NAD 3020 clone

[stoc005]

Here is the schematic from a 3020B that was simulated using modern transistors. I thought I would experience the equivalent of audio nirvana when I saw the gain/phase plots of this circuit but I was underwhelmed. To be honest, the gain/phase plots look awful compared to a typical Leach amp. Of course looking at a simulation doesn't tell the whole story. But if it looks and simulates less than stellar, why build it? Some of the blame could be the modern transistors used to make it. I am surprised by the number of small caps sprinkled around the circuit to shape the frequency response. Sure, it's flat +/- 3db 20 to 20K but the phase response is iffy and it appears that there will be stability issues at high frequencies. Blame the modern transistors, I guess. Has this amp been cloned? Does the NAD 3020 deserve it's cult-like reputation??

I guess the search for the audio holy grail continues......

[wg_ski]

Try bypassing the input filter. It intentionally limits the BW and is bypassable on the real amp (LAB input).

With modern transistors, it may require more agressive frequency compensation. Only experiment (not simulation) would be able to tell for sure. With only +/-30V supply it probably wouldn't blow up if it did oscillate.

[lineup]

Onlky to get the idea to make a clone of one classical Bjorn Erik Edvardsen (BEE) amplifier
shows you are a bright young man

Because this Norway guy makes some very good design
with corrent audio amplifier ideas!

The new classical amplifier from him
can be found here: nadelectronics.com/products/hifi-amplifiers/C325BEE-Stereo-Integrated-Amplifier/details

Why not liike I suggested
make a clown out of a more modern variant of 3020 : C320BEE
http://www.diyaudio.com/forums/showt...84#post1104784

Schemnatics can be found here:

ACD denmark member of diyaudio helpfull service

[stoc005]

Yes, the frequency response is different and smoother. Still has too much gain at -180 phase shift. I predict oscillations at some high frequency. But maybe the output inductor is what is needed. There is a 1uh at the speaker terminals.

[lineup]

I do not get your comment.
Is the problem the 3020 or the C320BEE circuit I refered to?

[stoc005]

Lineup: That comment was in reference to using the LAB input. Yes, it bypasses the input filtering and smooths things out

I did snag a copy of the SM for the 320BEE. That is a very interesting website for manuals. I find the 320BEE balanced design interesting. Quite a few more parts than the 3020 and different but there is a similarity to the original 3020 as well. The DC path for the input transistors is through the speaker output terminal, just like the 3020. Yes it's possible to clone it, I think. Maybe if I have time some day I'll see what it looks like in LTSpice. Shouldn't take but a hour or so to get some first pass results.

[stoc005]

WOW! With modern plain vanilla SMT transistors (not the 2SA/2SB/2SC/2SD types) this amp simulates very well. Flat to 100K. Excellent phase response at the ends and through the audio range. I like it. I want to build it. If you want a copy of the LT Spice schematic, email me at stoc005 (at) comcast (dot) net and request a copy. The reason I am not including a .pdf of the LT Spice schematic is because somebody could just copy it and sell boards/amps on EBay.