I can't speak for their "Statement"
but Naim's entry level and midrange products still seem to retain the hallmark topologies that we know well here, from clones of the early models, some of which were still in the range up to 2006, as I understand. I also read and heard reference in one of their You-tube videos that Naim have a long serving staffer who is designated "keeper of the Naim sound" which may be a humorous dig at a colleague but I assume there is truth to it otherwise many devotees and prospective customers would simply lose interest in new products if there was a perceived loss of the characteristic sound. By State of the Art Design, I mean within the frame of this forum where we deal mostly with conventional solid state linear power amplifiers in several classes but mostly AB and using easily manipulated, low cost components. I take the examples of competent engineers and designers who are members here and post designs that stretch the zeros of harmonic distortion out to parts per billion. Sometimes, this is seen to be a great achievement, worthy of attention and congratulations from peers. Audio seems to be the playground of engineers and some designs may not even have commercial significance as they are really laboratory class amplifiers but that's what I understand is considered SOTA here at least.
No one has ever made an amplifier with parts per billion harmonic distortion.
Besides which, the proof of the pudding is in the tasting. A constant problem I keep coming across here are "experts" who don't do listening tests; they "listen" with simulators and bench equipment...they might as well be deaf.
Besides which, the proof of the pudding is in the tasting. A constant problem I keep coming across here are "experts" who don't do listening tests; they "listen" with simulators and bench equipment...they might as well be deaf.
NAP100 uses complementary output devices: 2SA1386 and 2SC3519.
I sometimes think there is far too much folklore spouted about Naim.
Member
Joined 2009
Paid Member
In a VAS stage with a CCS load the current variation is tiny and the series resistor will simply generate a constant voltage drop so it has limited impact. In a bootstrapped VAS it's less so and in a symmetrical VAS the current is all over the map and you need a Hagerman (e.g. my TGM5) or similar to tame the issue.
Of course this resistor reduces the voltage across the Vbe device and depending on the Early Voltage for that device will impact it's operating point. It's a rather back-door way to control it's sensitivity perhaps.
Anyhow, as you say, only way to find out is try it !
I had planned to explore some tweaks ever since finishing it but time hans't been plentiful. My priority is to explore a change in capacitors at the input and the feedback shunt.
Last edited:
Thanks for the update. I'm not familiar with all their newest stuff.
And you are right there is a lot of folklore about Naim. Adds to the fun.
Hi, I cannot open the files that you've attached, only the schematic link. From my personal experience 0.55V is absolutely normal Vbe drop. Check on your board if you have two separate returns - one for input and one for output/supply. Probably you do, because the schematic is showing that (AGND and GND). If yes, you'll need to link them. One option is directly on the board, or better wherever your star ground point is.
See if that helps.
I am working on the BOM for my diy "kit" Nap 140. I've attached a copy (I hope) of the original Nap 250 schematic so that I can name all the components.
1. I can't make out the value for Resistor R114. Is it 390?
2. For capacitors, I believe the original used dry tantalum beads for C1, C2, C3, C6, and C7. All the others would have been film. Right?
3. The only high voltage cap I see is C105. All the others only see low voltage. Right?
4. The only power resistors are R12-15. These need to be 5 watt. Right?
1. I can't make out the value for Resistor R114. Is it 390?
2. For capacitors, I believe the original used dry tantalum beads for C1, C2, C3, C6, and C7. All the others would have been film. Right?
3. The only high voltage cap I see is C105. All the others only see low voltage. Right?
4. The only power resistors are R12-15. These need to be 5 watt. Right?
R114 is 390 ohms.
C1,C2,C3,C6,C7 solid tantalum. C1 & C2 need to be high voltage, like 35V (you can use 47uF for C2 to save cost). Note that both C1 and C2 positive terminals should be on the gnd side (your C1 is shown the wrong way around)
All the pF caps are polystyrene.
R12-15 are the only power resistors. 2.5W wire-wound is enough. Don't use thick film resistors.
You can also use BC550C instead of BC239C if easier. You may need several extras so you can hand select to ideally get the output dc offset below +/-10mV. Maybe buy some sockets to make this easier.
Also, I would be inclined to make C4 47pF to give more stability margin.
Your power supply is very important. Read advice here about this. In short (not a pun) ensure correct grounding paths and use a pair of good quality smoothing caps - 10,000uF is plenty - such as BHC or Mundorf or something expensive (and heavy - but they never tell you the weight on data sheets).
