DIYHiFi.org

What better to put the UCD400 up against, if not a former version of itself?

My collection of amps (and their current duties):

-The "Classd4sure"Homebrew 35 watt P2P UCD (honorably discharged)
-Hypex UCD 180ST (retired after suffering injuries in the course of duty-AKA death by butane iron)
-Hypex UCD400AD (serving another tour)
-Hypex UCD400HG (graduated from boot camp)
-Hypex UCD700AD (in the reserves)
-NewclassD “prototype” (Private Pyle is currently MIA; last sighted camped at WalMart, in greenery)

Other system specifics:
-Listening Room: typical worst case
-Speakers: Cerwin Vega VS-150 400W 4ohms 102dB @ 1W/1M
-Source: Computer based, EMU-1820M POS DAW
-Wires: 10AWG Zip wire for speakers and some cheap mic cable for interconnects
-Ambient dB level: Just right, between 3 and 6AM

"So what's the deal with the HG, any good?"
"What's with all the lights, like, does it, you know, do stuff??"
"Is it worth the money?"
"Is it a substantially new version or…?"
“What if I got a cheaper version and modified it like the HG, would it be as good?”

It's never been a secret that these UCD modules as they were sold to the do it yourself community weren’t up to their own potential. Hypex was always honest with that fact, and they were just as honest in disclosing the areas of interest concerning unlocking said potential. In the very least, we knew for certain it was a task worth undertaking, given a direction to follow, and had a solid foundation to work with. It proved all too tempting.

Here's a vague list of the modifications on my 400AD modules as they stand to give you some small idea of what I’m now dealing with:
-DC coupled
-LM4562 op amp
-Current regulating diodes biasing the LM4562 to their appropriate rail
-Current regulator diodes biasing the zener references for the onboard regulators
-Different decoupling network for the onboard regulation
-Different bypass caps
-Different bulk storage bypass caps
-Additional bypassing & decoupling re-rework

It’s a short list for a seemingly never ending task. Some of the modifications are more absolute than others in terms of performance. For instance DC coupling is a no brainer, if at all possible, or in the least doing something about the stock decoupling should be considered mandatory, but the selection of caps can be somewhat of a nightmare. This is especially true for those which must be electrolytic. A wide assortment at hand, and in the very least good availability, is realistically what’s required here to manage such a task even half proficiently. Lacking that as we do, we shouldn’t be too surprised to find it a difficult if not near impossible objective to find a result satisfactory enough to ever really consider it as a being a done deal.

I first tried a set of bulk storage caps with these modules that I knew for certain were at least decent, and certainly better than the stock ones, at least for what I was trying to achieve at the time. Certain things it seemed they did very poorly though, like drop loads of detail completely.

After several months of trying I was finally able to obtain the replacements I wanted and was at first overjoyed. They weren’t perfect at all either but featured a lot more benefits than not, as compared to the others I’ve tested, so they were a step in the right direction at least and a very welcome change after the six months worth of enduring the others.

I then happened upon some Hypex audio grade caps, the very same ones that are now found on the new HG modules. These proved to be a new plateau on the quality scale. A better cap than my "better cap”, without question, and while they weren’t dialed into the circuit by any means, this was immediately obvious.

Noise, coloration or odd artifacts were absent, seemingly entirely. What they didn’t lack at all that most others seem to was timbre, clarity, and detail. Now it was able to manage complex symphonic passages that would normally be less than appealing, more like thin wisps of lifeless noise than the rich, full bodied texture that was now shinning through in just the right way.

Vocals were as good, colorless, clear, full of body and lifelike, while pleasantly, but not overly smooth, just clean and natural, and it was starting to sound very convincing. Bass response was tight, hellishly potent with unusual clarity and depth. In short, performance was such that they shinned the light towards a new direction in which the system must be taken, one that includes these caps.

Of course it wasn’t “there” yet by any means. A small number of issues I had were obvious and simply amounted to the perhaps usual requirement for some fine tuning of other passives to best settle these caps into the modules. However there were also other odd factors that I initially had difficulty with putting my finger on, perhaps because I wasn’t even sure if anything I was hearing, and even not hearing, that didn’t rest well with me, could be directly attributed to the caps.

All doubting, scheming, and second guessing everything came to a sudden end some days later, after I tried playing it at a higher volume for an extended period, and paid that price with tense, burning ears. The following day I hesitantly turned the amp on again and that same pain returned, this time within minutes instead of hours. You can be certain there’s some real problem that exists when met with physical pain. The tired and panicked response to the pain was to swap in the second best caps I just recently enjoyed, which reaffirmed what deep down I already knew, that I’d already been completely spoiled by the performance of the Hypex caps. From them there could be no retreat, I had to find a way for them to live in my system.

I thought early on that I was possibly experiencing somewhat of a compound issue that was only now being brought to light given this higher level of clarity and definition. Interesting as the details may be it would make for too long of a post, but suffice it to say that turned out to be exactly the case. From the realization of exactly what was going wrong, it didn’t take long at all to solve the issues one by one and the Hypex caps found a permanent home in my modules.

