low the bass, how good the soundstage (which is what the amateurs do, isn't it), then IMO even in real listening those people will not be able to understand and their preference will always be subjective and changing (What if the music is different? What if there are more instruments?)
Not disagree. In simple words I was saying: "I have never seen a Naim in local audio shops, not in nearby countries."
Too many "if"s, man!
I'm old, old, old and not interested in theorizing life. Less "if", more "do" is my motto 🙂 If we start exchanging "if"s there won't be end to that thread.
These pages are more about informing and discussing technical matters. I'm sure anyone has some philosophy on the world (and sound) to share, but considering the distance and the different backgrounds... who cares?!
I, personally, am the most proud with my system, when completely random guest in my house says something good about how it sounds... All no-Naim gear in the last years, by the way (pun intended). And I definitely wouldn't consider a person to be sane if they give me some very detailed dissection and educate me on the understanding the sound... of my system. It hasn't happened yet - audiophiles around are rare animals, but you never know.
I'm old, old, old and not interested in theorizing life. Less "if", more "do" is my motto 🙂 If we start exchanging "if"s there won't be end to that thread.
These pages are more about informing and discussing technical matters. I'm sure anyone has some philosophy on the world (and sound) to share, but considering the distance and the different backgrounds... who cares?!
I, personally, am the most proud with my system, when completely random guest in my house says something good about how it sounds... All no-Naim gear in the last years, by the way (pun intended). And I definitely wouldn't consider a person to be sane if they give me some very detailed dissection and educate me on the understanding the sound... of my system. It hasn't happened yet - audiophiles around are rare animals, but you never know.
I couldn't put it better but the obstacle many audio fans and DIYs around the globe face, is the products are not available or affordable in their region, to experience and compare in reality.
Still, we cannot compromise standards simply because we don't have the goods, skills or system to perform tests credibly. Any opinion based on arbitrary internet sound quality, falls into the realms of its heavy psycho-acoustic sound processing which prevents comparisons of quality audio gear, even for the imaginative DIYs with good ears.
Have a listen to Naim's own promotional MP3 recordings. These are in high quality commercial You-tube format if your connection and device has the capability to receive it: https://www.youtube.com/user/naimaudio/videos
The amplifiers featured are newer products with less emphatic "sound quality" than earlier models like the NAP140 so there is still no real comparison to anything you may read here. You can't tell anything about the amplifiers either. It could be line-level audio from any low-noise source and unless I was visually prompted (a clue to a pointless evaluation) I would just be guessing - like the guys who register "like" for the awful sound from the audio systems that other guys upload to the 'net 🙄
Very good points as usual, Ian.
I plan to do an experiment. There are these video/audio clips on youtube that people make with Apogee One dongle. I call it dongle, because I don't know how else to call it. Like pocket home studio for 200-300 bucks! These videos sound terrific through headphones. Actually, better than many commercial recordings that I have. So, I want to download some and to play them through my audio system. I'm curious to see if they'll sound as good as on headphones.
I find that on headphones anything is good or very good. Maybe because the room acoustic is removed from the equation.
Deutsche Grammophon also try to sell me recently MP3 files that they claim are indistinguishable from a CD quality. I still don't have the courage to buy some, although to me a well recorded CD is everything that I need from a digital format. If they really match that, I'll switch to MP3 tomorrow.
BTW, I have some CDs issued by Naim Records and they all sound very, very good. Most places where I've heard Naim gear use them as demo records. Still, they don't represent what a "regular civilian" will hear from their regular recording collection through their Naim gear. I'm sure it won't be that impressive.
I plan to do an experiment. There are these video/audio clips on youtube that people make with Apogee One dongle. I call it dongle, because I don't know how else to call it. Like pocket home studio for 200-300 bucks! These videos sound terrific through headphones. Actually, better than many commercial recordings that I have. So, I want to download some and to play them through my audio system. I'm curious to see if they'll sound as good as on headphones.
