Just like the title says I haven't played with any class D amps for a long time. When I was listening to them they were great for the money and small size but still had a ways to go in the sound quality department. I'm curious if some of the DIY amps and premade boards have achieved the level of, lets say an F5 Pass Labs amplifier? Personally, I think that class D will begin to start taking over the market place once it has achieved that level or better.
I'm not sure what you mean by "the level of the F5".
Class D need not meet the specifications of an F5 to take over the marketplace IMHO. Those not wishing to spend $3K on an amp can a make a class D amp that sounds just as good to most ears but with much higher power, much lower output impedance, and similar levels of distortion at the power levels of the F5 for 1/5 the price of the F5. I haven't heard an F5, but I've heard a lot of amps, and the Hypex UcD modules sound great to my ears. If you can hear a difference between those and a $3000 F5 and think the F5 sounds better and have no qualms about spending $3000 then go for the F5.
But most people would not hear a difference in double blind A/B tests and most people don't have $3k to spend on an amplifier. Nobody can hear out to 100kHz, and speaker distortion dwarfs the 0.002% spec of the F5. The Hypex is not much higher at 0.01% at nominal power levels. Once the specs get good they become meaningless. Sure, if an amp has 10% THD it will sound bad. But if it's less than 1% or so there's no reason the amp can't sound great. A THD of 0.002% simply means the amp might sound good. Same thing with most other specs. Very poor specs almost always mean poor sound, but great specs don't necessarily equate to good sound. No doubt the F5 sounds great. But how "perfect" does an amp have to be to *sound* perfect? That's an individual opinion in most cases, but SURELY there is a point where an amplifier becomes so good that it can't be surpassed in the sense that the limitations of the human auditory system are physically incapable of detecting the difference between it and an even "better" amp. The F5 may sound like the perfect amp, but so do my cheap (relatively) Hypex modules, at least to my ears.
Without a double blind A/B comparison there's no real way to know if one amp sounds better than another once they become sonically close. It's been proven that if people *think* they are listening to or tasting or looking at something that is "better" they will choose that item when asked which they prefer even when the items are IDENTICAL. Your subconscious influences your choice. The ONLY way to eliminate this effect are double blind listening tests.
There are a lot of class D amps out there now, and unless you've heard all of them and compared them side by side in the same room with the same speakers you can't draw any conclusions about the superiority of a Pass Labs amp or any other amp for that matter over something you haven't heard. It's doubtful that very many people have the money, time, or inclination to double blind test every class D amp in comparison to respected reference amplifiers, and without that experience all you will get are subjective opinions like mine.
Last edited:
Pretty close
Spend $120 - about what decent heatsinks for an F5 cost if bought new - for something like a hifimediy T1 plus the appropriate Meanwell plus shipping plus a couple decent coupling caps. Then listen for yourself.
Sounds pretty darn good, and is a lot cheaper than an F5, and doesn't sit there burning several hundred Class A watts.
Hard to compare, though, very different approaches, hard to say which is "better".
On the Pass front, I'm waiting for Nelson to release the schematic for the J2 so I can play with the SemiSouth power JFETs.
Spend $120 - about what decent heatsinks for an F5 cost if bought new - for something like a hifimediy T1 plus the appropriate Meanwell plus shipping plus a couple decent coupling caps. Then listen for yourself.
Sounds pretty darn good, and is a lot cheaper than an F5, and doesn't sit there burning several hundred Class A watts.
Hard to compare, though, very different approaches, hard to say which is "better".
On the Pass front, I'm waiting for Nelson to release the schematic for the J2 so I can play with the SemiSouth power JFETs.
At this stage I would recommend buying a ready built/tested module.
Class D can be tricky to design and layout on a pcb for the uninititated.
I have 35 years of experience in electronics but got caught out when designing a irs2092 based amplifier. Even with a datasheet circuit to copy my first attempt didnt run. Problems with decoupling, inductors and layout were all problems.
Class D can be tricky to design and layout on a pcb for the uninititated.
I have 35 years of experience in electronics but got caught out when designing a irs2092 based amplifier. Even with a datasheet circuit to copy my first attempt didnt run. Problems with decoupling, inductors and layout were all problems.
Forget about the price aspect and THD. A good sounding amp is a good sounding amp. the best amp I have owned so far is the Decware Zen Select SET. I picked the F5 because it is a DIY amp that is relatively reasonably priced to build and has become somewhat of a standard. When class D first became popular there were a lot of terrible sounding class D amps around. I chalked this up to the fact that is was a relatively new amp design and had not been refined for as many years as tube and SS amps have been.
bdbender those power JFETs sound very interesting. I don't want to know if Class D is better then F5, I want to know if class D has reached a level of design and refinement where the difference would be a matter of personal taste. This is one of those things where describing audio in general becomes one big gray area. Would you say the T1 from hifimediy would be a good example of the best DIY class D amps offered currently?
bdbender those power JFETs sound very interesting. I don't want to know if Class D is better then F5, I want to know if class D has reached a level of design and refinement where the difference would be a matter of personal taste. This is one of those things where describing audio in general becomes one big gray area. Would you say the T1 from hifimediy would be a good example of the best DIY class D amps offered currently?
Last edited:
This is a serious logic fail. You compare the hypex diy units to a finished amplifier. One can build an F5 for around the same money (or less) than even the cheapest hypex unit.
As to the rest of your email, I have compared some very expensive D amps to the F5 (which should I prefer then?) and PREFERRED the F5. To my ears, it wasn't even close.
Like the original poster, I'd be interested in hearing about the latest and greatest!
Well, I don't want to pose as an expert on something I am not. I was collecting parts for an F5, have the cviller boards and the matched transistors and so forth, but stalled at the heatsinks, toroids, and case. The way I was going to do it they would have been $300 or $400 easy (yes I know you can cobble something together for less).
I got distracted by the hifimediy "new TK2050 board" thread, and after reading the whole thing, ordered a T2 plus Meanwell. I got the T2 instead of the T1 because I am running 4 ohm speakers. It sounds so good to me I have never looked back, and listen to it all the time now. The only mod I made was to add some 3.3mf Auricaps as the input caps (yes I know Bruno Putzeys says that's a no-no but I have them shielded with copper tape and they not only improve the sound, they run quite a bit louder indicating lower ESR.)
From reading around since, my understanding is that a number of folks think the Tripath TC2000 controller was the best sounding class D implementation so far. If I have it right, this controller used a spread spectrum sampling method, as opposed to the more common fixed frequency sampling, which may account for its sound.
And I'm having fun with all this, which is the point of DIY, no?
definitely yes , some class D amps are now as accurate as the best A/B amps.
I use Hypex modules with modded output cap ( that 's what Hexateq or Genesis have done ) and i've never heard a better amp
http://www.6moons.com/audioreviews/genesis5/6.jpg
i've replaced the output filter capwith 680nF tin foil KP one
for smaller cap , you can use PPS ( polyphenil-sulfide) one
That's not correct. I compare a completed amp USING Hypex modules to an F5 purchased pre-built which according to the websites I'm looking at is $3k. Yes, it can be built a lot cheaper, but, the F5 also weighs a ton compared to the Hypex, and it puts out a measly 25 watts per channel (assuming we're talking about the same F5). If the class D amp "wasn't even close" to the F5 then you are listening to the wrong class D amps IMHO. There should be at the MOST a subtle difference. If there's a big difference and the sound is "not even close" something is seriously wrong with one of the amps.
You can build a 180 watt/channel Hypex amp for $350 including the chassis. How much less expensive is a DIY F5? Less than $350 for all the parts new including the chassis (purchased not salvaged)??? Add to that the labor involved in constructing the F5 (a Hypex module based amp can be put together in one or two evenings) and the F5 becomes a much more expensive proposition in terms of money and TIME, unless you want to shell out $3k for a pre-built unit. The F5 also has less than 1/7 the power of the Hypex least expensive module (almost an order of magnitude) or about 9dB less volume at full power. The two amps aren't even in the same ballpark in terms of power output. If you want to do a fair comparison find a class A amp that puts out 180 watts RMS and compare THAT to the Hypex UcD180.
Did you conduct your tests double blind or did you know what you were listening to when you conducted your tests? Which "expensive" class D amps did you use for direct double blind A/B comparisons? Did you try the Hypex modules?
To my ears the latest versions of the Hypex modules sound better than any amp I've ever owned, and I've owned class A and A/B amps that cost a couple of thousand dollars that don't put out as much power and don't sound as good as the Hypex which is MUCH less expensive and EXTREMELY easy to build.
