Amplifier frequency range. Old Naim

[Arkless Electronics]

2011 NAP200 near as damn it same FR as 1979 NAP250. I reverse engineered that as well...

 

[Jim Audiomisc]

It slopes up from 1KHz in both directions... that's what I meant by "troughs at 1KHz"...

Interesting. The plot in the HFC shows a clear slope upwards between the two end rolloffs. May be an error, though. However a slope like this or shelf might explain some of the differences people report hearing.

 

[peterm]

I believe the "line stage", as I understand it the "main gain stage" was, the 321 board. The 729 boards were "time aligned" unity gain buffer boards used before the volume control and after the main gain stage on tape (i.e. 4 per pre amp) outputs as a way of catering for older tape recorders and input sources with different impedance characteristiscs compared with modern solid state items.

The earlier buffer boards were 324s as used in the 32/32.5.

Neil McBride reckoned your could remove all the buffer boards to advantage (using jumpers to complete the circuit in their absence).

 

[Tony L]

Neil McBride reckoned your could remove all the buffer boards to advantage (using jumpers to complete the circuit in their absence).

The buffer boards were just for the tape-loop weren’t they? It is a very long time since I had a 32.5 so can’t remember all the details, but it certainly sounded better the less you had in it!

 

[Arkless Electronics]

Now have the 729 board schematic but haven't got round as yet to putting them into the sim but I can say it's an active filter board...
It seems that in a NAC72 (this is the one described on a web site I looked at, "Acoustica") there are both the 729 active filter, the line amp board I already simmed, and it also mentions a "328 (Variable Level Line input with some HF filtering)" board as being fitted but I don't as yet have a schematic for that.... So it seems that I did sims on the boards which have a wide flat response but there ARE other boards which do some jiggery pokery with the frequency response...
There is a tape output buffer board, which is a good thing

 

[peterm]

The input buffer boards fed the volume control from either the phono boards or any of the line level inputs as selected by the selector switch.
The same 324/729 boards also fed the tape outputs.

Neil McBride said the phono boards, and any modern input source, could drive the volume pot perfectly well so the input buffers could be removed.

If you didn't use a tape recorder you could also remove the tape output buffers.

If you removed those you ended up with the main gain stage (321) per channel, plus a phono board per channel if you use vinyl.

Plus the output muting relay (which Neil also suggested removing, although it makes switching on/off a bit less convenient).

 

[Tony L]

it also mentions a "328 (Variable Level Line input with some HF filtering)" board as being fitted but I don't as yet have a schematic for that....

That was a “CD input” and it sounded dreadful. I guess it was intended to drop the level down to that of the vinyl and tuner input and to knock some top-end off in systems that were obviously optimised for vinyl.

 

[Jonathan]

That was a “CD input” and it sounded dreadful. I guess it was intended to drop the level down to that of the vinyl and tuner input and to knock some top-end off in systems that were obviously optimised for vinyl.

is there a better input to use (tuner?) for digital feed?

also - (addressed to anyone) would there be significant advantage to using a 72 with buffers if I'm using reel to reel (recording and playback) ?

 

[Arkless Electronics]

is there a better input to use (tuner?) for digital feed?

also - (addressed to anyone) would there be significant advantage to using a 72 with buffers if I'm using reel to reel (recording and playback) ?

Buffers are for recording and it will vary as to how much you need them with the exact R2R you are using. They buffer the signal driving the recorder so it is isolated from the source. Some recorders may have a lowish input impedance or high input capacitance etc and without buffers the source sees this directly and has to drive it... much worse still some machines when switched off can be a very non linear load and can cause considerable distortion and attenuation of the source. In practice problems are not that common but do happen now and then... I built a buffer for someone who was using tape out to drive into a HT set up and when switched off this caused just such a problem.
Tape out buffers removes the above issues from having any influence.

 

[peterm]

As Jez said!
Depends on your R2R.

The 328 boards were introduced in the days when (newly introduced) cd players had high outputs and sounded crap.

The 328s reduced levels and introduced filtering to make the cdp sound more civilised.
Theyr'e generally considered to be rubbish!

Any line level input can go to any input which has a direct feed to the selector switch, theyr'e all the same!

The only exceptions are the inputs which feed separate boards on the motherboard before going to the selector switch (Phono 1 & 2 on 32s, Phono & Aux or Phono & CD on late 32 and 72 depending on manufacture date) which may be fitted with 322 or 323 mm or mc phono boards, the afore mentioned 328 variable level "CD" boards, or the 326 straight through boards which are effectively just copper tracks linking the input and output pins of the board sockets to provide continuity of the circuit.

 

[George J]

I suppose band width limited is something that really needs defining. When Decca introduced Full Frequency Range Recording in the 1940s this was considered as very wide band indeed, and yet it extended up to about 14 kHz. This might be seen as a considerable advance on normal pre-1939 wax cut 78 records that ranged up to about 9 kHz.

As human hearing is [nominally from 20 Hz to 20 kHz, I would think that any recording or replay that manages this range should be thought of [nowadays] as being Full Frequency Range.