I find that on headphones anything is good or very good. Maybe because the room acoustic is removed from the equation.
Deutsche Grammophon also try to sell me recently MP3 files that they claim are indistinguishable from a CD quality. I still don't have the courage to buy some, although to me a well recorded CD is everything that I need from a digital format. If they really match that, I'll switch to MP3 tomorrow.
BTW, I have some CDs issued by Naim Records and they all sound very, very good. Most places where I've heard Naim gear use them as demo records. Still, they don't represent what a "regular civilian" will hear from their regular recording collection through their Naim gear. I'm sure it won't be that impressive.
Last edited:
Well, you can remove all the "ifs" which is used to make it more understandable 🙂. In short:
(1) No kidding.
(2) A little information can be useful for some and a lot of information can be useless for some.
(3) It is also important to know what to hear in a sound.
What are your amplifiers? There must be 300B hiding somewhere I believe? If you're happy with your system, I think that is the most important part
.What if the arbitrary opinion is based on real listening in controlled environment? Will the opinion matters to you? To me, they all have no meaning, unless the source of the opinion is known to me.
Listen to this channel where people comparing Bifrost vs Lumine, and tell me, do you think the comment regarding Bifrost preference is correct or wrong?
https://www.youtube.com/watch?v=1Fr9vV2t4BM
Music is about damping, phase etc. When there is a mistake, it will show up regardless of room effect, mic quality, computer soundcard, cheap headphone, etc. etc.
And this is what I like from a sound (listen to the song carefully if you're familiar with Dire Strait) and read the comment 😀. You cannot have this kind of sound from 2-way like Bifrost:
https://www.youtube.com/watch?v=HC-mViVQgIw
Now what is it that you can never heard from a usual 2-way speaker, beside the low bass?
The interesting mystery of the Naim amplifier is how little it changed. If I owned Naim the company it would change every week even if the prototypes never moved to production. It sort of supports the view of others that Julian didn't have the ability to move on or doubted he could. Whatever it was he kept the design and only thought of how power supplies could improve. I speculate if he had owned LAB 47 it would sound like a Naim very soon. I know one of the guys from LAB 47 and it rates very highly . You guys could clone that as it is all explained. And it would help your Naim clones.
As for low distortion of the Hitachi design. It comes with minimal complexity ( as LAB 47 ) and makes a lie of many statements made by others who almost seem to have made it a quest to decieve us. What FET's do not have is the slam as we called it of the Naim amps. However nor did a Sugden nor the Quad amps using Naim like devices.
The Naim sound to me is this. Projection of rhythm which slightly hides the recording engineers work. In my mind a good trade off. It can go wrong. Some Naim systems can sound limp and slightly sterile. We are talking real ones and NAP250 at that. I have no idea how that happens. What it shows is the average Naim dealer was a real music lover and Naim would have put that very high on the list of appointing a dealer. That dealer would work hard to ensure the Naim sound was had. Time goes by and non Naim dealers get involved and the magic is lost. Was it ever there? I think so. If you want it at it's best you want Nait ( 2 is fine ) and Quad ESL 57. It will take on anything and is the only example I know of the doesn't hide the recording engineers skills, that is space and interplay in lets say Beethoven. I always specualted that the correction networks we see in the NAP140 circuit are for the ESL 57, JLH sort of say similar things although doing it a different way. That Nait 57's combination will see off nearly anything at any price. With a little care it will remain safe and give bass. NAP250 suitably adjusted is good with 57's, not the vast difference over the Nait that one would think. Moving coil speakers are each on it's own merit. Spendor BC1 works and Spendor the company liked Naim. It won't be the usual Naim sound, it is a very musical one and will shame many so called accutate set ups. The Naim SBL has a very nice balance as do the cheaper versions. I seem to remember Naim always had Mordaunt Short MS20 as a cheap speaker one could use ( Isophon tweeter and MS paper version of KEF B200 on a Dalesford chassis with KEF approval ). I believe the SBL is that driver with improvements. I copied that idea on my nasty baffles and use a small paint roller to be a phase plug. Wonder where I got that idea from?
