Have you mounted the board with its heat spreader plate in a thick aluminum case or even on a separate heatsink yet? The heatspreader plate is not a heatsink in itself and of course, the amp will overheat otherwise.ZTX (VAS) transistors heat the all board. Also power transistors. All PCB temp. is not lower than 30-40C (room tempr. 22C). I believe this is not an accident. I made bias nearly 5mv (for one 0.22ohm). http://www.acoustica.org.uk/ says must be not less than 4.5mv. (make sure your Naim's output stage displays at least 4.5mV across a 0R22 emitter resistor that is, Iq around/up to 20mA..). I think there isno problem about bias and heat of amp.
I'm thinking about the transistor's effect in frequency range (or sound qualty) of the amp. I mean is it necessary to find low noise transistors in bc550 or bc239 or is MJE15030/31 or ztx's characteristic (Hfe or others) are important. Do I have to find the true BC550
That's right. Quad's isn't what I would call feedforward; it's more a cross-over distortion cancellation technique that employs a Wheatstone Bridge. One practical means to avoid bias adjustment issues in production. Well, sort of. Didn't really address audible distortion mechanisms in general, though, and had some stability challenges. Quad marketed the heck out of it.
I like Lipshitz & VanDerkooy; they seem to be able to do mathematics competently. They gave Prof. Hawksford a good spanking of reality too.
The challenge of feedforward is being able to predict the output error in advance. This is, of course, impossible to do when the only source of prediction is the output itself. So in practice you can apply some blunt adjustments using feedforward but to precisely reduce the error in real time you have to use feedback.
Member
Joined 2009
Paid Member
In proncipke at least, feedforward can perfectly cancel the distortion. There are however, two reasons feedback is needed for the Quad, a) to ensure a known fixed gain, b) to create a low impedance (voltage mode) output where the signals from the two amplifiers must be combined without mutual interference.
Lets call it a super conduction state... it means that at all frequency s, sound from the speaker is at its fullest transparency. Its so hard to catch that mode and stay there.
I dont know nothing that can do this sound.
Bandwidth, Q, Feedback, bias, harmonics, temperature, impedance, whatever and AC, Where it starts and where is the end ?
There is one more strange behaviour, i had 2 channels in one case, changed a BIAS bypass cap from 10V 47uf unknow tantalum to a panasonic 6V 47uf tant on a right channel.
After the turn on, a whole amplifier sound was different. I feel like there is some reflection going on, from the right and to the left, back and forward. basically everywhere in the case. It might be distance from one another or the wires,
It feels like we know nothing, its 2022 and still nothing ? but we can feel it...>P
EDIT: this amplifier is useful even if the transparency is not achieved, for example with harmonics on the output... they can be made "silently gentle", pleasant and who know whatever else.
EDIT2: Its simply universal. Most basic ground to learn AC behaviour and super rewarding. We dont need to buy this ****... better learn from it
I dont know nothing that can do this sound.
Bandwidth, Q, Feedback, bias, harmonics, temperature, impedance, whatever and AC, Where it starts and where is the end ?
There is one more strange behaviour, i had 2 channels in one case, changed a BIAS bypass cap from 10V 47uf unknow tantalum to a panasonic 6V 47uf tant on a right channel.
After the turn on, a whole amplifier sound was different. I feel like there is some reflection going on, from the right and to the left, back and forward. basically everywhere in the case. It might be distance from one another or the wires,
It feels like we know nothing, its 2022 and still nothing ?
but we can feel it...>PEDIT: this amplifier is useful even if the transparency is not achieved, for example with harmonics on the output... they can be made "silently gentle", pleasant and who know whatever else.
EDIT2: Its simply universal. Most basic ground to learn AC behaviour and super rewarding. We dont need to buy this ****... better learn from it
Last edited:
Quad's 405 was their boldest step in solid state design which I think, made it a target for critical analysis because it was radically different to your typical linear audio amp. and came with a technical paper that explained this in some detail. A proper design paper for a retail product is rare but definitely capable of scrutiny, so after almost 50 years, it maintains the 405 design as something of a punching bag for engineering students, designers and critical listeners.
Unfortunately, critics have generally remained focused on the elaborate proofs or otherwise regarding P. Walker's technical design details rather than elaborate on the tangible benefits to the buyer/listeners.
