Class D Kits Compare well against traditional linear amps?

Status
Not open for further replies.
Re: Re: Re: Re: Re: I would have to ...

soongsc said:


What linear amps did you think were the closest to the UCD?

Between Threshold T-200, Aleph30, Tripath, Panasonic SAXR70 the UCD is closest to the Threshold T-200. The Threshold T-200 is 100 watt channel SS Class A push/pull.

If there's a linear SS amp that someone believes is even slightly better than the UCD, I'd like to know what is is...

I'd love to hear a comparison of the Pass Firstwatt F3 and the UCD.
 
I now have a New Class D amp up and running. This is the first time I was able to listen first hand to a class D module. Despite all the rave about class D technology, I had some reservations as to how good this kind of amp really can sound. However, here is what I have experienced.

When the modules came in, it really took me by surprise and I had to scramble around to get two 24V 4amp transformers to work with right way. Assembling the modules took only one evening to get it up and running. First I looked at the outputs and was really worried about the 500KHz 2V p-p on the scope. But Lars told me it is normal, so I figured it must be working okay.

For listening, the source was an ABOSS DVD player HD-9000, the preamp is an Audible Illusions Modulus 2C, the NCD is using two 24V 4amp each transformer, inputs were configured into a switchable polarity configuration, the speakers wasAugene Lullby (my design, just getting ready for release when manufacturing process is stable). The reference amp is a Hafler XL-280 made from a kit, no special component modification, just left out the fuses from the speaker outputs. First impressions were as follows:
1. The gain seemed high, so any noise from the source chain comes though clearly. I was hearing noise originally inaudible while using the XL280.
2. The sound was balanced and pleasant for background listening.
3. In the beginning, the hall timbre is was not audible, but as the system warmed up, the timbre came out. It probably took about 30 minutes for it to get to a more stable operation condition. During this warm-up period, you can hear the piano sounds gradually change from muffled to more natural.
4. The detail was a bit lacking, this is noticeable in all instruments. This might also be referred by some people as a little ¡§dry¡¨. In piano (Hayman plays Duke Elington, RR-50DCD), the part where the key initially strikes seem a bit muffled even after 30 minutes, this also makes the piano not sound as dynamic as normally expected. The same is noticeable in the tambourine (Jazz at the Pawn Shop), the detail when the hand hits and the little things ring are slightly lost. This is also noticeable in timbre as well as reverberation (Jazz at the Pawn Shop), it sounds like everything is more in dead room, the difference between these recordings and old recording made in old studios are less.
5. Transients were not as sharp and quick. It sounds like some of our earlier versions of speaker with an average capacitor and inductor in the circuit and cone edge flexure induced effects.
6. When the polarity is switched, the polarity difference in sound is not as significant as when using the XL-280.
7. One funny thing is, although the gain (45db)is much higher than the XL-280 (sensitivity is 145W with 1.6Vrms input which is about 26.5db gain I think), it does not sound louder at the same volume settings.
Switching to the post filter feedback mode, first, the blue LEDs will flash and the trace on the scope goes way out of range when no load is added. I¡¦m not sure what this means. And it does sound like Mr. Nice guy as Lars had mentioned before. It¡¦s relaxing, but does not sound real at all. Reverse of polarity is even less distinguishable than the pre-filter feedback mode. More noise gets coupled in for some reason, and it changes if I move the wires around.

