advertisement


NA729 Board Whys and Wherefores

M

Mus

pfm Member
I eventually got round to making up some nice silver wire links, so I could remove the 729 boards from my 32.5. I popped in the links and put on a CD, the initial impression was wow! However, 15 seconds after sitting down I had the uneasy feeling something was very wrong. When the track finished I switched back to the 729s and the detail returned. I repeated the experiment four times, with different CDs, the result was always the same; removing the 729s sucked the life out of the music. The timing was shagged, female vocals sounded dreadful (almost distorted), fine details were masked and the music was boring.

The initial experiment was conducted using AAA5/APX2/SR/32.5/260Z/IBLs. The only boards in the 32.5 were relay, 321/5 and the 729s. Swapping my modded 321/5s for standard 321/5s made matters slightly worse. Swapping the 260Z for either a 140 or a 160 made no difference.

My problem with all this is that people, whose opinions I trust, swear blind removing the 729 is the way to go. What gives, why the discrepancy? Much as I hate poles may we have a show of hands for and against removing the 729s?

I know the subject has been touched upon before but there was no real consensus.

Regards,

Mus
 
Someone who's name starts with "V" and ends with "K" figures that silver wire screws up a system. I'm not sure that I agree with him, but have you tried some relatively-pure copper instead of the silver?

-=> Mike Hanson <=-
 
Someone else posted similar comments re removing the 729s - so you're not alone. Might have been Linnik, but have a feeling it was someone else. I have a tendency to agree as well, though I've not got anorakky enough to have formed a definite opinion.
 
Mike Hanson said:
Someone who's name starts with "V" and ends with "K" figures that silver wire screws up a system. I'm not sure that I agree with him, but have you tried some relatively-pure copper instead of the silver?
Click to expand...

I'll give it a bash in the morning; perhaps increasing the conductivity of the wire makes all the little electrons get over-tired.

Regards,

Mus
 
Is it possible the 729’s are acting as buffers. Maybe the AAA5 needs buffers to work with the 321’s. Have you tried refitting the buffers or using a Naim CD player.
 
fatcat said:
Is it possible the 729’s are acting as buffers. Maybe the AAA5 needs buffers to work with the 321’s. Have you tried refitting the buffers or using a Naim CD player.
Click to expand...

FatC,

Yes and no. 729s sound better than the 324s (buffers) which sound much better than a wire link. The CD input is attenuated internally by a pair of resistors mounted on a hacked straight through board; removing the attenuation made no difference. Substituting a Rega CDP for the AAA5 made the wire a bit more bearable, but not much. I have not tried a Naim CDP but I can't imagine that would make any difference.

Regards,

Mus
 
Its possibly a qustion of the relative level of your system.

the primary function of the 729's is to provide a time aligned impedence matching buffer stage.

Taking it out introduces an impedence mismatch which probably makes the sound a bit brighter and harsher (and possibly a bit clearer too if your 729's are due for a service or have a limited power supply).

In Mus's system I suspect that none of the above applies so he's genuinely hearing exactly what the 729 is supposed to do. Ie sweetens the sound by eliminating the impedence mismatch and preserves the timing - just like its designed to.

Strange how Naim designs are usually good isn't it ?
 
Hi Mus
I did not have access to 729 boards, so I did the same experiment with the 324. Sorry to tell you - it just belongs in there.
I ended up with attaching TPR reg boards to each one of the 324s and 321s - this gave a tremendous impact.
Yair
 
It's very simple thing as the buffer limits the freq. range outside our listening range.
Thus it will make the gain amp of NAC and the power amp to live in a FAR easier world. Both of those amps work much better being happy for easy living. Timing is one thing which gets better.

The freq. limiting buffer is one the Naim's secrets. The sound is made there. The time-aligned buffer makes the filtering more silky way than the older one in 42.5, 62 and 32.5. That is why I bought the time-aligned cards from a 72 owner and seated them into my 62. Much purer and cleaner highs with better tune.

Now I understand the reasoning of many people who aim to simpliest and shortest possible circuit. But it does not work that simple.

Without a preamp would be simpliest front end circuit but in numerous tests, most often a system sounds better with a good preamp than without a preamp.

Also one understandable cliché is that the best cap is no cap. But it is not true. The feedback needs a cap. Two caps (bypassing) sounds better. And now I hear 3 caps (double bypass) sounds even better than that.

There are many often mentioned principles in the art of electronics which are not true in some cases even if they are true in principle. That have I learned.

For me it sounds clear that the freq. limiting buffers are vital if we want to keep Naims as Naims. Or better, if we want to reach the Naim sound.

Oz
 
There's some intresting comments there.......

My two pennies worth would be to leave the silver links in for a couple of days soak...

That will change there charactor slightly.

You could just use the silver to jump the signal to each stage....best of both worlds.

What are you using ....single solid core in ptfe ?

Its been known for ages that the naim amp stages don't like to overdriven or exposed to ultra sonics.....A buffer sounds a bit on the heavy handed side ????? but I'm no expert.
 
zanash said:
What are you using ....single solid core in ptfe ?
Click to expand...

I have now tried multi-stranded & solid core with silver in PTFE. I have also tried pure copper multi-stranded & solid core. Made bugger all difference to my ears. I left a pair of the silver links in place for a few days, again no difference. For the time being my 729 are staying...

zanash said:
You could just use the silver to jump the signal to each stage....best of both worlds.
Click to expand...

Do you mean leave the links in and the 729s? Won't most of the signal take the path of least resistance down the wire?

zanash said:
Its been known for ages that the naim amp stages don't like to overdriven or exposed to ultra sonics.....A buffer sounds a bit on the heavy handed side???.
Click to expand...

729s sounded better than a link wire even when using a 260Z, so it's not just a Naim power amp thing.

Regards,

Mus
 
G'day Mus,
some past musings:
ok so you popped in the wire links, and it sounded worse. I agree that the 729s are better than the 324s. Didn't really notice much difference between buffers and none to be honest though; Bit thinner and more clinical but not particularly night/day stuff. So I agree, 729 better than 324 better than none.

but a question is 'how is your pre powered?'

I went from a hicap clone with 4 superregs in the pre powering gain and buffer boards with one winding from the *cap feeding the gain boards and 1 feeding the buffers. V good.
But found the optimal solution was to have one winding per superreg per gain board, period. Complete dual mono supplies. What you loose from removing the buffers you gain with interest by powering each gain board completely independently.

Effectively you trade a decrease in circuit performance for a bigger increase in power supply performance.
this would imply the biggest gain would be 4 windings feeding 4 superregs feeding 2 x gain and 2 x 729 boards - quad mono supplies- but not tried it.

So the question is... do you have a twin winding hicap clone to power the pre? If you do its quite simple to pop back in the link wires and rewire the regulation. What results you'll get with other regs, trannys or supplies I can't guarantee but worked beautifully in my setup and simple to try and reverse.
cheers
ced
 
Effectively you trade a decrease in circuit performance for a bigger increase in power supply performance.
this would imply the biggest gain would be 4 windings feeding 4 superregs feeding 2 x gain and 2 x 729 boards - quad mono supplies- but not tried it.
Click to expand...

I've done that using Avondale APX power supplies and four frame transformers so each board in my 72 (323s and 321's) has its own trannie, rectification, smoothing and regulation.
 
RichardH said:
Someone else posted similar comments re removing the 729s - so you're not alone. Might have been Linnik, but have a feeling it was someone else. I have a tendency to agree as well, though I've not got anorakky enough to have formed a definite opinion.
Click to expand...



Probably me, done this a few times, 72, hi, 250, used various links to try rule that out, to my ears there was more info coming through without the 729s, but it didn't sound right, something was wrong, the realism went AWOL, the life gone from the music, it reminded me of my old audiolab 8000a, real clarity but strangely monotone.

Regards
 
An entirely reversable mod for the 72, the result of which beggar belief, I started out by making a cable to feed the 72 from the hicap with 3mmsq cores x3 for the 24v + 0v lines, after chopping out the unused pins the cable was easy to fit in the plug.
IMG_2498.jpg


The cable contains no signal line, the signal to the 250 comes now from the 4 pin output on the pre amp.



Next, the same cable was used to supplement the 24v feed lines inside the pre amp, from input socket straight to the pins of the 321 and 729 boards.
Nac72.jpg

IMG_2515.jpg

opps, the yellow wire got in the way!

The mods effects were not something I was prepared for, and shows just how big the hi cap really is, it's a truly huge power supply and this shows it big time, the Naim character unaltered but now with ENORMOUS dynamics and scale, sibilance:zero, scale:enormous, speed:timewarping, the sound seems to have gone somewhat 3D.

Little effort is involved, so why not give it a go, it may look ridiculous but the results are out of proportion to expence by a 1000 fold.
 
ced said:
but a question is 'how is your pre powered?'
Click to expand...

Hi ced,

Thanks for the reply, you make some interesting points. I use an APX2 but I am just in the process of adding a second APX2. So I will end up with 8 windings feeding 4 APX boards feeding 2 x 321 and 2 x 729 boards. However, I will also try paralleling the output of the two APX2s to see what difference that makes. I have also been experimenting with using an APX2 to feed a pair of superregs, initial results are very promising indeed.

Regards,

Mus
 
You must log in or register to reply here.


advertisement


Back
Top