Anything that goes much higher probably might be characterise as ultra-sonic! Interesting but not necessarily significant as most of the classic microphone are not actually super-sonic in their abilities in the first place, so anything being reproduced at that range [above 20 kHz, nominally] is simply going to be recording and replay system noise from resistors and so forth ...

 

[Arkless Electronics]

For those on the edge of their seats anticipating the results of a sim on the 729 "time alignment filter" boards.... relax...

It's a consecutive pair of discrete buffers made from Sziklai pairs and constant current loaded. The first has a passive 1st order LP input filter -3dB @ 128KHz and this then feeds into a 3rd order LP active filter -3dB @ 45KHz and obviously rolls of rapidly above that. -1dB is 27KHz. Ultimate roll off rate appears to be around 22dB/octave > 150KHz. Square wave performance is good. THD sims at 0.001% @ 1V PP out.

Whats it for? You tell me.... It's a nearly ruler flat buffer 10Hz - 20KHz.... All I can think of is it would make damn sure any ultrasonic "hash" and lower RF frequencies were kept away from the relatively crude and slow matching power amp and prevent any slew rate limiting or the old "TID"... there is a 1st order filter at the input of the NAP250 though but it's at about 60KHz and would still allow a fair bit of >100KHz rubbish through.

@MJS Maybe Mark at Witch Hat would know what the intent of it was?

 

[Tony L]

is there a better input to use (tuner?) for digital feed?

Naim also made some ‘straight through’ boards that were effectively just jumpers in board form. I’d use these if you want to use the RCAs or BNCs, or just use one of the DIN line-ins, which are the same thing. Its actually really easy to jump the terminals internally without the ‘straight through’ boards, you just need a wire a side with appropriate clip terminals attached and to know which goes where (which I’ve forgotten).

 

[Jonathan]

i seem to recall Naim (in their product literature) making a big deal out of the wide bandwitdth of their amplifiers ... DC to 40 or 100K or something along those lines ... i'm CERTAIN it's documented somewhere

 

[Hempknight]

I think one of my Japanese amps has a supposed frequency range up 500k (don't know why), and it still sounds 'punchy'

 

[Arkless Electronics]

i seem to recall Naim (in their product literature) making a big deal out of the wide bandwitdth of their amplifiers ... DC to 40 or 100K or something along those lines ... i'm CERTAIN it's documented somewhere

Erm... no.. You did read all the posts of the actual frequency response of the various stages above? They are capacitor coupled between stages so DC impossible... and the power amps are slow old fashioned things which can't handle much above 30-40KHz or so comfortably and have a 60KHz filter on the input anyway.

 

[Arkless Electronics]

OK whole Naim pre and power with MC board, 729 board, line stage and NAP250 power amp, no vol control just straight through is....... 1dB down @ 20KHz, 3 dB down at 34KHz and drops like a brick around 45KHz. In the bass 1dB down @ 44Hz and 3dB down @ 19Hz.

 

[Jonathan]

Erm... no.. You did read all the posts of the actual frequency response of the various stages above? They are capacitor coupled between stages so DC impossible... and the power amps are slow old fashioned things which can't handle much above 30-40KHz or so comfortably and have a 60KHz filter on the input anyway.

yes indeed i read them - and I'm just adding my bit - that i DISTINCTLY remember Naim touting this at one point ... it could well have been 'only' to 40KHz but that was an extremely impressive figure at the time and why I became interested in the Naim gear ... maybe i can find some of this literature however

and yes - see - you're just confirming what I'd read ...

 

[Tony L]

OK whole Naim pre and power with MC board, 729 board, line stage and NAP250 power amp, no vol control just straight through is....... 1dB down @ 20KHz, 3 dB down at 34KHz and drops like a brick around 45KHz. In the bass 1dB down @ 44Hz and 3dB down @ 19Hz.

I suspect if you were to model the line inputs you’d find the LPF remained (it keeps the power amp stable), but I suspect the HPF was a very deliberate rumble/warp-filter and only exists on the phono stage. The only thing I’m not sure about is whether the design fundamentally changed between the ‘chrome bumper’ kit I know well and the later olive 82, 52 etc. I’d certainly be surprised if the classic 32.5/HiCap/250 had a bass shelf on the line input, there was nothing ‘rolled off’ about the bass that came out of my Isobariks!

PS There were a lot of dodgy “wide band” preamps back in 80s that ensured huge amounts of cone-flap etc with warped records, foot-fall or anything less than a mathematically perfect mass/compliance match. It really makes no sense to allow sub 20Hz through on a vinyl input as the record doesn’t contain it and it wastes a lot of energy/is potentially damaging to loudspeakers trying to reproduce what are really noise/fault conditions. Obviously digital plays by different rules as it has the bandwidth.

 

[Darth Vader]

yes indeed i read them - and I'm just adding my bit - that i DISTINCTLY remember Naim touting this at one point ... it could well have been 'only' to 40KHz but that was an extremely impressive figure at the time and why I became interested in the Naim gear ... maybe i can find some of this literature however

and yes - see - you're just confirming what I'd read ...

I have a Naim catalogue that states the frequency response of the NAP180 and up as -3db at 3Hz and 40KHz. Hope that helps.

Cheers,

DV