I knew I had to get higher voltage transformers to let the NCD operate at a more optimum condition. So I went and ordered acquired one O-core and one R-core transformer each rated at 400VA using Z9 core. The first transformer that went into the system was the O-core. Some first impression with the O-core transformers wit the amp still in post filter feedback mode just listening to ¡§Jazz at the Pawn Shop¡¨ and ¡§NoJima Plays Liszt ¡§(Reference Recordings RR25CD):
1. The power and dynamics are improved, but still sounded a bit like it was not coming all out free. It sounds like transients being soft clipped.
2. Absolute polarity difference is very obvious, but I think some may prefer one way and some others the other. The correct polarity reveals better harmonics of the piano, especially the initial ringing of the strings right after the key hits on the strong notes. The other way around the strikes are softer rounded, but you feel more power in the piano.
3. Taken the gain is 45db, the loudness still does not seem that much louder, but the noise passes through very easily. My speakers are only around 85db, and wire movement in the chassis you can hear noise change.
4. The detail was still missing. Tambourines still lacked that initial impact realism, and the trailing rings gets mixed up.
5. Focus and image width is correct and quite good. But the sound is too clean and lacking air. The delayed reverberation of instruments are lacking as well as the trailing timbre.
I thought it would be a good idea to have the same internal input wires as the XL280, but after switching to the same internal wiring used in my Hafler, the difference is not that much.
Switching back to the pre-filter feedback mode, the reverberation and trailing timbre is improved, but instruments such as cymbals and tambourines still get s little mixed up and the onset transients are still a bit lacking. But the music is enjoyable in this mode. At this point, the NCD seemed very close to the Hafler XL280 just missing it by that little detail and transient onsets.
I tried to run a null difference test, but it could not be done. The amp just kept turning itself on and off. I also took the preamp out of the loop because it was the source of the noise, but no further improvement in the comparison.
At this point, I was about to give up. Then I looked through the NCD cookbook and found it mentioning that an aluminum case is necessary to minimize the chassis effect on the output core magnetic field, so it made sense to get a new chassis to try out.
Not having listened to the NCD for a while, just before I was going to switch to the aluminum chassis, I was a bit curious as to what would happen of a reversed the leads in one of the transformer 42V, so I did that and powered it up. To my surprise, that little lacking in detail and transient onsets disappeared! Now listening to the NCD feels just the same as listening to the XL280. If I had to make a choice, I would have to upgrade my speakers first because I think the speakers will prevent a more accurate comparison. I have not tried the R-core transformer yet, but my guess is I would need speakers with better low frequency performance to notice the difference.
Based on the NCD cookbook recommending using aluminum chassis, I switched to an aluminum chassis, changed the O-core transformer to an R-core transformer, and added multi-input selector. In this configuration, the sound was just not right. Much of the bass was lost, the sound was not real, and switching absolute polarity yielded no difference in sound. It sounded like when I started out with two 24V transformers. So I decided to first switch back to the O-core transformer. The first thing I noticed was the 4A fuse would blow out when I powered up; this did not happen when I used the R-core transformer of the same rating. So this means there must be significant transient characteristic differences. After changing the fuse to a 6A rating, the system powered up smoothly, and the sound is back to what it should be from a brief listening session.
So now I believe that class D technology is good enough to be called HiFi, and based on the things going on, I think we will see even more improvement in this technology, and the real limiting device would still be the speakers.
 
Daveis: " ... Between Threshold T-200, Aleph30, Tripath, Panasonic SAXR70 the UCD is closest to the Threshold T-200. ..."

I have a Panasonic XR70 as well, plus a set of AussieAmplifiers.com NX-150 . How would you rate the Panasonic as a comparison?? (Any guess about a comparison w/ the NX150?)

Also, I appreciate that the UCD is a good design, are there any specific areas where you believe that the UCD excells?? Specs?? Audio quality?? Size or configuration?? .. etc.

For comparisons sake, what is you test or comparison source? (vinyl TT, CD player, DVD-A, signal generator, etc ...)

:smash:
 
The Panasonic SA-XR70 receiver to me is a slightly better than average sub $500 receiver. It's fairly neutral sounding, but in particular it seems to have problems driving 12" subs with any authority. For the time being I am using it as a tweeter amp. As an added bonus, the headphone amp isnt too bad.

I would be interested in whether you thought the Pansonic was as good as the NX150.

At times when using 3 SA-XR70's in a 3-way active PC crossover system I thought the sound was going to keep me happy for awhile.... but I grew discontent with it.

My main issue with Class-A amps is the heat (especially when using 8 channels). At one time, I had 6 channels of Class A amps, but the heat was getting to me(over 1000watts). I could only really run the setup in the winter. During the summer I'd switch back to the Panasonics. Then I discovered UCD. UCD to me is the first digital amp I would consider as a replacement for my Class-A amps. The first words that come to mind are clean, accurate, effortless power with a good dose of detail. While the Tripath's do things that I appreciate, the sound of the UCD is more familiar to me... more like the other amps I've got.

The most endearing thing to me is that I can use a UCD amp alongside a good Class A pushpull amp and not feel too bad about the sound.

I've gotten to the point where even if the Class-A is a bit better than the UCD, it's not enough better to warrant the heat and losing the benefits of active crossover.

The comparison I'd be interested in is between UCD and NX150.

My comparisons have been mainly CD. I have a few DVD-A's.
 
Hi to all,

I have an old Audio research D-100B magnificent powerfull sound, but a bit unforgiving and lacking detail. I have two 30W class A hybrid monoblocks after 'John van der sluis' design, very musical sound with a lot of presence. It has a switch to go from class A to A-B and with a lot of effect. All these amps I bought when I worked for a hifi shop for several years at the service department. But with these amps detail and precision leave great wishes open when compared to the Tripath amps I own (Audiodigit MC4x100, SI 5066:dead:, eight AMP32 amps for all my friends....) they can only stand in the shadow...