I should say at this time that the audiodock of the 1820M has also undergone similarly sensible upgrades. This was required to ensure that it wasn't presenting itself as the bottleneck of my system, or as big a bottleneck, I should say. To ignore the card while working towards improving the amp seems to me like working out your left side only.

There’s now two pairs of modified channels on the audiodock to compare with one another, component by component if need be, along with a few sets of stock channels which serve to base myself as to the sound of the flattest response, ensuring that I’m tweaking the response of the amp by tweaking the source as little as possible.

The two pairs of modified channels are kept only slightly different from one another which help me see how the amps might sound when driven from a different source, and it’s quick to swap over. The single difference between them is low ESR VS general purpose bypass electrolytic caps on the output stages which produces the kind of difference you would expect from that, both being high quality, semi decent sounding caps.

By the time I put the modules down long enough to let some dust collect, they weren't AD versions at all, but sensibly modified modules that have come a long way over their stock form, but are still never quite “there”. Much of the credit for their present level of refinement has to be given to the Hypex caps though, which are the best I’ve had the chance to try and by no small margin. I’d quantify the kind of improvement they make right along with say, DC coupling, the LM op amp or CRD biased zeners.

I first opted to plug in the new 400HG modules into my existing power supply for evaluation which isn't a scientific blind test, but nor does it have to be. It serves me well in that only one parameter in the system changes allowing me to evaluate them relative to what I'm already familiar with, and this is enough to get the feel of it.

The existing power supply is by far no slouch, consisting of a 500VA Plitron with an epoxy filled center, static screen and magnetic shielding. It feeds a pair of 68A FRED monolithic bridge rectifiers which are snubbed by an R+C laying flat across them, and in turn feed paralleled Jensen 15 000uF four pole cap networks that are wired in a particular fashion which provides forced equalized charging/discharging paths, or just very low noise and great coherency. The enclosure is 1” thick gold, well actually it’s just a steel plate from an old receiver enclosure rigged up as a test jig that I’ve grown attached to.

The power rails rest around 47Vdc as this was originally designed for the UCD180 modules. The modules were wired to it in stereo and set up in opposing phase to help with low frequency reinforcement, and help tame any possibility of supply pumping, especially since the now DC coupled audiodock introduces some offset. Of course the HG’s were connected to it in the same fashion as the older modules.

Sometimes it's the smaller things that leave you with the bigger impressions. One of the very first things I noticed of the HG modules was that the ~1.5s turn on delay is gone, as it only occurs on initial power up, so now the /enable functions instantly. Even though I understood the function of the circuit and knew for certain this was never any kind of issue at all, I was never able to shake the cheap feel it gave having speakers activate at different times for having thrown a single switch, even if they only differed by a split second, anymore than I could not spend that ~1.5 seconds thinking of the worst possible outcomes. Maybe this is conditioning, but I’m certainly no optimist.

When I throw the switch now both speakers are active that very instant, and by far the only sound that can be heard is that of the switch itself, pop-less (de)activation as always with UCD. This all leaves me with a reassured feel of a serious, dependable high quality toy and leaves me no time for doubting it. I find that’s a much improved “first” impression, and one that’s revisited each time I listen to it. It’s just more polished and sophisticated, so nice move!

The AD modules never got to keep their coupling caps for even a first trial. I’d seen that on the 180 modules already and wasn’t interested in hearing those electrolytic caps again. With the improved setup found on the HG modules I thought it stands some chance at sounding alright so I figured I would give them a listen.

This was some of the most correct and convincing sound I have experienced from audio gear. It also introduced good regulation to me for the first time, via the HXR regulators, which still impress me to no end. The soundstage is rock solid, with a strong sense of real ambience, that all arrives to the ear in perfect time, giving you the sensation it’s all happening right there. It doesn’t lazily emanate from the speakers off in the distance, or from reverberating all around you, it solidly creates a high degree of ambience and brings it right to you.

Between being musical or analytical, it’s just neutral. I find it walks that line between the two perfectly, where it’s not at all sweetened or hindered, just pleasant and involving, the sort of thing you can easily listen to all day and keep you smiling as long. Should you find it too far on one side or the other you might consider looking at your implementation of it.

I managed to keep it AC coupled for almost a week and enjoyed every second of it, giving me a thrill every time I listened and I always walked away pleased. To say the least, no longer will people who can’t solder and are forced to contend with AC coupled UCD modules have to suffer outside of the pack!

However the curiosity of DC coupling got the better of me before too long, film caps might be superior to electrolytics but they aren’t a wire by any means. Replacing them with one really exploded the sound in every way and every direction. I told a few people of my initial impression when I first heard it AC coupled, that it dropped my jaw and blew my mind within the first five seconds. I had some high expectations of them given the latest level of performance from my modified modules, which it exceeded easily even while AC coupled, embarrassing the sound I’d come to know, and now just days later I was picking my jaw up off the floor a second time and nodding in approval.