I always got the idea Naim made speakers because they wanted to. I never felt they really believed in them. So strange as I have used them in systems nothing like Naim and found them totally excellent. They had crossover upgrades etc which sort of said it was a learning curve that the customer joined in with. KEF also when KEF 104 or 104 AB so not just Naim. I like the 104 standard, it does work with Naim. The Linn Isobarik is 4 x KEF Kit 3 and got the Belgian tweeter later on . Sara and earlier Kans work well with Naim. NAP250, Kans with later Belgian tweeter work very well. Most other speakers to my ears don't work. The strangest exception being Klipsche Forte 2 which is so like the sound I own. Nasty and real. Like the best fairground speakers with a little refinement. I never tried Dynaco A25, I feel they should work. They are a Heybrook HB1 ,Spendor hybrid to my ears. My own speaker were inspired by the A25 although more like 1950's cinema baffles ( poormans 57's with bullet proof ability ). I have a hunch Japanese speaker FAL would work. I suspect their cheaper speaker is the best value the world has to offer. It has the weirdest measurments ( worse than mine ! ) and the most detail I ever heard and shouldn't if you read the graphs. When fitted with the compression tweeter a master class in how to do it.
As for low distortion of the Hitachi design. It comes with minimal complexity ( as LAB 47 ) and makes a lie of many statements made by others who almost seem to have made it a quest to decieve us. What FET's do not have is the slam as we called it of the Naim amps. However nor did a Sugden nor the Quad amps using Naim like devices.
The Naim sound to me is this. Projection of rhythm which slightly hides the recording engineers work. In my mind a good trade off. It can go wrong. Some Naim systems can sound limp and slightly sterile. We are talking real ones and NAP250 at that. I have no idea how that happens. What it shows is the average Naim dealer was a real music lover and Naim would have put that very high on the list of appointing a dealer. That dealer would work hard to ensure the Naim sound was had. Time goes by and non Naim dealers get involved and the magic is lost. Was it ever there? I think so. If you want it at it's best you want Nait ( 2 is fine ) and Quad ESL 57. It will take on anything and is the only example I know of the doesn't hide the recording engineers skills, that is space and interplay in lets say Beethoven. I always specualted that the correction networks we see in the NAP140 circuit are for the ESL 57, JLH sort of say similar things although doing it a different way. That Nait 57's combination will see off nearly anything at any price. With a little care it will remain safe and give bass. NAP250 suitably adjusted is good with 57's, not the vast difference over the Nait that one would think. Moving coil speakers are each on it's own merit. Spendor BC1 works and Spendor the company liked Naim. It won't be the usual Naim sound, it is a very musical one and will shame many so called accutate set ups. The Naim SBL has a very nice balance as do the cheaper versions. I seem to remember Naim always had Mordaunt Short MS20 as a cheap speaker one could use ( Isophon tweeter and MS paper version of KEF B200 on a Dalesford chassis with KEF approval ). I believe the SBL is that driver with improvements. I copied that idea on my nasty baffles and use a small paint roller to be a phase plug. Wonder where I got that idea from?
I always got the idea Naim made speakers because they wanted to. I never felt they really believed in them. So strange as I have used them in systems nothing like Naim and found them totally excellent. They had crossover upgrades etc which sort of said it was a learning curve that the customer joined in with. KEF also when KEF 104 or 104 AB so not just Naim. I like the 104 standard, it does work with Naim. The Linn Isobarik is 4 x KEF Kit 3 and got the Belgian tweeter later on . Sara and earlier Kans work well with Naim. NAP250, Kans with later Belgian tweeter work very well. Most other speakers to my ears don't work. The strangest exception being Klipsche Forte 2 which is so like the sound I own. Nasty and real. Like the best fairground speakers with a little refinement. I never tried Dynaco A25, I feel they should work. They are a Heybrook HB1 ,Spendor hybrid to my ears. My own speaker were inspired by the A25 although more like 1950's cinema baffles ( poormans 57's with bullet proof ability ). I have a hunch Japanese speaker FAL would work. I suspect their cheaper speaker is the best value the world has to offer. It has the weirdest measurments ( worse than mine ! ) and the most detail I ever heard and shouldn't if you read the graphs. When fitted with the compression tweeter a master class in how to do it.