The latter are so well detailed and seem as if to smooth out every shortcomings from my humble speakers, the precision and soundstage is simply overwhelming. On top of this I feel (and I don't feel this quickly...) comfortable listening to these amps for long sessions...

The first one I tried was a SI (offcourse 😛), this one made me go through all my CD's with my mouth open (drooling)...unfortunately all of them I owned have deceased...:RIP:....but there was something unpleasant also, the sound struck me 'fearse' at some times and gave fatiguing effects. Especially the mid-range, the highs seemed more ok actually.

The next one I bought is a four channel amp from Audiodigit and is never fatiguing, quite powerfull and sounds a lot more grown up. But still highly detailed and precise (especially in the lower frequencies). The only down side I can think of now -after months of listening-is that it strikes me as a tad dull when I compare it to my new AMP32's. The amp does seem very durable, I tortured it quite a lot with shorting it and connecting ESL loudspeakers, taking it to several places. It seems to be able to handle a lot of abuse....
Awaiting a new package from 41hz.com with two more AMP32 kits AND an AMP9, which uses the same chip as the Audiodigit but costs 1/4 the price!

The AMP32's sound really exciting and musical, though they seem to be out of breath a lot quicker (offcourse) when compared to the Audiodigit. At moderate listening levels-tested this on 85dB speakers...- the sound is very very good detailed and with excellent
soundstage. I think a couple of high sensitivity (hifi) speakers can show the potential of this amp a lot better. I did connect the little amp to two large PA subs (Community's) and high sensitivity PA tops (Granite) at work (I'm a broadcasting audio engineer) and it rattled the cages (at the other end of the industrial area 😀 ).
My collegues were all very surprised (many DIY-ers there in the technical department).

It's just pretty hard to describe how revealing class D (T😛) has been to me, I have listened to all kinds of manufacturers designs on a switch panel (not ideal offcourse) in the shop where I worked for years. Had Mark levinson, Dyson (bad by the way), Vanmedevoort, Quad, Denon (also bad...), Linn, AR, etc. -amps combined with JMlab,
B&W, Chario, Dali, you name it brands at the push of a button there... But the first time something shocked me was listening to a pair of Esl-57 speakers (knowing how old the design was...hey Martin Logan does the trick, but can't afford them frankly...) and the second time something shocked me was the little T (just forgetting the down-sides for a minute...)

D-amps DO for sure produce a picture of reality a lot better in many ways than anything I ever heard...it just makes me feel a lot more pleasant listening to my records than before....there's a certain honesty to the sound... a sense of purity and easyness (to produce instrument on top of instrument without standing in eachothers way....).

I wonder how the standards of fidelity will move in the future, can it get much better? I bet it can.....
 
I have not traced the NCD circuit board, but my initial guess would be the electro-magnetic effects in the transformer. This makes me think whether I should go back and give the R-core transformer a try by swapping the secondaries. I am also guessing that whichever way the startup current has the largest peak, then that should be the best way to connect. If someone does a measurement and hears a difference, please post it.
 
niiico said:


Can you please tell us more about this?

Measures, physical facts, how to determine the right connexion, etc.


Please look at the picture (if I succeeded to attach it)
 

Attachments

  • picture.jpg
    picture.jpg
    81.3 KB · Views: 585
soongsc sounded like one-in-a-billion, ok, maybe 1-in-a-million.

Highly subjective listen may be the reason for low volume in high-end gear besides the price. It's hard to get two audiophiles to agree on the quality of a system. The amp only contributes a small portion of sound reproduction quality. Speakers and room much more.

However audiophiles swap lower cost items (amp, op amp, cables etc...) and not speakers or room because they're lower cost.
 
Daveis: " ... The Panasonic SA-XR70 receiver to me is a slightly better than average sub $500 receiver. It's fairly neutral sounding, but in particular it seems to have problems driving 12" subs with any authority. For the time being I am using it as a tweeter amp. ... I would be interested in whether you thought the Pansonic was as good as the NX150. ..."

A tweeter amp? seems like a waste of those extra channels :bigeyes:

As for my comparisons re:
Panasonic XR70 v. AussieAplifiers.com dual mono-block NX150s
<100W @ .01 into 8 ohms v. <150W @ .005 THD into 8 ohms
class-d > 5 or 8 channels v. class-ab Power MOSFET into 2 channels
lots of features v. line level input amplifiers
US$350 to 500 v. ~US$330 DIY modules w/o case or PS

Panasonic xr70 Listening: decent to good quality, good clear output into a variety of speakers. I notice some "fatigue" after lengthy listening at higher levels into 5 channels. Low level listening requires high sensitivity woofers. Many equalization options and channel mixes require various adjustments for various, alternate sources. "Pure" stereo from 24/96k sources through two decent mains is an option, but some what anemic in the base range (gutless?). Quality L/R mains plus decent L/R "surrounds" and center for best results = very good Home Theater sound. Suggested tweaks for 2 channel mains only, would be more capacitance or better quality caps in the PS and some way to "decouple" the output, although the digital controller would never allow full rated power into just two channels, even when "bridged" (one of the many options).