Unfortunately, critics have generally remained focused on the elaborate proofs or otherwise regarding P. Walker's technical design details rather than elaborate on the tangible benefits to the buyer/listeners.
Last edited:
Fair question. I was alluding to the benefits being unconvincing myself, because there are several reasons, some non-technical and unrelated, for an ongoing fascination with Quad "current dumpers". Introducing a virtually unique audio design approach plus being a focus for nostalgia and preservationist efforts to mark the halcyon days of British hi-fi and collecting related printed articles being just a couple.
I agree that the sound quality of Naim's products wasn't there in Quad's 405 but it does have something else that that for me, comes out better in the later 606 and particularly 306 models. You get a lively, somewhat finger-tapping sense of musical engagement that is absent from your typical densely populated, over-built linear amps with multiple feedback paths that don't seem to achieve anything better than keep THD and noise down to undetectable levels. It's obscure to me where this effect comes from but I quite like it because it can breathe a little life into some otherwise dull recordings.
Last edited:
Quad started out as a valve amp maker. I have never heard one but they had a very good reputation. You can imagine what happened in the early 1970s when Japanese transistor amps started flooding the UK market. Cheap, unbreakable, small, light, reliable. A huge shock to the existing industry, I'm sure. It was this time of technology disruption that forced Quad into making transistor amps. At nearly the same time, Vereker spotted an opportunity to improve the sound of these new transistor amps, which he found highly lacking; he was raised on valve amps after all. Peter Walker must have been very frustrated by how hard it was to make transistor amps sound as good as valve amps. He was not alone in this struggle: many brands tried and made various actual steps forwards and also a lot of marketing BS. But I think Vereker did the best of the lot because, by cunning or luck, he discovered something the others overlooked. His second break was to win the ear of Linn and partner with them to supply whole systems. I guess Linn and Naim had nearly 20 years of dominance in the UK.
Naim's problem was that they didn't innovate enough. People got tired of the eccentric product forms - they were and still are sort of Jules Verne. Big, black, ugly. The antithesis of modern audio electronics. Linn dropped them and made their own sexy boxes and adopted more modern interfaces and styles. Naim toyed with speakers and digital inputs (the latter they got quite good at) before running out of money and then selling themselves to Focal (they claim a merger but they effectively sold). Naim is no longer a Vereker child and their designs are still stalled, sound-quality-wise, if not worse. New designers came in and applied Horowitz & Hill (as it were ) and lost the recipe. The Statement was an attempt to fit in with Focal's eccentric, high price wares. An insider told me they had a lot of trouble trying to tweak the Statement to sound acceptable, N.B. their design theory was not good enough.
It is frustrating to observe because Naim didn't spot the opportunity to have the greatest sound on the planet and be contemporary in style and features at the same time. It seems to me they are now just another boutique box supplier for the rich. They aren't even British anymore.
Naim's problem was that they didn't innovate enough. People got tired of the eccentric product forms - they were and still are sort of Jules Verne. Big, black, ugly. The antithesis of modern audio electronics. Linn dropped them and made their own sexy boxes and adopted more modern interfaces and styles. Naim toyed with speakers and digital inputs (the latter they got quite good at) before running out of money and then selling themselves to Focal (they claim a merger but they effectively sold). Naim is no longer a Vereker child and their designs are still stalled, sound-quality-wise, if not worse. New designers came in and applied Horowitz & Hill (as it were
) and lost the recipe. The Statement was an attempt to fit in with Focal's eccentric, high price wares. An insider told me they had a lot of trouble trying to tweak the Statement to sound acceptable, N.B. their design theory was not good enough.It is frustrating to observe because Naim didn't spot the opportunity to have the greatest sound on the planet and be contemporary in style and features at the same time. It seems to me they are now just another boutique box supplier for the rich. They aren't even British anymore
.
Interesting commentary. Naim now seem to focus on selling all-in-one players as much as component power amps but I ask myself why they virtually shifted the core, power amp. design direction sideways to accommodate conventional engineering practices, over a unique type of design that had been working admirably and receiving sympathetic upgrades, over decades. This is about listening qualities and whatever the assumption that all power amps should sound the same implies, a common reason for buying a particular brand/model is that they don't, so our choices are not trivial and neither is design continuity.
