advertisement


nac 102

say it as it is

pfm Member
ok this may sound crazy but seeing and hearing what the RSL boards do to a 72 is there any way to use them with a 102?

I have heard of 821 boards being attached to 102 mainboard that seems to have similar effect.

thinking was the RSL boards are very well regulated if they could be incorporated into the 102 with a well regulated power supply like TPR4 or RSL or Teddycap.

This possible or just not viable due to construction of the 102
 
Perfectly do-able but there wouldn't be a lot of the 102 in use at the end of it. Switching and remote volume/balance only.

EDIT: Oh and the tape output buffers if you use those. I use a heavily modified 102 myself so nothing wrong with the idea.
 
The 102 has everything on the one pcb - it doesn't use plug-in boards like a 62/72 etc - so yes, you could do it, but it'd be an invasive /destructive mod.
 
It looks like both the 112 and 112x have cmos switching on the inputs as opposed to relays.
I'd probably avoid them on those grounds.
122x reverts to relays.
 
i have a 102 so just looking to see what I can do with it have a few decent power supplies so thought it might be cheaper than the mods suggested eg 7 internal regs and chang ng lots of components on the board.
looking for cheap upgrade options to match the 72
 
If you're interested, the mods I've done on my 102 mostly it involved improving the earthing structure and separating digital and analogue sections.
They are quite destructive though so would ruin any resale value.
It runs off 2 Hicaps, one for the TA section and one for the Gain section. Digital is run off a TeddyCap.

I also did the TA mod that brings it into line with the 82 & 72. It has been suggested that Naim hobbled the 102 in this regard. Mod described here:
https://pinkfishmedia.net/forum/threads/time-aligned-filter-query.31217/
and here:
https://pinkfishmedia.net/forum/threads/time-aligned-filter-upgrade-really-good.37707/

It's an ongoing project so doesn't look that pretty and I'd probably do things a bit differently if I were starting from scratch.
 
hi so your 102 now has 3 power supplies 2 hicap and 1 tedycap?

cost mounts up if i want all the power supplies to match as i assume matching is required?

what regulators would be needed if i went the whole hog and did the 7 regs with raw dc supply option?

that would still leave the in 72 terms excuse my ignorance the 321 and 729 sections to be upgraded plus what are the earthing mods?

curious if you were starting from scratch what you would do differently?
 
Let me answer the easy question first, power supplies don't neccesarily have to match. You could run a Hicap on the gain stage and a TPX1.5 on the TA section or the other way around or any other similar PSU.
I used a teddycap on the digital side because I had one - the exact type is even less critical on there.
Due to the way I'm powering the various analogue stages, directly from the Hicaps, I had to provide 24v for the output relay and that came from the Teddycap also powering the digital side. Hence needing a 24v supply instead of the napsc.

The earthing mods are a bit more involved - I changed the reference point for the potentiometers and the gain stage feedback capacitors to the main star earth.
Theoretically, this should be better because of power currents flowing from the main star earth to the sub-stars by the gain stages. Whether this actually makes a difference, I don't know as I did all the mods at once.

As well at the TA mods mentioned above, I also doubled the size of the tantalum input and output capacitors. Electrolytic caps create distortion dependant on the AC voltage across them. Increase the size of the cap and the distortion they add to the signal decreases.
Same is true for Polyester caps but the effect only shows up at larger AC voltages. I haven't seen anyone test whether the same is true for tantalum caps but I thought I'd experiment.

If I was starting again, I'd probably leave the earthing mods out and modify the HiCap for the gain stage into two separate 24v supplies with separate earths each going to the respective sub-stars by the gain stages.
This would remove the psu currents from the link between the sub-stars and the main star. The pots are currently referenced to the sub stars. I would add a direct link from the feedback capacitor to its respective sub star.

If this all sounds a bit too much, there are a set of other mods for the 102 that were written up into a pdf by a fellow PFM member that are quite easy to follow. I can't find the exact link to it at the moment but I'll have a further look.

