advertisement


Which PSU board for NAP140?

coredump

coredump

Press <ALT-F4> to continue.
Hi,

I own a CB NAP140 with NCC200 boards and Keindeil caps. It sounds very good and is a quite iconic item.

However, it still has the old PSU board (including rectifiers) and the old cables, which I would like to change. Currently looking at Avondale's offerings, but unsure which one to pick from the physical size POV.

The NAP110 PSU upgrade looks most promising so far, as I'm not sure the Minicap 6 could be squeezed into a 140.

Alternative offers from China offered via eBay are currently not considered, as some of them seem to place the drilling holes for the internal mounting wrongly (sporting PCB tracks where holes should be), and none of them have the CLC filtering.

So, anybody with first-hand experience? :)

Best,
Oliver
 
Minicap 6 will fit, but you'd need to drill holes in the chassis. You also loose the 24v supply for the pre.

If you search here for mono 110's there should be some images.
 
Hi Mr Deepdivedog,

Thanks for your replay. Loss of 24V line is not a problem, since two of these can be provided by an SPX1. Would the 110 PSU without any mechanical modifications. If not, I might be tempted to go for the Minicap.

Thanks again,
Oliver

Edit: better choice of words
 
The Avondale 110 PSU fits with no mechanical modifications, their MiniCap6 needs two or four countersunk holes to be drilled in the base of the sled. Comparing the two the MiniCap6 has better (Schottky) rectifiers and is a CLCLC arrangement whilst the 110 module is CLC only. I'll add some photos shortly. The photos are both of NAP110s but the principles, layout and space are identical to the NAP140. The second photo is one of a pair of NAP110 mono amplifiers.

DSCF0989 by Malcolm Connah, on Flickr

IMG_0845 by Malcolm Connah, on Flickr
 
Here is the way I did it (to a 110, but the same on a 140).
I mounted the Minicap6 upside down and used the 2 central mounting holes.
This allows you to have the output wiring at the same height as the existing loom. And direct soldering to the mc6 tabs.
This is the result.

15qdt11.jpg


You do need to countersink the underside of the chassis for the 3mm spacers, otherwise straight forward. I added the tubes just for extra stability, as I had them. Washers would work just as well.

A few more pics and measurements here: https://goo.gl/photos/A8Do7Yi77PK2LUQW6

Beaten to it by Malcolm...
 
Thank you for these very informative replies! The Minicap 6 certainly looks a tight fit, while the 110 PSU looks like a breeze to work on (same dimensions as old PSU, proper terminals).

I'll have to think about this a bit. This may appear slightly chicken-hearted, but this is my main power amp (the stand-by being a Quad 306), and I'd like to keep it in as much in perfect shape as possible.
 
Hi,

I am also considering to replace my NAP140 PSU. I love the shoebox form factor. I currently build a HackerNAP PSU with 6800uF Kendeil Caps and Cree Schottky diodes. Still the MiniCap as well as the HackerCap means sharing the PSU by both Channels. Of course you can make a mono amp, but thats another story. The unique feature of the NAP140 Amp was the separated PSU rails for each of the channels, only the transformer is shared. I am considering to build a PSU using Nichicon 6800uF caps. They are only 25mm in diameter. I heared great things from the Nichicons in terms of audio quality. The height is a tight fit but it should work out. I plan to use 2 caps per rail and channel making altogether 8 caps. Still this is in an early stage of planning. With regard to the CLC filter: what is the purpose? To filter out the ripples at 100Hz and higher harmonics you would need at least 10 mH inductors. With the Ampere rating required they would be huge. In a Spice simulation I observed almost no visible effect from the inductor values used in the CLC equipped PSUs.

Jan
 
coredump said:
Thank you for these very informative replies! The Minicap 6 certainly looks a tight fit, while the 110 PSU looks like a breeze to work on (same dimensions as old PSU, proper terminals).

I'll have to think about this a bit. This may appear slightly chicken-hearted, but this is my main power amp (the stand-by being a Quad 306), and I'd like to keep it in as much in perfect shape as possible.
Click to expand...

The MiniCap6 is a perfect fit and I'd rather have direct soldered connections than anything else. Failing that I'd prefer push on spade terminals over screw terminals.
 
You certainly have a point there, Malcolm.

On the other hand, I do not mind screwed connections - if the cables' ends are crimped and the screws are tightened properly.

But spades are probably still better. So, basically the decision seems to be "separated reservoir caps per channel" vs. "CLCLC filtering".
 