When a new team throws out the mould and starts afresh with what they call "the world class circuit" you can almost feel the BS as it permeates the product line, marketing and even their fan club. It seems Naim are still looking to make a fresh impact with relatively conventional designs in the uncertain, fickle world of promotional wordsmithing at the top end where the rich are happy to buy mutton dressed up as lamb and what should be called out as mediocre may well be promoted/perceived as exquisite. It's fortunate that few people take the promotional claptrap seriously but it would be nice occasionally, to see some honest, listener opinions expressed too.
When a new team throws out the mould and starts afresh with what they call "the world class circuit" you can almost feel the BS as it permeates the product line, marketing and even their fan club. It seems Naim are still looking to make a fresh impact with relatively conventional designs in the uncertain, fickle world of promotional wordsmithing at the top end where the rich are happy to buy mutton dressed up as lamb and what should be called out as mediocre may well be promoted/perceived as exquisite. It's fortunate that few people take the promotional claptrap seriously but it would be nice occasionally, to see some honest, listener opinions expressed too.
Last edited:
My best practice; using upper diagram,
TR1,TR2, BC550C or BC239C similar results, similar Hfe
TR3 MPSA06,
TR4, TR6 ZTX753&ZTX653
TR5 ZTX108 or ZTX384 (hfe >200 which I used),
TR9&TR10 MJE15030&MJE15031, I think MJE243&MJE253 doesn't change the result. May be hotter.
TR11&TR12 for 40v BDY58 or MJ15003 or MJ15023?(22?), 2SC2922 (for 35v). In my music system (dac&speaker) does not change the sound.
C1 tantal&100nf parallel,100nf change the HF.
C3 tantal 100uf
C7 elect. caps 47uf, 100nf parallel (I don't say this config must be. I used
R2 22k
R3 4k7
R6 you can change it for idle voltage. test it. 560, 620
I didn't use c2. But if you use preamp NAC42.5 use 2x 100nf caps between input +,- to ground. this decrease the electrical noise.
At output I used a small coil parallel to 15R.
TR1,TR2, BC550C or BC239C similar results, similar Hfe
TR3 MPSA06,
TR4, TR6 ZTX753&ZTX653
TR5 ZTX108 or ZTX384 (hfe >200 which I used),
TR9&TR10 MJE15030&MJE15031, I think MJE243&MJE253 doesn't change the result. May be hotter.
TR11&TR12 for 40v BDY58 or MJ15003 or MJ15023?(22?), 2SC2922 (for 35v). In my music system (dac&speaker) does not change the sound.
C1 tantal&100nf parallel,100nf change the HF.
C3 tantal 100uf
C7 elect. caps 47uf, 100nf parallel (I don't say this config must be. I used
R2 22k
R3 4k7
R6 you can change it for idle voltage. test it. 560, 620
I didn't use c2. But if you use preamp NAC42.5 use 2x 100nf caps between input +,- to ground. this decrease the electrical noise.
At output I used a small coil parallel to 15R.
"preamp NAC42.5 use 2xMy best practice; using upper diagram,
TR1,TR2, BC550C or BC239C similar results, similar Hfe
TR3 MPSA06,
TR4, TR6 ZTX753&ZTX653
TR5 ZTX108 or ZTX384 (hfe >200 which I used),
TR9&TR10 MJE15030&MJE15031, I think MJE243&MJE253 doesn't change the result. May be hotter.
TR11&TR12 for 40v BDY58 or MJ15003 or MJ15023?(22?), 2SC2922 (for 35v). In my music system (dac&speaker) does not change the sound.
C1 tantal&100nf parallel,100nf change the HF.
C3 tantal 100uf
C7 elect. caps 47uf, 100nf parallel (I don't say this config must be. I used
R2 22k
R3 4k7
R6 you can change it for idle voltage. test it. 560, 620
I didn't use c2. But if you use preamp NAC42.5 use 2x 100nf caps between input +,- to ground. this decrease the electrical noise.
At output I used a small coil parallel to 15R.
I think it would be better to discuss Jeff Young's modified NAP 250 design in his or a separate thread. This schematic has no date reference and it's a modified (separate, regulated power to the front end) design that was amended several times in re-development so it probably isn't relevant to the discussion. It was also important to Jeff that the semis resembled the original TO3 types to appear authentic as much as equal their performance.
FYI, the later, much improved versions of NAP250, all used Sanken LAPT and related types of high grade power transistors. That's what Naim still use in their more popular models and if you are looking to get the best Naim style sound quality in a clone, start there with genuine Sanken, Onsemi or Toshiba's top audio grade power transistors. Some may be expensive but there's no point in substituting cheap copies, fakes or ancient switching type TO3 relics.
