Ncore Improvements

Ric Schultz · « **Reply #180 on:** 13 Jul 2012, 09:39 pm »

My friend with the first set of Ncores I made finally listened and then tried AMR fuses. He says the fuses made the amps fuller, cleaner and had more air. And this was using the amps on a two way bi-amped speaker on the bottom end. Basically running 20hz to 850hz......48db per octave xover. He loves the control and slam of the N-cores but feel his highly modified Hurricane amps have more air, texture and harmonics.

I will be firing up another pair of N-Cores in the next few days and I will try Furutech fuses and also bypassing the fuses with a piece of wire underneath the board to give a reference. Then I will be removing (ripping off) the .68uf output caps and dremeling off the sides of the caps and removing the steel leads and replacing them with OFC leads and marking the caps for outside foil. I will then replace the caps underneath the board (outside foil oriented to ground). I will also try removing one of the caps and see if it is still stable and if so will listen with 1.3uf instead of 2uf. If I like this effect, then I will replace the two .68 mylars with three modified Wima .33 polyprops.......lots to try. I will let everyone know what I hear and if the cap mod works great I will show pics of what I have done.

serengetiplains · « **Reply #181 on:** 13 Jul 2012, 10:34 pm »

Excellent, Ric. I think those mylar caps have to go and will be interested to hear what you find by replacing them as you intend.

serengetiplains · « **Reply #182 on:** 14 Jul 2012, 12:12 am »

Btw, here is a quote from John Curl on the matter of bypassing, err, lousy capacitors (from his Blowtorch thread on DIY):

Quote

For Vendetta Research, I had to use a more sophisticated approach, in order to make it extremely low noise, and not contribute a sonic 'character' to the circuit. I found that ANY aluminum or tantalum cap that was used directly across the input circuit power supply would change the sound. Therefore, I had to find a quality film cap that would work OK.

serengetiplains · « **Reply #183 on:** 14 Jul 2012, 12:17 am »

And another:

Quote

Once I heard differences in electrolytic caps in my Vendetta input circuit (thanks, Peter Morcrieff for showing this to me) over 20 years ago, I had to design them out. The 0.1 polystyrene cap on the output is not doing much at audio frequencies, and it sounds good as well, for some reason.

And here's a quote re power supply regulation:

Quote

The power supply is the biggest part of the active circuitry for the BLOWTORCH. It is composed of 3 passive stages, then 3 active stages of regulation and noise reduction, on each channel and each supply voltage, before reaching the actual gain stage that amplifies the music. The gain stage is about 2 square inches (rough estimate) for each channel. The rest is servos, and power supply.

And another:

Quote

I can hear the 'sound contribution' of the output electrolytic cap when it powered my folded cascode circuit. As an engineer, I would not have thought it possible, but I heard it in a direct AB test with a large value film cap, that sounded better. This forced me to use another approach ...

serengetiplains · « **Reply #184 on:** 14 Jul 2012, 12:30 am »

And here is JC on DA:

Quote

'Smear' is a pretty good description of DA. I once used the term 'echo' for DA effects, more than 20 years ago in a LTE to 'Wireless World' or 'Hi Fi news'. I never heard the end of it from Doug Self, etc. , but 'echo' is a good first approximation. It is signal AFTER the original input has gone, and there should be only silence.

cab · « **Reply #185 on:** 14 Jul 2012, 01:22 am »

JC is also an ardent bybee fan.....

serengetiplains · « **Reply #186 on:** 14 Jul 2012, 01:52 am »

So am I. I once put one one the output of my OR5, and yes it created presumed reflections I could hear as jitter---so much so that my Tact would at times lose signal lock. I didn't like the jitter, but the sound was improved in desirable ways. The overall effect was almost a wash.

cab · « **Reply #187 on:** 14 Jul 2012, 02:04 am »

Must have been the quantum gravity well disassociating the matter and anti-matter....

But seriously, with your ability to precisely hear and pinpoint all this minutia, you must have the most sensitive hearing of anyone I have ever heard of.....

serengetiplains · « **Reply #188 on:** 14 Jul 2012, 02:10 am »

1/f noise is probably closer.

Ric Schultz · « **Reply #189 on:** 15 Jul 2012, 09:11 pm »

It's in the fine details where major progress will be made to make these modules more pure:

1. All brands of surface mount resistors have a different sound (they all measure the same). Did Bruno A/B several brands to find out the one that is the most transparent? For instance, I tried a Dale surface mount resistor on the output of one of my clocks (100 ohm) and it sounded way worse than a Susumu resistor. Both are .1% thin film same size (0605) resistors....look identical. So, there might be major gains in replacing the resistors in the module.

2. The output filter caps are metalized mylar with steel leads. Also they are not oriented so the outside foil is toward ground. Better caps properly oriented here could make a major difference. Even larger more inductive parts here could make the sound much better. Is it better to have a super low inductance part that is non linear....or a higher inductive part that is more pure? I bet the purer part will give the better results. Obviously, you would not want the leads on these parts really long or the module might be unstable. Lots to play with here.

3. Bypass caps. All the amps that recently have been said to give better air, harmonics and slam versus the N-core I bet are using all film and electrolytic caps in the power supplies. Probably no ceramic caps in these other amps. Ceramic caps can sound very pinched sounding. The tiny surface mount ceramics are low inductance and fast but they still have the "ceramic" sound. There might be a couple of tiny ceramics on the input of the amps as well. Removing nome of these ceramics and changing to Teflon, styrene or modified Wima, etc. could open up the sound mucho.....and of course, just bypassing the power supplies with added Teflons as Serengetiplains is doing could also improve the sound. However, I bet the amp would sound better without so many ceramics on it.