Jan,

thanks for your input. No easy way with these nice little cases, it seems.
Edit: the inductors used by Les seem to be 12mH/4A types.
 
moebiusband said:
Hi,
With regard to the CLC filter: what is the purpose? To filter out the ripples at 100Hz and higher harmonics you would need at least 10 mH inductors. With the Ampere rating required they would be huge. In a Spice simulation I observed almost no visible effect from the inductor values used in the CLC equipped PSUs.

Jan
Click to expand...

The inductor here is not intended to provide ripple suppression - as you observe, it would be huge if so. The 12uH value instead deals with incoming HF noise and especially, HF rectification noise - above the audio band, and has a marked effect owing to the NAP circuits modest PSRR (which falls with frequency, of course)

Incidentally the value range of 10-12uH works well for another reason: with the typical size of the reservoir capacity, the potential series resonance is inherently over-damped by its own resistance and the capacitor ESR - no other measures required.
 
If you were converting NAP140 to monoblocs it would be relatively straightforward to convert a normal NAP140 power supply into a cap4 style board. Simply change polarity of a couple of caps, remove a leg of each of the rectifiers and link the caps with an inductor.
 
@martin: Thanks for those explanations. I already heard the HF noise reason, but need to think about it to fully understand the benefit. Thank you for giving me the starting points on what to look for :).

Jan
 
I should have added: 12uH and, say, 4700uF has the filter 'knee' around 650Hz: and with the resulting 40dB/decade roll-off, this will take a big bite out of fed-through PSU noise across the midrange where the ear is most sensitive, and even more from the treble. It's a very effective little trick.
 
Thanks for your replies. Just ordered a Minicap 6 from Les. Malcolm, thanks for reminding me that DIY is also about avoiding unnecessary compromises.

Best,
Oliver
 
Hi again,
I have been rebuilding the NAP and fired it up for a first test this morning; unfortunately, I'm measuring the full negative voltage at the output. The Minicap itself measures like it should (-39V/0V/+39V), and the cabling seems to be correct (checked it multiple times).

Things that struck me:

* The input signal socket has no 0V connection, only via the "screen" terminal; I disconnected this one as no internal +24V PSU exists; however, Les mentioned in one of his information snippets that this connection should be maintained
* The NCC200 boards currently have only one dedicated 0V connection, but two terminals for it. How crucial is the second connection?

Of course, no other equipment was connected to the amp during the test. The boards worked perfectly well with the old NAP PSU and have not been fettled with in the meantime.

I'd be thankful for any hints.

TIA & best regards,
Oliver
 
Hi Oliver,
Before you do anything else make sure all the capacitors are 'drained'. (Measure 0 volts to ground both sides +ve and -ve.) Do not disconnect anything until they are. It is possible to damage the NAP/NCC boards if you do not.

Yes I think you are just missing the ground to the front end of the NCC boards, it comes from the input connection where the screened wires join it. Put the green wire back there and take it to the ground / 0 volt connection of the minicap. Like the pictures 4 and 5, I posted earlier, here: https://goo.gl/photos/A8Do7Yi77PK2LUQW6

Alan
 
Hi Alan,

I have just drained them using a 10W/20R resistor (I always do this since a cap destroyed the tip of one of my screwdrivers). Thanks for the heads up. :)

I measured the exact difference between input signal 0V and Minicap 0V - chances are - as you mentioned - that this is the problem. I'll report back when the connection is soldered.

Thanks for your assistance!
 
Amp in situ again and playing nicer than ever. Sonic memories are not trustworthy, but I think this amp has come a long way since I bought it. The sonics are so much more sophisticated (read: more detailed, better colour, no trace of added harshness). The Minicap is definitely the icing on the cake now.

Just as a potential reference in case somebody needs the info - the original ballast resistor before the green LED is 3K3. It is situated "behind" the W06 diode in the original PSU that isolates the +24V line from the +35V line.

I decided to use the resistor that I found closest to the spec, but slightly bigger in size. So a 0.5W 3K1 carbon comp (Allen Bradley) went in, along with a new LED socket, new cable and a new LED. Connected to +35V and 0V, it shines a bit brighter than before, but still OK. To my eyes, it is much less intrusive than the blue LEDs of, e.g. a Trichord Dino or Beresford Caiman.

Thanks for the help!
 
You must log in or register to reply here.


advertisement


Back
Top