FYI, the later, much improved versions of NAP250, all used Sanken LAPT and related types of high grade power transistors. That's what Naim still use in their more popular models and if you are looking to get the best Naim style sound quality in a clone, start there with genuine Sanken, Onsemi or Toshiba's top audio grade power transistors. Some may be expensive but there's no point in substituting cheap copies, fakes or ancient switching type TO3 relics.
+1 Ian and traderbam!!!
I think Unity is saving their game. It opened the company to much larger audience than their "hi-end" components. It's kind of funny, because they were claiming some 20 years ago how Naim will never go into digital, because it was inferior. Look at them now. Streaming one-box digital device is their main game
I think Unity is saving their game. It opened the company to much larger audience than their "hi-end" components. It's kind of funny, because they were claiming some 20 years ago how Naim will never go into digital, because it was inferior. Look at them now. Streaming one-box digital device is their main game
Yes, it really has been an about-face direction for Naim. I guess market research got through to management with the message that digital music is now largely delivered as streams and purchased as downloads so the best disc player technology in the world wouldn't have helped Naim stay afloat. Building equipment that only meets the needs of vinyl revivalists or CD collectors won't pay the bills any better than Vereker's eccentric multi-box, backward-compatible component audio systems of the past could in today's multimedia environment.
The current Uniti Nova model sells for around US $7,000 which places it in the top echelon of compact audio systems but certainly not the most expensive. I refer to the US market there because I think that's where Uniti product is ultimately aimed and should sell strongly as long as its features are adequate and reviews support it, whatever the ticket prices.
The current Uniti Nova model sells for around US $7,000 which places it in the top echelon of compact audio systems but certainly not the most expensive. I refer to the US market there because I think that's where Uniti product is ultimately aimed and should sell strongly as long as its features are adequate and reviews support it, whatever the ticket prices.
I gave that example schematic, bec. it's look is better, all parts named. I don't want to discuss the Young's schematic.
Drawing a PCB for this design is an art of currents, voltages and impedance overall. I think we should keep it simple and dont go into details... cuz devil is there, he got me twice
Something with the phase-shift networks, driver capacitance and theyr 100R local feedback resistors..
Track width and thickness along with the input capacitance(RC) filter will dictate amplifier harmonic structure on the output.
Powersupply lines do not affect much if you dont twist them into different shapes. But they will play a huge role if being manipulated with, for example twisted, angled or anything else.
Capacitors on the powersupply lines... theyr impedance is not only determined by the manufacturer. They layout overall and the diodes too.
Diode capacitance vs output transistor capacitance and the distance between these two components ?
EDIT> i forgot ground, ground is important, i mean... it is crucial in this design... and it doesnt matter how u do it... cuz layout determines the grounding technique.
This is crazy and super fun
Something with the phase-shift networks, driver capacitance and theyr 100R local feedback resistors..
Track width and thickness along with the input capacitance(RC) filter will dictate amplifier harmonic structure on the output.
Powersupply lines do not affect much if you dont twist them into different shapes. But they will play a huge role if being manipulated with, for example twisted, angled or anything else.
Capacitors on the powersupply lines... theyr impedance is not only determined by the manufacturer. They layout overall and the diodes too.
Diode capacitance vs output transistor capacitance and the distance between these two components ?
EDIT> i forgot ground, ground is important, i mean... it is crucial in this design... and it doesnt matter how u do it... cuz layout determines the grounding technique.
This is crazy and super fun
Last edited:
What are the currents and why they affect sound in the speaker... How much current needs to flow through the Powersupply diodes or output transistors ?
Is 38mA bias the "END" ? I think not.. 5mA can also be the END for the best sound, 20mA, 1mA ?
Example: Lets say i pull 60mA from the diodes on psu line and 5mA from the outputs or 50mA from the diodes and 10mA from the outputs, will the results be the same on the speaker ?
are Harmonics being manipulated with the change of currents ?
Is 38mA bias the "END" ? I think not.. 5mA can also be the END for the best sound, 20mA, 1mA ?
Example: Lets say i pull 60mA from the diodes on psu line and 5mA from the outputs or 50mA from the diodes and 10mA from the outputs, will the results be the same on the speaker ?
are Harmonics being manipulated with the change of currents ?
Last edited: