advertisement


HackerNAP feedback caps?

Lokidtc99

What does this button do...?
Hi all,

Been a while since I've fiddled with my hifi, but the tinkering bug is back. I'm fairly happy with my current setup, but I've always wanted to find out what difference cap rolling would make to the HackerNAP. I built mine with wire links, so I've never actually heard the effect of caps in it. I'm aware that there are a myriad different possibilities, and I'm not going to do something as silly as to ask which sounds "best", but I'd like to get a good overview of the different flavours available.

So my question is: if I were to buy 3 or 4 pairs of caps, in order to get a good idea of what effect caps can have, which ones should I try?
 
(wipes egg off face)

Yup, of course you're absolutely right. Wire links in C1 position.

So, question number 0 then - should I roll C1 caps or C2 caps (which has the greater impact on flavour)? And then on to the second question - which caps should I try?

Sorry for the confusion!
 
Phew!

There's no better "cap" than a wire link for C1. I honestly wouldn't bother cap rolling C1 because nothing will sound as good as what you've already got.

For C2 my favourite is an 100uF wet axial tant. After that there's a lot to be said for a good quality 68uF (if you can find 100uF, even better) polyester or polyprop. Silmics were rubbish, as were Oscons.
 
Phew!

There's no better "cap" than a wire link for C1. I honestly wouldn't bother cap rolling C1 because nothing will sound as good as what you've already got.

For C2 my favourite is an 100uF wet axial tant. After that there's a lot to be said for a good quality 68uF (if you can find 100uF, even better) polyester or polyprop. Silmics were rubbish, as were Oscons.

Agree Mil spec wet Tant in Feed back position Ive user 68 & 82Uf and both sound great

http://rover.ebay.com/rover/1/710-5...0001&campid=5338728743&icep_item=191664934253
http://rover.ebay.com/rover/1/710-5...0001&campid=5338728743&icep_item=262364614462

Alan
 
This site contains affiliate links for which pink fish media may be compensated.
Isn't it better to put link wire at the output cap on the B4 and use a cap on C1?
Or can we dispense of both caps?
 
Thanks chaps.

Not trying to bias my ears or anything, but what would you say are the relative sonic differences between wet tants and polyester/polyprop?
 
Thanks chaps.

Not trying to bias my ears or anything, but what would you say are the relative sonic differences between wet tants and polyester/polyprop?

Warning: subjective waffle ahead.

Fullness of bass is the big difference for me. With film caps the bass is certainly tight and accurate, but it just seems to lack the warmth that the tants bring. The tone of the tants is warmer, richer, fuller than film caps that - to me - draws the listener into the music more.
 
I much prefer the film caps, they seem much more real and seem to allow a window into the music much more.
 
I ended up with 68uF film caps in mine. I did try 2x 47uF wet tants in // as well, but preferred the films.
 
All of which goes to show that much of the effect of cap rolling is subject to taste. My final advice would be: try to have as few preconceived notions as possible, try as many caps as you can, and go with your favourite!

For testing purposes, it may be advisable to use short fly leads to the C2 position on the PCB to make it easy to swap the caps in/out without having to disassemble the amp every time. Try to keep the leads as short as possible.
 
I used to have 47uF Ansar polprops in the feedback loop of my Onix but they are so big I had to put them on top of the case. Did sound very good though...

Don't suppose anyone has any spare AVX polys? LINK
 
Yes, I'm not surprised it's totally down to personal preference. I'll certainly give various options a listen. I'm just hesitant when 'audiophile' film caps can cost £30-50 each and I know ultimately I'll prefer just one pair but I don't want to have spent hundreds of pounds finding out which!!!
 
In the feedback loop, don't fit any boutique-but-physically-huge caps; the loop area matters, as hacker already put it.

Once you get the uF rating up where the AC voltage across it is very small in the audio band - for a NAP-alike with 1K 'lower' FB resistor I'd call that 47uF or more* - the cap's distortion contribution essentially goes away. For any reasonable cap - some will make odd choices of course ;)

I do agree, my preference if you must use them would be polypropylene (or if not available in usable form/size/cost, say next larger value of PET/ MKT.)


47uF/ 1K = ~3hz turnover for -3dB point. 24Hz is 3 octaves away, so no effect there - not that many systems usefully go that low!
 
Evox MMK, AVX FFV3 and wet tants are probably the most popular feedback caps in Naim pre/power amps.
Ansar are really good in my Onix, so you'd think they'd be good in the Hackernap BUT you just never know till you try (they are absolutely massive though). Don't bother with Mundorf m-cap (the cheap white ones) they sound very rough and thin for some reason, but oddly don't sound anything like that in a speaker crossover.
 
In the feedback loop, don't fit any boutique-but-physically-huge caps; the loop area matters, as hacker already put it.

How much does the loop area matter compared with using a quality cap though? And what would it do to the sound if you increase the loop area?
Thinking about it, I'm tempted to try some really long cable to hear what the subjective difference is.

In my Onix that I've recently rebuilt I fitted a 100uF Nichicon KZ in the feedback position, obviously quite tight to the board. The sound was a bit bright and messy in the upper midrange. I replaced the KZ with an Evox MMK (47uF) on flying leads of about 4" and the messy sound has gone but the upper mids are still a touch bright. It's a real shame because the sound is almost perfect.

Before I blew a channel on the Onix, I had Ansar polyprops on flying leads of at least 6" long. The sound was crystal clear without a hint of distortion.

Think I'm going to have to do do some experimenting...
 
I replaced the KZ with an Evox MMK (47uF) on flying leads of about 4" and the messy sound has gone but the upper mids are still a touch bright. It's a real shame because the sound is almost perfect.

Make sure the caps are in the same direction going by the print on the casing and then try flipping round the other direction - this might cure the brightness. BTW I think I have some spare 68uF AVXs so I could look for them when I get back from holiday if you drop me a PM.
 
I'm tempted to try some really long cable to hear what the subjective difference is.

No good can come of this... but... if you keep the leads tightly twisted you'll probably be fine in terms of keeping the amp stable. Probably. Maybe. I don't know what would happen with big wide loops of wire though... seems like a good way to pick up noise and inject it as feedback :eek:
 
I am running 47uF films on the board and 47uF in parallel on flying leads under the board!
Sounds good but I might remove the flying lead ones.

Pete
 


advertisement


Back
Top