The low frequencies never lacked in reach, but now picked up some bolder power. It’s able to hit those notes that make one feel uncomfortable with authority and insanely tight grip that’s just fun to experience.

The level of air conveyed is simply correct and convincing whether it’s with vocals, wind, strings, or cymbals. One of the neatest things is the impression of hearing the sticks themselves as they crash into the cymbals. Give 10000 Days Wings Pt 2 by Tool a try and maybe you’ll see what I mean.

The soundstage is just as dimensionless as while AC coupled, only now it has exploded and dissolves the room entirely, replacing it with pure ambience that only changes along with the album being played. You really get the impression of true ambience as it's recorded into the source, it’s a treat.

This amp has good authority and remarkable control. The level of refinement towards a convincing reproduction and natural, correct sound is astonishing. Hypex did an excellent job of voicing it via their selection of components. The skin on this drum is tight and it sounds like it. This degree of refinement could almost take getting used to if it weren’t just so correct, it ends up pleasing out of the box and after just a few days it’s hard to imagine liking anything else.

As to whether or not a cheaper version of module could be modified similarly to the HG, I think you can see now I find it unlikely, for at best with extreme determination, a lot of time, and a lot of luck, the best you could do is get it close, and by no means would such a venture be cheaper. I think it best to save up a little extra and splurge on the HG, at which point if you find it weak in the bass, look towards your EQ and leave the boutique-audiophile caps alone. Ask yourselves how many people with a McLaren F1 take them over to PepBoys or Canadian Tire looking for ground effects packages. The UCD High Grade is one such performer.

In the coming months I've got quite a list of things to try. The HG modules will be enjoying full mono block construction along with the HG supplies and potted Hypex transformers. I’m curious to see how their power supplies will fair against my ol’ beastly T-network tank.

I’ll also be installing the HXR regulators in the older modified modules, maybe even in the audiodock, and doing a little more rework on the modules before seeing how the “AD” and HG fair against each other again, perhaps side by side this time. Based on the improved modulator I expect the HG will always have the edge though.

Until then,

Chris

Chris, excellent report! Your class D work, experience and expertise comes through solidly.

Thank you kindly Tom, but if that's all that came though, I must have missed the mark

There will certainly be more to come in this thread anyway.

Cheers

Hi Chris,

Great report, thanks!

I have a similar setup to yours, modded UCD400 modules (with LM4562 buffers, CRD's, Panny bypass caps, DC coupled) and a slightly modded EMU 1820M. I do use a NOS TDA1541 DAC for the mains, the EMU does duty for SPDIF and center/rears. I do have the luxury of decent speakers (Dunlavy SC-IVa - 10 years old and still the most amazing speakers I have auditioned) and a purpose built audio room, both of which I found improve the listening experience more than amp tweaks.

So you think the magic in these modules come from the modulator changes, as I'd expect the use of better bypass caps & the CRD tweak in the standard UCD to be similar to what Hypex have done on the HG?

Regards,
Dean

Hi Dean,

I think the magic is the sum of the whole. The solidity of the soundstage must be largely due to the HXR regulators however.

Also despite the similarities between the modified module, which are especially strong in my case given that it already had the Hypex caps, the final arrangement of passives on the HG give it a sound that I find is orders of magnitudes more convincing in terms of it being that much closer to having perfect pitch, especially up in the highs where there's all kinds of new ability for timbre.

You can easily change that in it if you start messing with cap upgrades, for instance I know I can give it the same color the older module has which doesn't have the little bit of noise electrolytics have so it's got some additional clarity, but the pitch is off and it doesn't sound near as convincing. It's even got the additional clarity of the crd biased output stage but isn't as enjoyable given that the pitch isn't as right with it.

Needless to say the HG remains stock for now and it's been impressing. It feels like it can hit 10Hz now, one song I've tested with recently has electronic bass that plays with the woofer and you can feel it slowly bring it to a dead stop, then speed it up again. Spanning whatever range of lows it does there's no sense of standing waves in the room, boomy response or loss of control. Everything that's heard is also felt in perfect time from the woofer comming to a dead stop right up to the very highs in a soundstage that just never wavers at all, things the modified AD won't do just yet.

I expect that were they to be voiced identically given all the same passives, along with the additional use of the HXR regulators, the HG should still have an edge in clarity given the modulator itself, but to what extent that's true I can't say right now.

BTW while this thread isn't going to be about tweaking, this might be a little too good to pass up. You already gain an advantage by implementing the auxiliary supply to power the input stage, but you also render the on board pre regulator redundant. In this pre regulator you'll find the CRD's that once relocated will be a nice "free" tweak to bias the output stage of the LM op amp, the solder pads are already there, just use the right ones.. that bias it to the positive rail.