NX150 Listening: Top quality output into healthy, high response mains. No "fatigue" noticed when power supply has plenty of filtration/capacitance (40,000 uF per channel in tests). Low level sound is superb, clear without noticable distortion at any frequencies, 5 htz through 25k htz. (My 'scope" is not good enough to see any.)

This is a bit of an unfair comparison = a multichannel, mass produced class-d verses DIY 2 channel class-ab power MOSFET kit. The very real audio quality differences are tainted somewhat by my subjective observations = I did not expect the Panasonic Home Theater receiver to keep up with the NX150s in stereo only ... that's not what the Panasonic was designed to do.

(My wish list now includes something like a Panasonic "home theater" controller pre-amp (for all those great features) that I can hook to several higher power amps for 4 or 5 channels and some bi- or tri- amped speakers. The XR70 just has speaker outputs with a huge variety of source inputs, but no "line level" outputs other than a "tape I/O".)
 
koolkid731 said:
soongsc sounded like one-in-a-billion, ok, maybe 1-in-a-million.

Highly subjective listen may be the reason for low volume in high-end gear besides the price. It's hard to get two audiophiles to agree on the quality of a system. The amp only contributes a small portion of sound reproduction quality. Speakers and room much more.

However audiophiles swap lower cost items (amp, op amp, cables etc...) and not speakers or room because they're lower cost.

I'm trying to get more exposure to all different kinds of systems and just listen to understand when I hear a difference, what could technically be causing it. If I find the technical reason and can fix it, it just brings a system a little closer to what it should be.

Having read somewhere that the Nautilus being the most accurate speaker a magazine editor had heard, recently I listened to a pair of Nautilus just to find out how good it sounds. It sounded good, but for some reason not really as good as I would expect from the looks of the design. There are some places where cone resonance can be heard. I think it would be a blast if the drivers could be slightly modified.

I think we get lost in all this "taste" thing until we suddenly listen to something that really is different. About a week ago, A few people gathered at an audiophiles listening place to listen to some speakers and amps (one tube amp was interesting, you could plug in whatever tube that will fit, and it would still play quite well). Each speaker had a different sound, and everyone was traing to say good things about each one played. I kept my mouth shut because each did have very significant characteristics, but really was in no way near to being correct. When they stuck my little speakers up there, everyone became quiet, I don't know what it means, but I kept quiet too. After that some people just told me the speakers were really good considering the little size, and another asked me if I could to an enclosure for a set of driver he had. One person did comment to me privately about the timbre of some instruments which I pretty much agree with.

I must say that many friends provided me comments in a truthfull manner that help me learn more how to detect possible areas of improvement along the way. This is something that anyone can learn if an open mind is always kept.

I'm not sure what the situation is in other countries, but here, many really high priced equipment are not easy to get an ear on, and I probably know the reason why now. Once systems get bigger and complicated, it is very hard to get it to sound correct, it just sounds bigger.
 
soongsc said:


Have you actually measured it to be like that? I'm under the impresson that the ripple is always the same direction because of the way the cap discharges during the negative slope of a half sine wave.

Of course I have looked at these wave forms with a dual trace oscilloscope.As in the correct wiring the ripple is in opposite phase on + and - voltage they cancel each other to some degree I believe.
"Correct me if I am wrong". 😉
 
Anzgar said:


Of course I have looked at these wave forms with a dual trace oscilloscope.As in the correct wiring the ripple is in opposite phase on + and - voltage they cancel each other to some degree I believe.
"Correct me if I am wrong". 😉

Now I'm tempted to stick it on my scope the next time I switch around.🙂 Maybe it's a shift in timing rather the shape itself?
 
soongsc: " ... I'm trying to get more exposure to all different kinds of systems and just listen to understand when I hear a difference, what could technically be causing it. If I find the technical reason and can fix it, it just brings a system a little closer to what it should be. ...
... I think we get lost in all this "taste" thing until we suddenly listen to something that really is different. ..."

Totally agree about the subjective v. objective v. taste thing.

I thought I had a decent sounding setup until I listened to my cousin's US$30K high res speakers and $US ?? K tube amps and US$18K turntable in his US$ too many K custom listening room. ... I may never even come close to his setup in this lifetime :bawling: . The positive side, I get to visit him about three times a year for a listening session, just to keep everything here in perspective. 🙂
 
Status
Not open for further replies.