DIY Power Supplies

What I did sense though, maybe incorrectly, was that you were trying to create the impression that you knew more than the designers and any other contributors to this forum, which without a demonstration of a proven track record in a relevant field (i.e. Audio) is untenable.

If I've misread the intention of your posts Bam then I apologise unreservedly, but comments like: -

"Send me the design and I'll be happy to help you figure out what the problem is." read, without the benefit of face-to-face contact, exactly this way. It may be a genuine offer of help, but a question rather than a demand would have been better!

Maybe more careful choice of language should be considered, and less literal interpretation

Since your posts demonstrate that you do not know the specifics of many of Naim's designs, purporting to be able to correct their errors is an impossible position to uphold. Since I do not feel it appropriate to discuss or post specific circuits that are Naim's intellectual property I can't really pursue this line any further on the forum. I am happy to continue privately though.

quote:

---

I submit one of Naim's key differentiators from competitors is the DIY nature of their systems - the allure of the tweak and add-on

---

That's a reasonable line to take, although as a proportion of Naim's user base I'd suggest (as a straw poll of contributors to this thread and other similar ones would attest), that few Naim users take any advantage of this.

The sonic benefits of Naim's approach is obvious to most with ears, but I'm certain that in most UK homes the appeal of less boxes, smaller units and tidy cabling is greater.

I'm sure if Naim could acheive the same sonic performance from a more elegant package, I feel the potential user base could be increased. Apart from those who see Hi-Fi as jewellery, the more discrete the better, I'd have thought.

The only good thing B&O ever came up with was the quote about their TV's, which is applicable to other items also, namely "One has to look at the TV, even when it's off"

But hell, what do I know, I'm not Naim's marketing department either

Andy.

quote:

---

I'm sure it is. Naim engineers have other priorities as well as sound quality. The home designer just has to worry about sound quality

---

bam,

you obviously feel that ultimate cost-no-object sound quality could be better with inboard supplies than outboard ones.

Why?

cheers, Martin

Martin,
My reason to favour on-board regulation is because, particularly with single-ended circuits, the PSRR is relatively poor. The psu voltage is influenced both by psu noise, noise picked up on the interconnect and artifacts of the amplified signal that modulate the psu voltage. I think it is imperative to keep the audio signal as clean as possible and to prevent any psu noise, regardless of source, from polluting it.

Having the final regulation stage close to the circuit reduces this noise in two ways. Firstly, it blocks noise picked up on the interconnect and secondly it minimizes the impedance seen by the circuit thus minimizing the modulation of the psu by the circuit itself or other noise sources.

I accept that having radiating components like the transformer and bridge rectifier too close to circuitry can induce noise. However, wherever these are placed the regulation should, ideally, be near to the circuit being regulated. This is why I say it ought to be better to have a *cap contain just transformer and rectifier and some caps and have the regulation done in the pre-amp. On the issue of noise from ac components I don't think it is optimum for the signal cable to be routed via the *cap - that seems to go against the idea of minimizing noise pick-up.

I hope this is clear! Sort of hard to explain in this manner. I guess all I'm saying is that I think there are compromises being made that are not in the interests of sound quality but are in the interests of creating an easily upgradable system architecture and in the interests of minimizing manufacturing cost. Of course, having a *cap is better than not having one, it's just not, in my opinion, going to get the most out of the system. And it doesn't surprise me that few other mfrs have chosen Naim's approach.

BAM

quote:

---

As I suggested earlier, they do indeed comprise single-ended common emitter stages followed by common collector buffers. The output buffers are biased using current sources. Am I looking at the same circuits as you, Andrew?

---

You sure are - you just used totally the wrong circuit description though.

The circuit is known as a complementary (or series) feedback pair, and is more than a little more complicated than 'single-ended common emitter stages' and 'common collector buffers'.

quote:

---

Having the final regulation stage close to the circuit reduces this noise in two ways. Firstly, it blocks noise picked up on the interconnect and secondly it minimizes the impedance seen by the circuit thus minimizing the modulation of the psu by the circuit itself or other noise sources.

---

But if your circuits are correct you'll see that the DC impedance of the supply is purposely increased, by design, somewhat negating this argument.

I'd agree this is counter-intuitive - I thought so to at first, yet it seems to work quite well.

Andy.

Provided the grounding is correctly starred in the pre-amp I don't see a problem with shunted current. A possible issue is the heat dissipated by the shunt - since this is a sort of class A regulator if you will.

In principle it seems like a good idea to me.

It's on my (ever increasing) list of things to try. Theoretically it is possible to achieve a very low and constant impedance over a very wide bandwidth.

The only shunt I've tried was an active noise shunt, which shunted the noise component only of the incoming supply. It modelled and measured superb acheiving in excess of 40dB of noise reduction compared to the standard circuit used.

Andy.