Dean, do you use the analog outputs on your audiodock ? I've discovered EMU's output stage on it has the hopefully unique feature of taking shield noise and feeding it into the output differentially.

Solved by removing the mute transistor on the output for the shield only, which makes it easier to interconnect without issue as the shield stays earthed, but disconnects it from the output that's later inverted to drive the hot output.

Mute still functions too as the the other transistors need not be touched.

Chris,

Sorry for the late reply, I have been off the air with a motherboard failure, so I thought OK, time for a Core 2 Duo upgrade so went out and bought all the bits only to waste more time with faulty new memory modules.... OK, now back on topic...

You have me interested in the new Hypex modules, as it seems your audio philosophy is similar to mine (I'm recalling several threads we participated in with another forum), and we have similar equipment setups. What I don't like about my current Hypex amp setup is that the bass does not feel as impactful as it does with the stock setup and I suspect it is the LM4562. I have added a tant bypass to the LM4562, which from memory I think was a previous recommendation from you, and it has improved a little, but does not have the authority a 400W amp should have. It still moves the woofers with ease, but something is missing. I use a Rythmik Audio DIY servo subwoofer setup for music (and a digital delay using patchmix) and it integrates well, giving me the "punch" and air movement that adds that authority back. Ideally I'd like to just use the sub for HT as my Dunlavy has 4 x 10" woofers which should be up for the job (it does 20Hz at -3db and still going at 15Hz) but my modded Hypex modules don't energise the room as much. Your feedback indicates that the new modules do a better job in the lower octaves compared to the old AD modules, which is encouraging.

Can you also confirm that the HG modules do not have the CRD bias for the LM, I would have thought that would be a mandatory tweak, as it was built into the older AD modules? Assuming Bruno did a lot of component optimising for the new HG, he would have surely used the CRD, and maybe it did not work as well as it did with the older modules?

Re: Analog outputs, yes, I do use the balanced outputs, which is a good match for the balanced inputs of the Hypex modules. Not sure I follow you about the shield mute transistor, are you saying that there is an issue when mute as the shield is not connected to earth? I don't have problems with earth loops.

Also, while we are talking EMU, I have put some time into learning patchmix and it is a seriously good application with a wide arrays of filters, delays and effects that work at the driver level, and are very high quality. I use patchmix to do all the HT processing (bass management, delays, subwoofer mixing) and it works very well.

BTW (and more offtopic...) - you really do have to look at the acoustics of the room to get the max from your setup. After playing around with audio for many years now I have come to the conclusion about the following in order of importance:
1) Speakers. These make the biggest impact. I favour time aligned & well engineered speakers over fancy drivers.
2) Room. I was amazed at the difference this made once I finished my home theatre with significant acoustic treatment (minimised reflections, bass & midrange damping, staggered walls/floor/ceiling). Biggest impact is the soundstage, bass and image placement. With a good stereo recording, it sounds like the sound is all enveloping, and as my centre channel is below my theater screen there is an illusion that the centre channel is on when there is only a stereo source.
3) Source. Has to be good to start with as nothing else upstream will make it better. I go to pains to ensure my 16 bit CD sources ripped to hard disk is coming out of my HTPC pristine, and it does sound good! Unfortunately most of the music I listen to (contemporary rock) has not been well mastered to start with.
4) DAC/AMP. I do appreciate the wider soundstage & more solid bass of Class D, but the impact of the amp and amp / DAC tweaks seem small, but still discernable, hence the interest in improving it. My next tweak will be moving from a NOS 16 bit TDA1541 DAC to a better 24 bit DAC but I have not done the research on the best chip to start with.

Hi,
The LM4562 will shift the overall response to a more correct one VS what the AD op amp reproduced, but as to where the actual problem you have stems from, it isn't the LM4562, swapping it just made you notice something else that needed work.

Looking towards caps was a good idea to try solving it, but that’s where the nightmare begins too. You may also recall that I was still very much working on that tantalum thing with my 180 when my far too hot butane iron fried the PCB on it which brought that story to an end.

You can do worse than a tantalum, but you can sure do better. They add distortion and leave you with a noisier, thin, cold and hard sound that's unpleasant, but they also have admirable qualities, and then it will also depend on the exact selection and final configuration. You can do better with a decent electrolytic easily enough.

The only point of worth to those posts about the tantalums back then was not so much that they were the magic answer, or even something decent, but more so that it was an area worth working on towards improvement. Sorry if I led you to believe otherwise, they really had no other value to speak of, just like most of all the other posts on caps in that thread. I think it was more the removal of the X7R I was impressed with anyway, which might have become the “X7R mod” or something.

Your subwoofer solution was better for sanity than messing with the caps to get it right. Bi-amping could have been another method.. $$. Like you though, with my speakers being what they are, I should just not require anything additional in order to enjoy some bass, and I went for the caps since I wanted a better understanding of things.