EDIT: PDF was by Neil James - possible link to it here:
https://pinkfishmedia.net/forum/threads/naim-102.154646/
 
We are contemplating a conversion kit to 4x24V, as an add-on to our DualTeddyCap.
Will be announced in our new (and upcoming) website).
y4mnB0vnj8sCfcAxdeI6fV0n1Qrj3dKbm2fen19Ek7FBB0f9udP4nAKxuI5Lqi_JmZf3SGN79ytUrxjbV3vgPB7T0ZzefLDVxW3RLIF6G0sdw4T81KlHLTytwwZ943dEY2ajGaUoHlyt6XiA3zKUkwtOLH7TJesQ3s_eDqc93F1bDsAXIiF3ZOwt4aY-dZ80avK
 
It looks like both the 112 and 112x have cmos switching on the inputs as opposed to relays.
I'd probably avoid them on those grounds.
122x reverts to relays.

Yes, probably safer BUT the CMOS generate noise when switching so when in use, e.g. input switched and steady, they should be pretty benign in theory as they have a low resistance when 'On'.

Should be relatively simple to mod a change to relays though.....but as this is a major digression I'll stop here.
 
The big problem with cmos switches is that the R 'on' changes with applied signal voltage so their thd is quite large when just used as simple switches. Doug Self did some measurements and with a 2v signal (say from cd or dac) a 4016 with a 47k load resistance has getting on for 0.03% distortion. A 4066 drops this to 0.01% but it would still be a major contributor to the distortion figures for a preamp.

There are techniques to mitigate some of these problems but that would depend on the exact implementation Naim used on those models.

And yes, sorry to the OP for the digression.
 
The 102 is a bit of an odd one in the Naim range. I have one, but have recently brought into service a NAC 12 with NJ cards and it sounds really good, but I do miss having 6 x inputs; I haven't quite worked out how I'm going to plug everything into the NAC 12 yet!
 
The big problem with cmos switches is that the R 'on' changes with applied signal voltage so their thd is quite large when just used as simple switches. Doug Self did some measurements and with a 2v signal (say from cd or dac) a 4016 with a 47k load resistance has getting on for 0.03% distortion. A 4066 drops this to 0.01% but it would still be a major contributor to the distortion figures for a preamp.

There are techniques to mitigate some of these problems but that would depend on the exact implementation Naim used on those models.

And yes, sorry to the OP for the digression.


However in mitigation maximum distortion occurs at maximum voltage swing I.e at the loudest point in the music, so will generate low order harmonics which are far less objectionable to the ear, high order harmonic distortion can sound very unpleasant as that is caused by errors near the quietest part of the signal where the ear is very sensitive.
 
Maximum distortion does occur with maximum voltage swing, I agree, but as these are input switches they recieve the un-attenuated output from the , say, DAC or CD.
Given the current loudness wars, on some music a substantial amount of signal is going to be up there.
Also, the figures I quoted above are from single tones, you then have to start thinking about intermodulation distortion as well.
I think the point is, that these are avoidable distortions. Using a switch or a relay removes these non linearities.

I'm afraid I don't follow the second part of your argument. Low or high order harmonics can be generated anywhere there is non linearity in the transfer function and is dependant on the details of the non linearity.
As a rule, in any class A circuitry, both low and high order harmonics decrease with signal level i.e. when the transfer function approximates closer and closer to a straight line.
In the specific case of there being crossover distortion present, then yes, high order harmonics can be highest at low signal levels but in a preamp, there is no need for for this ever to be present.
 
right last shout before I give and accept a lousy p/x price on the 102

who is up for doing some mods to 102 ?

either

Neil Jadman style mods have a pdf with the details eg changing resistors and caps and earthing arrangement and fitting internal regs for each rail powered by a raw dc supply. (I have a power supply for this a diy Hicap with a big Canterbury winding tx don't have any other components needed)

or

Attaching 821A boards in place of the 321 circuit, either modifying or linking out the 729 circuits, supplying the 821A with dedicated power supply, if 729 still utilised again dedicated power supply. (for this I have the 821A boards and several power supplies TPX1, TPX 1.5, Sonic Bliss, Diy Hicap)

be happy to pay for some ones time as I don want to just give away the 102 and lose remote function.

cheers
 


advertisement


Back
Top