4. The output coil: Looks like litz wire to me....this is very good. However, would a toroidal cored inductor sound better? When I was modding the first generation UCD modules I changed the output coils to toroids wound with litz wire. It certainly was better. However, I don't think the first generation modules had litz wire in the coils.

5. Electrolytic cap sound: I remember changing the main power supply electrolytics in the original UCD modules out for some Rubycon ZL?..... was way too bright and nasty, so I put the original caps back.....so these caps contribute to the sound of the amps. This is another thing to try.

I will be starting my work on the output filter caps and then work on the power supply bypasses. The resistors are a hard one as there is no schematic and it would be hard to figure out which resistors are the most important....not to mention the difficulty of soldering on such a tiny board.

Phil · « **Reply #190 on:** 15 Jul 2012, 09:45 pm »

with all due respect to those with considerable modding skills, it might be very worthwhile to check out the links in Jason's post:

http://www.audiocircle.com/index.php?topic=108167.0

This is not to imply anything cannot be improved, but that many, many years of experimentation and development has gone into the little Ncore modules.

serengetiplains · « **Reply #191 on:** 15 Jul 2012, 09:53 pm »

Phil, I only monkey with gear I consider first rate.

kevinh · « **Reply #192 on:** 15 Jul 2012, 10:44 pm »

Quote from: Phil on 15 Jul 2012, 09:45 pm

with all due respect to those with considerable modding skills, it might be very worthwhile to check out the links in Jason's post:

http://www.audiocircle.com/index.php?topic=108167.0

This is not to imply anything cannot be improved, but that many, many years of experimentation and development has gone into the little Ncore modules.

You are correct Phil Bruno discussed the issues surround the PCB layout and the effort that went into it controlling the emf and so on.

I would recommend that the 'modders' listen to the designer talk about the amp he designed.

Playing with the wiring going to and from the board, enclosure and so on is fine. tampering with the PCB is a no no if you want to keep the performance.

Phil · « **Reply #193 on:** 15 Jul 2012, 11:11 pm »

splains,
no criticism meant, just that Bruno's talk, as Kevin points out, might provide useful information about tight tolerances and the like. It would seem that Bruno has tried and measured quite a few options and designs what does not meet the needed specs.

cab · « **Reply #194 on:** 15 Jul 2012, 11:30 pm »

It also might be fruitful to pay attention to his comments about power supply and ncore interaction......

serengetiplains · « **Reply #195 on:** 15 Jul 2012, 11:40 pm »

Phil, cab and kevin, I understand where you're coming from. Apart from adding some off-board bypass capacitors and regulated supplies, I personally don't plan on changing smd components or board layout, except perhaps to replace a few critical resistors with Vishay smds if possible. I'm unsure I can do so as the smallest Vishay smd is 0805. But in that event, maybe I can scratch some traces to fit these. We'll see.

*Scotty* · « **Reply #196 on:** 16 Jul 2012, 12:13 am »

serengetiplains, I not sure you can answer this,but has the HF "lift" you heard due to the pwrs bypass caps subsided yet? I'm not sure if you have the bypass caps in place on the bridged Ncores or not. I was thinking that if the amps were left on 24/7, the hours on the bypass caps would quickly mount up and the breakin irregularities should be going away.
Scotty

serengetiplains · « **Reply #197 on:** 16 Jul 2012, 12:59 am »

It did subside ... maybe 50%. The amps were sprawled on boards in my living room, so I dismantled them a few days ago. In the time they were operating, HF lift and stridency reduced as I thought they would, and due, I suspect, to capacitor break-in. Teflon capacitors to me typically sound awful when first used---edgy as hell despite immediately performing better in other regards. I typically have to prevent myself from judging their use a failure, and just be patient. Then after 100, 200, 300 hours ...

I was also using a cheap DAC which in headphone listening I found edgy (for me predominantly a HF phenomenon).

All that said, I liked what I heard. Given probable further improvements in capacitor operation and a better DAC, which will be an R2R topology, I think HF lift and edge will be sufficiently ameliorated to allow rather enjoyable listening. Btw, the amps sounded wonderful at low volume where HF stridency didn't so much irritate.

serengetiplains · « **Reply #198 on:** 16 Jul 2012, 01:15 am »

I was also using cheap speaker wire, psu wiring was not ideal, etc.

bruno · « **Reply #199 on:** 16 Jul 2012, 07:37 am »

I see I have so much to learn from the experts who really know what matters: big caps and fancy materials, the sort of things that are simply over the head of the poor bloke who merely got the thing to work.

Still, from this humble position I would like to offer some interesting tidbits:
*the power rails are LC filtered. Not much PWM current gets out through the supply wires, it's mostly just the average (audio frequency) component. It's called "decoupling". As in "making sure circuits are no longer coupled".
*The electrolytic capacitor bank was designed to have a damped impulse response. Also, the series inductance of the capacitor bank as measured at the power stage works out as <1nH. The large film capacitors shown in this thread have an inductance that's fully determined by their size and attached wiring, and are probably more like 100nH. They won't be very effective bypass caps. That's putting it mildly. To get low inductance, get a small loop area which is not compatible with big components.
*DA (soaking) is irrelevant in DC applications. Besides, it isn't in any way improved by adding a small low DA cap across it...
*FET linearity is not dependent of heatsinking. FETs just aren't linear at all. Besides, Ncore is a class D amp... Not much sense in making FETs more linear then.

Give me some credit. There's still some wisdom left in this old brain.