"I can't believe your speakers can even play that kind of bass" was something I heard when I had it playing some kind of weird skin stretching drums. That was said before the HG's were ever a part of it too, once I started to settle in the Hypex caps to the old modules actually. My speakers shouldn’t be able to play bass like that so convincly either, but UCD hardly seems to care.

So, you really find that class d has “a wider soundstage”? If you do, isn’t that a little suspicious? It’s not as though it’s remastering what’s on the source right? So what’s the problem then? Maybe it’s the microphonic noise created via the HF ripple in inferior electrolytics that’s messing with the micro detail? I hope you weren’t too in love with it, because this is the type of noise the Hypex caps don’t have! Personally I don’t miss it, it’s been replaced with true, clean, ambient filled micro detail, and I find it to be an acceptable trade.

When limiting myself to higher quality MP3’s I find the bigger factor is still the original mastering. For instance, I don’t care how bit perfect a copy of anything from the red hot chili peppers is, it won’t be played on my system, ever. Better quality MP3’s aren’t far off from the already often horrible originals, so I don’t find the MP3’s are the real crime. However if I play a DVD it’s like a whole new system that I can hardly believe. That’s also why I use certain soundtracks for testing, even if it’s an MP3 of it, you get a much higher quality sound given the excellent mastering then normal off of usual every day kind of material.

About the CRD: It performs excellent with them. The pads are right there for you, and so are the CRD’s if you implement the aux supply to free them up and then swap them over like I said. Ideally you’d want the aux supply anyway, so I consider that CRD mod a nice freebie, as the part would otherwise go to waste.

Hypex has to watch how much gold they sprinkle on the modules or nobody could afford them, and they’re not the cheapest components. The 4562 already sounds fairly clean by itself, so there’s far more to be gained by using them in the onboard pre regulator instead, where they really aren’t an option. Good choice I think.

I’m aware that speakers should be the bottleneck in a system, along with the room. Pratically I think that’s only the case if you already have top notch electronics, but then with top notch electronics, they matter that much less too. I have transients that move around in a spherical soundfield that fill more than the center of the room, where ambience conveyed knows no bounds.

The sound is no longer "all around you", to me that's feeling the room without true ambience. I still feel the room on my modified AD this way, which presently lacks the level of microdetail required to wipe the walls from existance. The bone stock HG does this with ease, where all you know is ambience without walls.

With the HG even when I hear components of the sound being reflected directly off the wall because I’m too close to it, I just notice it coming from a particular direction apart from the illusionary sphere of ambience, but there’s no sense at all of the wall ever being right there. That reflected sound still conveys much of the same ambient filled micro detail, all in near perfect time, so you’re not “feeling the wall” and instead you still get the illusion of the true ambience faithfully reproduced. That’s my guess anyway.

Also there's very little sense of where the speakers are. I have no doubt of the room hurting focus, and the speakers more, but they're still good enough for the imaging to be getting carved out of thin air and is holographic like crazy.

I have a much greater chance of further improving focus via the electronics right now than I ever will with this room. Better speakers would help greatly, and immediately, but they're still good enough to not be of great concern. In fact I never liked them as much until I got the HG's. It seems like I've had them pushing fifteen years now and they haven't impressed me until last month. They've never sounded anything like this, and they've been in all kinds of rooms.

Lesser electronics would never do this at all. Imagine some bargain bin electronics like a sonic impact driven by an Ipod with the very best speakers in a “perfect” room, you could very accurately tell what junk your electronics were, and you wouldn't be bragging to your friends about the room or the speakers given that sound quality.

I'm thinking with your list being in such an order of importance, along with the statement that the mods to your amp didn't produce much compared to room treatment, could be indicative of a problem with things somewhere. Could be your 16 bit NOS DAC which I know nothing of, could be your implementation of the modules which I know nothing of, could veeeeeeeery likely be your audiodock. It could be all three.

Feel free to email me if you want more info about getting your hands dirty with the audiodock. The worst that happens is you kill it, and knowing it like I do, I don't think that's the worst that could happen.

Cheers

Chris, thanks for the comprehensive reply.

Did you do a before/after assessment when adding the CRD to the LM4562?

What was the x7r mod? Is this the resistor in the power supply lines to the opamp that helps decouple the power supply with a RC filter?

I'm not sure I want to invest more time/$$ in the 1820m, it does fine for my surrounds, but has an annoying problem with switching off when warm, especially after a reboot, probably a bad joint but I can't find it. If I get this problem I put the audiodock in the freezer for 5 minutes and we are up and going again. I'd like to try the Lynx 2B but I can't find it at a reasonable price and I'd have to do something different with the digital filters as I've grown quite attached to patchmix. I am comfortable modding the audiodock, it is a good platform for modding (eg. can add auxillary supply) and is something I might do over the Xmas break if I don't get the Lynx.