Last edited:
Here's a smoothing cap in a NAP250DR.
http://www.oliveandblack.co.uk/ALC10D1019DF063-5.pdf
http://www.oliveandblack.co.uk/ALC10D1019DF063-5.pdf
Attachments
Yes, ALC10D series though, seem to be custom products and there are few sellers at reasonable prices. This Naim service agency asks £15 each + extras. Naim Audio Hi Fi Servicing Components and Capacitors | Worldwide shipping
It makes a recap with original parts rather expensive unless you are in the UK - certainly disproportionate to the value and expectations for clone kits or for curious DIY builders on small budgets.
It makes a recap with original parts rather expensive unless you are in the UK - certainly disproportionate to the value and expectations for clone kits or for curious DIY builders on small budgets.
I see a couple ironies in these statements.
1. I noticed when listening to a modest (192kbps) MP3 that I could hear things clearly that I had not noticed listening to the uncompressed version. So it seemed that the compression was helping my ears, not hindering them. Point being that, yes, MP3 is about economy of resources, but the study of psycho-acoustics that produced MP3 has discovered that perfect reproduction is not the ultimate "performance" goal.
2. MP3 and digital media compression, including digital video technology, was not originally a corporate goal. It was the product of a bunch of personal computer hackers who wanted to store and share media files over the News net and bulletin boards. OK, some of those people were also employed by large corporations where they did boring work on payroll and data bases and such, things their employer was interested in.
3. Resistors have changed, mostly in the 60s. Carbon-composition resistors had great high frequency characteristics (low inductance) but they were noisy and unstable. It was common for Carbon composition resistors to drift down to a much lower resistance if they ran warm. And surface mount technology has eliminated the wires. Today, a ~medicine bottle of 0201 or 0402 resistors looks like a pepper shaker, and it is very difficult to work them by hand.
For me, SOTA amplifiers must fulfill some criteria for some variables. I don't want to start suggesting these criteria, but someone can initiate a thread that will compile these criteria.
First variable is the classic THD at 1kHz, 1W. My criterion is that it should be equal or less than 0.002%. This will disqualify many valve and class-A amplifiers. Old Naim will be disqualified too, unfortunately.
Another Naim weakness is the performance related to phase. Hardly audible but think about SOTA!
Amplifier performance can be improved, BUT human ability to discern the improvement is close to NULL.
The Mundorf MLytic AG 10000uF 63V are also about £15 in the UK.
DIYAudio could do with a parts guide showing quality ranking and prices of capacitors. As long as I do the ranking
. Seriously, I think DIYAudio may be missing a trick.I wonder what people are buying NAP kits for? I mean if it is just to build something which is based on an audiophile brand then sure. I mean for the fun of the build. Very satisfying. But if it is to try to get Naim sound on the cheap then I don't think it is a good investment without spending more on replacing parts and good psu and so on. I suppose a stereo kit and all new caps and good power supply is going to cost £150 to £200 + case, maybe. It's still a tenth of the price of a real Naim and won't be as good (unless you know the design and have T&M equipment).
If you think I could get parts cheaper for you from within the EU I'd be happy to oblige.
Maybe. Audio systems are complicated, I find you can improve one part and the whole can seem different rather than improved, or you remove one class of distortion only to have brief joy as another class is now prominent.
MP3 removes information from the music. It may be that this allows you to notice things that remain better on an imperfect system. Could be.
It could be that distrotion introduced by the lossy compression algorithm may be giving the impression of hearing something not in the original.
I have found that my improvements to my system cause both a removal of sounds and a clarity of that which remains. For example, some capacitors can have the effect of adding a sort of high frequency hash to the music that can be a little like there is more cymbal going on or a sort of bathroom echo effect that can seem like there is more music getting through. Hard to describe it. But a better capacitor will leave the hash out and what remains is sort of less overall but much more realistic and clear.
No doubt. Now the internet is fast Apple & others still sell only lossy compressed and I imagine this has a lot to do with copyright protection.
Definitely. I'm no advocate of old for the sake of it either. If I were updating an ancient tube amp (maybe one day) I would use the best modern parts and try to hide them inside the shells of the old ones. I still prefer through-hole just for easy of handling but SMT are better for size and very high f circuits due to lower inductance.
Thanks for the kind offer. In buying the kits, I think that most folk just take an opportunity to hear the Naim sound (or that of other highly regarded audio amplifiers) for a very low cost - pocket money to some. So why not build it and see if it lives up to the reputation of the original product? Perhaps it will also sound good enough to warrant a nice case but I don't think there's much more to it, other than it's a low cost power amplifier, which can be built with cheap parts in an easy assembly, through-hole format.