300B. I love EL34, 84. Wouldn't want most 300B except a generic made by Kron that he said he couldn't give away. My best memory of Mr Kron is him saying to me " What's all this valve nonsense about, surely transistors are better " ? To which I said " they are ". I like people who have a sense of humour. He then explained to me why all 300B made today are wrong. It is simple, the real RCA had thermal compenstion for the heater. 300B is a great driver valve. EL34 can do all it does and better and costs £14 as Ruby brand. I designed a triode version that no one seeems to be interested in. It has the g2 and g3 removed and slightly better position of g1 if possible. Triode strapped EL34 is not quite a true triode. EL84 in the Leak Stereo 20 is almost a Naim sound. EL84 is almost a transistor to me.
If you ever wan't a valve sound build RH34/84 and forget the rest. Then learn. All Naim lovers should love it also if speakers suit. 57's do. No damping factor so won't suit all. 2A3 is the valve of choice. If you listen quitely to suit neighbours it might be your better sound.
If you ever wan't a valve sound build RH34/84 and forget the rest. Then learn. All Naim lovers should love it also if speakers suit. 57's do. No damping factor so won't suit all. 2A3 is the valve of choice. If you listen quitely to suit neighbours it might be your better sound.
I hear a strange acoustic with high reverberation which makes the music sound badly recorded and the instruments sound disconnected.
I certainly would not want to judge amp differences using youtube vids like that!
Food for thought. I imagine Naim have the resources to use types other than tantalum too. But they keep using them. Why would they not agree with you and choose something "better"?
PS: Are you using all the original Naim parts in your build where you can?
The most important part is the transformer. I'm not sure anyone (outside of Naim) knows the spec, never mind where to get one.
I have a very rough idea. It is nicely made and very low DCR as low as you dare go. What this low DCR does it mimic the way a very large transformer works without the bulk or cost. It cause a few problems. One is they react very badly to what people call DC on the mains. The NAP250 transformer was at least twice as big at it's worst case use of current. Lets say 500VA working like 3000VA in some ways. I work with people who design transformers. I doubt they would think this the best way to do it. They more likely would use an electrostaic shield. Not sure who I side with. Perhaps both.
I guess the low DCR is the same number of turns with thicker wire. In the first 60 mS ( arrived at on this forum once ) the inductance overcomes the DCR if the fuse has survived. Personally I would add a 30 amp relay, resistor and NE555 ( 1M/1uF ). I like 1 second as it's long enough to hear it happened.I guess make that resistor 230V 10 amps or whatever. That's circa 20 ohms. I would use 2 x 10 ohm 15 amp thermistors as they could work even if the relay fails. Some type of thermal safety also would be required ( Micromark thermal fuse as Naim used perhaps ).
I guess the low DCR is the same number of turns with thicker wire. In the first 60 mS ( arrived at on this forum once ) the inductance overcomes the DCR if the fuse has survived. Personally I would add a 30 amp relay, resistor and NE555 ( 1M/1uF ). I like 1 second as it's long enough to hear it happened.I guess make that resistor 230V 10 amps or whatever. That's circa 20 ohms. I would use 2 x 10 ohm 15 amp thermistors as they could work even if the relay fails. Some type of thermal safety also would be required ( Micromark thermal fuse as Naim used perhaps ).
Last edited:
- Status
- Not open for further replies.