I agree Red Hot Chilli Peppers has awful mastering, mastered for FM radio and sounds flat on a decent system. Do you listen to all your music with MP3, even with higher bitrate encodes? Lossless would be a better option especially as a medium for archiving & transcoding. I can tell the difference on my system between lossless & 320Kb MP3, so you may have a bottleneck there, and with all the effort on the DAC/amp/speaker/room, its best to use as pristine a source as possible - heck disk space is cheap now. I have 400 odd CDs on a 200Gb disk with lossless storage, well worth it.

Last night I was listening to the new Santana HD-DVD which has the new DTS-HD codec (2Mbps bit rate and efficient encoding) and it was amazing, the first time I have heard a concert DVD (HD-DVD actually) that could rival a well mastered stereo CD, so I do listen to lossy sources

I don't think the problem with my setup is with the soundstage, I have listened to a number of digital amps and have noted that the two characteristics of them is better bass control (less sloppy) and better soundstage. I suspect that I like stronger bass and as the Hypex amp and Dunlavy speakers are very accurate, which sounds thin to me, and the subwoofer boosts the bass which I can confirm with room measurements, but I like it like that. The sub moves so much air it also shakes the chair which adds bass realism and weight, because you can feel each hit on a drum, an effect that I only ever felt when previously listening to live music, and of course is a great effect for home theater.

In another thread in this forum there is talk about weak midrange for class D and I don't hear that. I do agree with you that if the room & speakers are right and you have good sources, then the deficiencies in the DAC and amplifier become more important.

Regards,
Dean

"Did you do a before/after assessment when adding the CRD to the LM4562?"
This is pretty hard science and not a subjective thing at all. No need for in depth listening sessions (which I'd done in the past just to prove it to myself anyway), it's one of those plug and go tweaks, like I was saying in the first post, a no brainer.

"What was the x7r mod? Is this the resistor in the power supply lines to the opamp that helps decouple the power supply with a RC filter?"
No, something unique to the older 180 modules I had at the time. What you're talking about though is something else I looked after when modifying my 400AD modules.

"I'm not sure I want to invest more time/$$ in the 1820m, it does fine for my surrounds, but has an annoying problem with switching off when warm, especially after a reboot, probably a bad joint but I can't find it. If I get this problem I put the audiodock in the freezer for 5 minutes and we are up and going again."

That's not at all good for condensation. Yeah I really wouldnt' recommend investing a cent or a second into the audiodock if you can replace it with something better. Their newer card, 1616M, might not suck so very bad, but I couldn't say from experience. I declined a big discount on one from them after I told them about my issues with this thing... still paying off this junk, dont' want another.

"I am comfortable modding the audiodock, it is a good platform for modding (eg. can add auxillary supply) and is something I might do over the Xmas break if I don't get the Lynx."

Not really... not really. Not if you can't help what's broken by design very easily.

"I agree Red Hot Chilli Peppers has awful mastering, mastered for FM radio and sounds flat"

Flat? OK, if flat is the point where they hard clipped it. My ears will ring listening to that at even a low volume.

"Do you listen to all your music with MP3, even with higher bitrate encodes? Lossless would be a better option especially as a medium for archiving & transcoding. I can tell the difference on my system between lossless & 320Kb MP3, so you may have a bottleneck there, and with all the effort on the DAC/amp/speaker/room, its best to use as pristine a source as possible - heck disk space is cheap now. I have 400 odd CDs on a 200Gb disk with lossless storage, well worth it."

No not all, I would say the 99.8%, yeah. The ~16 songs I like that are mastered good enough to make it worth my while I'll have in lossless, they are very, very few. I can easily tell the difference between MP3 and lossless as well, it's just not worth the effort to me, for a more pristine copy of junk is still junk. At this high a level of sound reproduction, most would be throwing their CD's in the trash in the first place, not archiving them in lossless format.

When typical mastering quality of what I like to listen to is pristine, I will typically acquire the pristine copy. For the very rare higher quality masterings, much of said higher quality is still conveyed by the MP3, even though lossless will convey more info still. What MP3 misses though, isn't greatly missed, certainly not enough for me to want to apply lossless format to everything just for those very few rare occasions of a very good mastering. So you see? It's not that I don't miss out on any sound using higher quality Mp3, I just dont' miss what I'm missing much, and with the usual garbage quality of masterings, it's no loss at all

"Last night I was listening to the new Santana HD-DVD which has the new DTS-HD codec (2Mbps bit rate and efficient encoding) and it was amazing, the first time I have heard a concert DVD (HD-DVD actually) that could rival a well mastered stereo CD, so I do listen to lossy sources"

Doesn't sound like you missed that loss much then


"I don't think the problem with my setup is with the soundstage, I have listened to a number of digital amps
and have noted that the two characteristics of them is better bass control (less sloppy) and better soundstage."