If we expected to build a passable clone from the PCB and supplied components, then we were either ill-informed about the product, couldn't interpret the pics or not even thinking that day. In response, much of this thread is about the facts and fantasies spread by both kitsellers and builders, then trying to help people come to to terms with what they actually received in their kit. The issues tend to be basic rather than in the nuances of sound quality.
Few builders will have any idea what the sound quality characteristics should be, so what they hear is without reference but to their measure, must be correct and they will pass judgements like Naim sound is old hat, not as good as familiar products based on chipamps etc. It's a conundrum and part of the problem with most clone builds. They may be close to original and even measure extremely well but still sound significantly different.
I think that when you throw in all the expensive goodies that will fit an Ebay clone, you get what may only be different sound, not necessarily better. Initially, parts changes make a focus of interest but it fades as another idea and change takes precedence. So, spending hundreds on recapping a $25 clone is fraught with ifs, buts and maybes. Opinions on caps are everywhere but most would appear to be guesswork and parroting. They are popular, easy targets for changing on a whim but swapping them based on popularity is likely costing the price of the clone many times over. Is it worth it when we still don't know what the original sounds like?
Hi! I have followed this thread a bit now and I have just ordered this NAP200 PCB kit from eBay. It seems to be the "Zero Zone" (GZLOZONE) PCB.
I have also ordered a custom toroidal transformer (450VA) from Tiger in the UK, a Modushop enclosure and I have ordered the missing parts for the preamp supply on the PCB.
Now the description of the eBay PCB kit says: "Test the 0.22R 3W Resistor (near the JE243G) voltage V1.adjust the Adjustable resistor to make V1 about 2mV."
Stupid question: Do they mean R12 (see picture for reference – it’s from this thread here)? And do they mean the voltage across the resistor or from one side of the resistor to ground?
Anything else I have to do? I have read quite some things here, but if I conclude correctly it seems that people have successfully built their NAP 200 clones without necessarily replacing components that came with the kit? I’m not sure about the BC239Cs that will come with the kit. I could buy some BC550Cs and use the ones that have HFE around 500 with TR1 a bit higher than TR2, right?
And as some have stated that the trimpots are low quality I will replace these.
And then i remember that there were problems with the fuse blowing? And then it got a but too much for my limited understanding of electronics. Can’t I just use a 3.15A T fuse for the mains? The transformer will be 450VA and will have the 2x 28V and 1x26V secondaries, just as the NAIM transformer.
I have also ordered a custom toroidal transformer (450VA) from Tiger in the UK, a Modushop enclosure and I have ordered the missing parts for the preamp supply on the PCB.
Now the description of the eBay PCB kit says: "Test the 0.22R 3W Resistor (near the JE243G) voltage V1.adjust the Adjustable resistor to make V1 about 2mV."
Stupid question: Do they mean R12 (see picture for reference – it’s from this thread here)? And do they mean the voltage across the resistor or from one side of the resistor to ground?
Anything else I have to do? I have read quite some things here, but if I conclude correctly it seems that people have successfully built their NAP 200 clones without necessarily replacing components that came with the kit? I’m not sure about the BC239Cs that will come with the kit. I could buy some BC550Cs and use the ones that have HFE around 500 with TR1 a bit higher than TR2, right?
And as some have stated that the trimpots are low quality I will replace these.
And then i remember that there were problems with the fuse blowing? And then it got a but too much for my limited understanding of electronics. Can’t I just use a 3.15A T fuse for the mains? The transformer will be 450VA and will have the 2x 28V and 1x26V secondaries, just as the NAIM transformer.
They mean adjust VR1 until you get 2mV across R12 (or R13, either would do).
As R12=0.22 ohms, 2mV across it means 9mA is flowing through it. Now I think that's a bit low for the final setting after everything has settled and warmed up. I would go for 5mV across R12 (about 23mA).
These component changes we prattle on about are more to do with optimizing the sound than getting it working. It ought to work ok if you use the parts it comes with.
I am not familiar with the fuse blowing issue; I'm sure Ian will know. I believe a "T" fuse is a quick-blow so you don't want one on the mains side of the transformer because the transformer will draw a large surge current when you switch it on.
An important thing about selecting TR1 & TR2 is to make sure the dc offset at the amplifier output is not too high. Ideally, it should be <10mV (+/-). You could buy some TO-92 sockets so you can just plug transistors in and out and try lots of combinations. You may need a few extra transistors. I think either BC239C or BC550C are ok - obviously make TR1 & TR2 the same type.