You're question presupposes that there are a lot of people doing personal double blind listening tests of various class D modules with various class A or A/B amps. I wish I had the time and money to do just that, but I had to rely on various reviews and specifications, as well as the papers written by Bruno Putzeys and those of other manufacturers to make my purchase decision.
I have read most of the class D manufacturer websites and none of the technical descriptions of the amps they make come close to the level of detail given by Putzeys and the reasons the Hypex modules are designed the way they are. Based on what I have read the Hypex modules are the best. To my ears they sound awesome. If you give those a try and you still believe the F5 sounds better then you can conclude that for YOUR purposes that class D is inferior to class A and the technology has yet to yield sonic performance equal to or better than class A amps like the F5.
I don't need to hear an F5 to know that the Hypex modules are suitable for MY most demanding application which is studio monitoring of recorded tracks and completed mixes with as little coloration as possible, a ruler flat response, and very low distortion (as both measured AND perceived).
To sum up, in my opinion class D has surpassed all other topologies THAT I HAVE COMPARED IT TO. This is coming from a guy who up until I built the Hypex amp was a die hard tube amp fan, constructing amps with the parts alone costing $1500 for a stereo amp that weighs 50 lbs. and construction requiring at least 100+ hours to build and fine tune. I put together my Hypex amp in two evenings.
Time is money, and that's something you fail to take into account when you compare a DIY F5 to a DIY Hypex class D. The construction time for an F5 is much longer and more labor intensive than a DIY class D using modules and prebuilt SMPS. You also fail to take into account that the Hypex is in a totally different league in terms of power output. In short the comparisons are not valid.
Forgetting the F5, the Hypex class D modules sound excellent, and if there is a difference in the sound between those and the best amps made regardless of cost, my bet is that in double blind testing the Hypex would rival some of the best class A and A/B amps in the world. I didn't say surpass, I said rival, meaning any difference in the sound would be subtle and audible only to a good pair of ears.
What s**ks about these forums is that it will never be possible to accurately test the perceptions and conclusions of the people who post here. If ONLY we could get together, put the speakers behind an acoustically transparent curtain, and compare the amps blind, we could reach some firm conclusions, at least a lot firmer than anecdotal "evidence" that one amp sounds better than another from non-scientific and uncontrolled comparisons.
Interesting comments.... Is Bel Canto considered good class-D well it was handed it's **** vs an S500 Threshold and most recently a modded Krell KSA200S.
The biggest issue for me with the class-d stuff is at the freq extremes, bass/highs. The highs are splashy with no air and the bass is pulled and lack detail vs the others ..
As usual type of speakers used ( load) will make a difference in these comparisions ,Does this mean class=d will sound worse than an F5?
Not sure never heard one before.....
The biggest issue for me with the class-d stuff is at the freq extremes, bass/highs. The highs are splashy with no air and the bass is pulled and lack detail vs the others ..
As usual type of speakers used ( load) will make a difference in these comparisions ,Does this mean class=d will sound worse than an F5?
Not sure never heard one before.....
If I'm not mistaking, Bel Canto is using B&O ICEpower ASP modules, which are not the best sounding ICEpower modules to my ears. Yet it receives very good reviews.... Personally I would consider Bel Canto products as "good enough" class D, but far from reference sound quality. Far from Hypex. Then I like what Rowland do with ICEpower modules a lot better, since they use better sounding ICEpower modules than the ASP's.
I agree on your comment on lacking highs and lows with most class-D designs. I have not heard them all, but the ones that I have are somewhat limited here. But that does not apply to the latest stuff from Hypex, I think those problems are completely solved by Hypex.
I agree on your comment on lacking highs and lows with most class-D designs. I have not heard them all, but the ones that I have are somewhat limited here. But that does not apply to the latest stuff from Hypex, I think those problems are completely solved by Hypex.
There's a lot of datasheets on Hypex here:
https://www.hypexshop.com/
Hypex are definitelly among the best of class-D... but there might be other options, like Tripath. I have not heard Tripath stuff so I cant tell for sure...
https://www.hypexshop.com/
Hypex are definitelly among the best of class-D... but there might be other options, like Tripath. I have not heard Tripath stuff so I cant tell for sure...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.