Could be all you heard there was a switching power supply, which are characteristics I expect their regulation would improve on, even though the 'net consensus on that seems to be they have worse bass control, or just power.... makes no sense but that's the rumor. You can't really attribute those characteristics universally to the topology itself though without implying some for of distortion unique to the classification as a whole. However I find with subpar electrolytics in the power stage the level of microphony in them agitated by the high frequency switching ripple messes with the microdynamics enough to create a kind of distrubed ambience, one that's different from the source. It can be more minor or severe enough to start dropping macro detail. My nutty theory.

"I suspect that I like stronger bass and as the Hypex amp and Dunlavy speakers are very accurate, which sounds thin to me, and the subwoofer boosts the bass which I can confirm with room measurements, but I like it like that. The sub moves so much air it also shakes the chair which adds bass realism and weight, because you can feel each hit on a drum, an effect that I only ever felt when previously listening to live music, and of course is a great effect for home theater."

Shakes the chair?? I can shake the cement floor a good 12 M away, and it's right in step with the tunes. There should be nothing but good old canadian shield under that. Feels like a minor earth quake, I'm not sure if it adds to realism or not. I like it. The problem starts when you have that happening at low volumes though, it's not realistic. More like car audio, where fronts are powered by say "25" watts and the subs by few 1000 watts.

As far as feeling every beat HG delivers it all and in perfect time with what you hear. Hell I feel each string too, like a kick in the chest, but its' from the midrange now. Felt your jaw rattle yet? That's fun too.

"In another thread in this forum there is talk about weak midrange for class D and I don't hear that. I do agree with you that if the room & speakers are right and you have good sources, then the deficiencies in the DAC and amplifier become more important."

mmmm I think they're more important to start with. As they improve further, the deficiencies in room and speakers will take over and tip the bottleneck scale into their favor, but I think, even though it's all one system in the end, that the electronics have to be very good to get to that point.

My point really is, most people don't have a choice what they can do with their room besides where to hang the odd picture, or throw a rug, and things are so non ideal that I'd have to be dreaming if I heard any difference doing that here. There's a lot I can do with the amps and what not though that will improve things more, and more, and more aaaaaaaaaand more, without the room or speakers ever changing.

How much of a bottleneck is that then? It hurts focus, and perhaps the truest illusion of ambience, etc... does not ruin enjoyment. For example, let's see if I dont' get more focus with further improved regulation (auxiliary supplies implemented) and going full monoblock for a very first, in this very same room with these very same speakers. Do you think the room and speakers will hold those little tweaks back from being very audible? Sure they'll hold it back a little bit, but they're going to blow the walls down that much more anyway. Just like adding the CRD's did, I sure heard that.

BTW, my favorite "room treatment"... closing the windows. Unless it's -30C outside, where it would then be opening the windows. Both are kind of hard not to notice a difference.. and that's all the treatment this particular room will ever get.

Cheers

Chris,

Regarding room effects, I have a similar set of speakers (Dunlavy Aletha) for my living room, where I definitely have less options for room treatment, and use a HTPC with a Panasonic XR45, sure not the same calibre as the Hypex setup, but not bad either, and there is a big difference in sound quality compared to my media room. The room and speakers IMO is by far the biggest tweak you have, so I will have to differ with you in this respect, but it is based on my experiences.

Don't get me wrong about the Hypex modules - I think my setup for my media room is great, I have not heard a better system and I have auditioned a few over the years. In my Hypex amp I use BHC 4 pole caps on the power supply, with split rails, and dual mono setup. Its pretty decent. Although having a high quality aux supply for the Hypex preamp opamp would help.

Shake the cement floor 12M away, now that must be Canadian humour, or you have subwoofers coupled to the floor. Isn't the ground in some parts of Canada permanently frozen - maybe that creates strange acoustic effects for a concrete floor

You close the windows only when it gets -30C outside? Wow, I can't imagine that, its going into summer here and it is 35C outside at the moment down here.....

Hi,

Regarding:
"So, you really find that class d has “a wider soundstage”? If you do, isn’t that a little suspicious? It’s not as though it’s remastering what’s on the source right? So what’s the problem then? Maybe it’s the microphonic noise created via the HF ripple in inferior electrolytics that’s messing with the micro detail? I hope you weren’t too in love with it, because this is the type of noise the Hypex caps don’t have! Personally I don’t miss it, it’s been replaced with true, clean, ambient filled micro detail, and I find it to be an acceptable trade."

Written by someone who's obviously more involved with the production aspect than we who concentrate on the reproduction:

"You can use noise as the modulator in FM patches to produce eerie, atmospheric sounds such as submarine sonar pings and vocal choir impressions. Because noise is a complex modulator, it is best used to modulate a sine wave or some other low-harmonic-content waveform. Noise modulation adds an evocative dimension to the sound, giving it a mysterious reverb quality (see Web Clips 5 and 6)."

You can see how he's enthused with this but I tend to doubt he would apply it to everything he hears universally, or it would quickly lose the appeal. For instance with an amplifier that's guilty of the same type of thing, the only thing it can evoke is you eventually getting up to walk away.

Here's clip 6 where I find it's an obvious enough difference that even those with not so clean systems might pick up on it:

http://emusician.com/web_clips/SDW%20Dec05%20WC06.mp3

Things to note here are how pure and rich the tone is before the noise FM kicks in, where it then suddenly becomes etched and thin, not at all pure, and is no longer anywhere as near relaxing on the ears (imagine that effect continuously while trying to listen to orchestral music). Also, the sudden appearance of an odd reverb like echo... true ambience as it was recorded has been grossely altered if not lost entirely, and a very bad cap will not only do exactly that but even start dropping some of the macro detail. The selection of the cap alone will effect the degree to which this problem exists, or to which it exists as a problem, where most caps seem to exhibit this to a large enough extent to be audible and therefor problematic if you at all care for true reproduction and not some euphonic "painted all with the same brush mess.

It is this exact kind of difference you hear with lesser caps. No other cap I have tried in their place can produce the rich tonal purity as the Hypex cap, where all others sound thin and etched and also include odd reverb like effects all exactly as experienced in the second half of that clip #6.

Once the caps are this good, then things like the room and speakers will play a greater roll in what the true bottleneck to enjoyment happens to be, and before that, they hardly matter much at all.

Also, I find that the odds of the amateur tweaker finding a cap as good are astronomical.

I have actually compared two caps so similar that they're identical on paper, however their physical contruction differed enough that the difference in sound was again very much like that of clip #6. I think this comparison is good enough to display that it is in fact the mechanical properties that dictate how much of a problem it is going to be.

One would think such a phenomena could be measured, though not very easily, and would probably be no where near as effectively observed as it is just by using the old ears.

Cheers

Chris

No. It is simply a matter of frequency response.

Jocko

That depends on what exactly you're talking about, can you elaborate at all?

Keep in mind this discussion relates to an amplifier for which the frequency response compares to the very best, where it is virtually ideal up to a respectful frequency, and even has a very respectable impulse response.

I can see the soundstage lacking height, and therefore appearing wider, when all the high's get chopped off including a lot of the microdetail with it, this too however is in relation to the selection of caps.

Swapping those alone should have little effect on the frequency response of the amp.

The sense of dimension derived of the signal leaving this amp in particular would be more readily effected by a simple change of caps than for say minor tweaks to the bandwidth.

What "frequency response" doesn't cover as a simple matter is the sonic correlation of say.. how two "identical" caps can sound so drastically different in the very same circuit/system.

1. How come it suddenly lost so much focus and tonal purity.
2. How come one is soo thin and so heavily etched, where did those detailed, rich harmonies go?
3. Where in the hell did that crazy reverb come from, and how this "helps" me locate the walls around me.
4. Where did the recorded ambience go, why is what's left feel glued to the walls.
5. What the hell is that noise.

6. How these things come and go to varying degree's (more reverb now?) just by swapping caps, yet with the one decent cap all vanish completely below audible levels, where the sonic change is at least as drastic as that in the clip I gave you if not more.


The sound clip offered demonstrates one degree of many of these very same artifacts and it was created in a very specific way that lends itself to explaining these very artifacts in class d amplifiers..... that are of high enough quality to portray them in the first place. EMI inferno's or subwoofer amps won't be them.

PS:

I have found you can mimic certain changes in sound you'll get from a swap in caps with the use of an equalizer, so that's certainly frequency response related.

It's not those changes I was discussing though, but others that fall to the physical characteristics of caps that don't really have anything to do with frequency response, like the level and character of the granularity and reverb like effects.

These too are physically related and known cap effects, but I think in class d there are additional problems. The reverb like effect in particular is a factor to caps already and even some super audiphile caps exhibit it to a gross extent (real garbage here are silmic II like Fred always said), but what you get in the class d amp isn't that same reverb effect, which is very clean, smooth and unbroken, it's more like you hear from the recording, where it's possibly a ghost image brought on by the noise envelope modulating the signal, and I believe it to be (re)created in a more similar fashion as seen in class d amps as well.

EDIT: which I should have mentioned was created by:
"This clip demonstrates the effect of noise FM on a simple sine wave patch. The unaffected patch is heard first, followed by the noise frequency modulated version "

http://emusician.com/web_clips/web_clips_december_2005/

Check out clip 5 too.

Cheers

Chris, I think Jocko mentioned frequency response because class D amps are very speaker dependent and frequency response changes according to the impedance of the speakers it is driving.
Although I think the Hypex amps have some way to deal with this and minimize the problem, isn't that right?

Btw you can never compare the effect of DC coupling or better PSU caps with what you can achieve with an equalizer.
With EQ you can eventually get to the same... huh... "tonality" but phase shifts and a bunch of more active and passive components in the signal path just kill the harmonics in the music.
Music is not just bass, mids and treble. There's much more than that and you know it.
PS: oh, you call it reverb - that's fine.

Here's another important one: treble precision.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein