Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Low THD PWM audio amp, DIY friendly (diyclassd.com)
24 points by catseyechandra on Jan 14, 2023 | hide | past | favorite | 48 comments


DIY as in assemble yourself? I tried finding parts or info or schematics or designs for their other amps, but nothing of any real detail seems available. Cool potential offering here, but based on their other offerings, looks like it will probably be phenomenally expensive even in kit form. And not at all in the kind of hacker-ish open form we might hope.

It'd be great to see some real open-source wonkery start eith some of the modern chipamp chips. TI's latest tpa3255 is a beastly chip, and there's some pretty good sub $100 ~400W designs about based on it's advanced magic. Trying to get a bit more geeky, to share & tune & profile boards & builds together for modern chip-amps could really open up some new frontiers of affordable quality.


My money is on "tax and/or tariff avoidance ploy."

Taxes and tariffs are responsible for "tariff engineering" to get around the definitions used for many tariffs. For example, a lot of early DSLRs couldn't record video continuously for more than 20 minutes, because this would qualify them as camcorders.

Either that or they're selling this for people who have the skills to create a chassis (or have one made) and then put these guts into. Sort of a "crate engine" of the audio world, I guess.

I liked the "affordable" claims. The complete kit is $1200...


DIY in the site name is very misleading. There is nothing DIY there, they sell already assembled kits, no schematics or component list. From the site it looks like they work closely with NCORE which produces very good amps.

Edit: clarification


There are quite a few very, very good class D amplifier designs out there. “Very good” means highly efficient and with good enough fidelity (noise, distortion, etc) that no one will ever be able to distinguish them from perfect.

But, IMO quite annoyingly, these things generally have analog inputs. This means you need a good DAC (which exists and isn’t terribly expensive) and a good connection from the DAC to the amp. You can make a good connection out of inexpensive cables, but you still need to worry about ground loops and possibly signal levels.

So why aren’t there comparably good amplifiers with digital inputs and a DAC integrated directly? Fancy preamp features are unnecessary — if the input resolution is 24-bit, then volume control and such can be done digitally with effectively no loss.


Given that it's just phase-correct PWM, you'd think feeding the bits you'd stuff into a DAC directly into the PWM counter would do the job nicely.

You could probably do that with a fast microcontroller these days. Some golden-eared audiophool pillock would complain they could hear jitter caused by the pins not being gold-plated or something, but who cares?


Not at all my expertise, but:

The good class D amplifiers are closed-loop control systems. They don’t just create a output a fixed sequence of highs and lows for a given piece of music — they determine when to switch based on some combination of the input signal, the observed output voltage and current, and various state variables (for example, for protection). And they operate at frequencies of several megahertz, possibly with rather high precision pulse widths needed.

So one could do this with an ADC, a microcontroller, and a single output bit, but it would need to run the ADC at quite high frequency and possibly have multiple input channels (for current and voltage). This sounds expensive.

A TPA3255 costs $12.50 (qty 1 from Digikey). Add a dollar or two a nice DAC. I don’t think a 20-bit ADC, a DSP running at a couple hundred MHz, and most of the same driver and power circuitry that’s already in the TPA3255 would come out at the same price.


It takes a lot of computing power to reconstruct an audio sample well. Once reconstructed it takes massive massive amounts of storage.

It's all about the Shannon Nyquist sampling theorum.

"Strictly speaking, the theorem only applies to a class of mathematical functions having a Fourier transform that is zero outside of a finite region of frequencies. Intuitively we expect that when one reduces a continuous function to a discrete sequence and interpolates back to a continuous function, the fidelity of the result depends on the density (or sample rate) of the original samples. The sampling theorem introduces the concept of a sample rate that is sufficient for perfect fidelity for the class of functions that are band-limited to a given bandwidth, such that no actual information is lost in the sampling process. It expresses the sufficient sample rate in terms of the bandwidth for the class of functions. The theorem also leads to a formula for perfectly reconstructing the original continuous-time function from the samples." [1]

But, HOLD ON: "Practical digital-to-analog converters produce neither scaled and delayed sinc functions, nor ideal Dirac pulses. Instead they produce a piecewise-constant sequence of scaled and delayed rectangular pulses (the zero-order hold), usually followed by a lowpass filter (called an "anti-imaging filter") to remove spurious high-frequency replicas (images) of the original baseband signal." [1]

In case that was too verbose.....

"sample rate that is sufficient for perfect fidelity".... "such that no actual information is lost in the sampling process"... "theorem also leads to a formula for perfectly reconstructing the original continuous-time function from the samples"

PERFECTLY!!! Zero information loss between sampling and reconstruction. Still WTFs me to this day.

So yeah, you can feed bits to a PWM. But that not music.

[1] https://en.m.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_samp...


The thing is, if you design your input and output filter correctly then any sampled signal will be indistinguishable from the analogue source passed through the same set of lowpass filters.

It does not take a lot of computing power, but it does take a couple of opamps.


I think you misunderstand sampling theory. Sampling theory allows the sampling and recreation of ALL information, not just the 44100 samples, but datapoints in between. Sorry about the potential bad formatting, it's from a pdf.

"By contrast, the waveform reconstruction process – digital-to-analogue conversion – is not so simple. In an ideal world it would be achieved by generating an impulse of appropriate amplitude for each sampling point and passing the train of regularly spaced impulses through an ideal low-pass filter with its passband upper edge set to half the sampling rate. The impulse response of such a filter is the sinc function, so each impulse would generate a sinc waveform of the necessary amplitude, and the train of sinc functions would sum to recreate the original waveform.

This is illustrated in Fig 2, which shows six successive sinc functions of different amplitude and their sum (the black trace). Each sinc function contributes nothing to the summed signal amplitude at other sampling points, but does contribute to the waveform between the sampling points. It’s a common misunderstanding of sampling to suppose that the waveform between sampling points is unknowable but that is not true – provided that the input signal is bandlimited and sampled at least twice as fast as its highest component frequency, as required by the Shannon sampling process. In that case the waveform between sampling points can be reconstructed unambiguously. Importantly, the wave shape depends not just on the value of nearby samples but, ultimately, on the pattern of samples throughout the sampled signal" [1]

[1] https://drive.google.com/open?id=1QhJo_Rh8sQsItFvJ1g-MnNlg9p...

Or click the link here: https://chordelectronics.co.uk/explained-rob-watts-filter-te...


Not affiliated, just a happy customer, but anyway: I had good luck building around a DIY low noise class-D design (modulus 86) made by a fellow from Canada, branding himself as neurochrome audio. It was the first amp I could find that had a low enough noise floor to keep my high efficiency horns from hissing audibly. Mine is a multi-amp setup with active crossover, so I just use these dual monos for the tweeter channels, but they're great.

edit: My mistake. When I first read your post I thought you said there were quite few as opposed to quite a few good designs, hence the recommendation.


It sounds like what you want is a dedicated class D power amp section and fully digitized dac/preamp section.

I think you’ll have more success focusing on them as separate components as the power amp will always be in the analog domain


This is exactly what I’m objecting to.

A good amplifier has better than 100 dB SNR. Your ear may have better than 100 dB dynamic range. Your sound system may have a bit less useful range, since your goal isn’t actual to destroy your ears. But 100 dB is a vaguely credible target.

So somehow you’re supposed to get a preamp and an amp, and you’re supposed to connect them, usually with RCA (unbalanced) cables. You’re supposed to get both devices to agree as to what voltage is full scale. And you’re supposed to get a noise floor, on the interconnect, better than -100dBFS. And you are supposed to do this with at least two separate cables plus whatever ground (or neutral) connections exist to both chassis, and this is all done near 60Hz currents that induce a voltage around every loop in question.

This is a tall order. At least the pros use balanced connections.

Or the industry could move to digital interconnects, where the DAC and the amp are connected by short traces on a PCB with a proper ground plane. And full scale on the DAC could be matched to the amp’s input.


Searching for "plate amplifier module" you can find integrated digital/amp/crossover units suitable for DIY builds.


That’s a great form factor for building a speaker but a rather annoying form factor for driving an existing passive speaker.


From the title, I expected an audio amp project with schematics and layouts available that was "DIY friendly" to build yourself. I was left disappointed…


Don't see the point of this when I can just build a Kevin Gilmore Dynalo or Dynahi. Very low part count, almost BOM golf like; multiple variations by him and others to cover any case. Closest you will ever get to a wire with gain.

Not to slight Hypex (the company behind the Ncore family of Class D chip amps), mind you, they're probably the only chip amp to ever succeed in making low distortion high output high efficiency chip amps (and if it is a class D and doesn't have a Ncore, it probably sucks, unless it's that new THX design family); but it will never outperform a Dynalo/Dynahi built correctly.


aren't the Dnyaxx amps headphone amps?


Yes and no. People have built speaker amps out of Dynahis. Due to the small size of the PCBs, it is trivial to do bi/tri-amped designs (with the woofer using a different amp or multiple stacked Dynahis). The Dynahi is, admittedly, a ~25w amp (at least, at near-zero distortion and clipping conditions) when firing into an 8 ohm typical load, which means it isn't a headphone amp either; the Dynamite configuration (balanced Dynahi) doubles this output.

Note: the best tweeters ever made have rather strange electrical characteristics, in the same vein that some headphones do. The Dynaxx family tends to handle anything, high or low ohm (even speakers that are 4ohm nominal or headphones that are 1000+ohm), but also ones with ridiculous impedance curves. People have realized that due to its exceptional tolerance of weirdly designed speakers, it makes for a good speaker amp too. 25w is enough to drive these tweeters loud enough, and some mids as well; woofers, no, not with a single board.

Kevin Gilmore also has sequels to his Dynaxx family (these being, themselves, improved versions of the KG Dynamic), due to some parts being replaced as they've fallen out of production and are hard to buy now.


Too bad it's not actually DIY.

Does anyone remember the 80s-era catalog of DIY HiFi kits? I can't remember the name, it was electronics geek+audiophile porn. Maybe it was Craven? something like that?



They seem too focused on automotive/car stereos to be it.

The catalog I'm recalling was cover to cover populated amplifier circuit boards accompanied by specs and prices. Lots of huge toroidal transformers... maybe it was a local thing back in IL.

The cover was mostly black with glowy green text IIRC, nothing like the vintage Crutchfield catalogs I can find via GIS/ebay. Oh well.


Possibly Madisound. Their site says they've been around since '72. I have vague recollections of them having a print catalog in the 90s before everything went online.


It might have been an 80s refresh hifi-leaning variant of eico's catalog:

https://www.ebay.com/itm/255457806571

That cover seems very familiar to me, and I do remember there being a bunch of green on a black background.


Google or Youtube GR Research. They supply bookshelf speaker kits and large open baffle towers.


I’m a huge fan of class D amps, although usually on the cheap end of the spectrum.

Recently I purchased a “fosi audio” (brand names seem to rotate) stereo amp with bluetooth. It’s about the size of a raspberry pi.

It is plenty powerful to drive a pair of bookshelf speakers, sounds good, and connects to a turntable and bluetooth source at the same time.

$25 on ebay. Really hard to beat for the price and size


I can reliably distinguish even very high-end Class D from a decent class AB amp. I perceive sand-grain in the sound of class D.


I'm both an IT and audio geek, does HN likes audio geekery? I'm not affiliated with the company behind this website BTW..


> does HN likes audio geekery

Only if scientifically backed.


Hahaha! I can't imagine discussing the finer points of optimization on here.

"the electromagnetic signal of course rides half outside the wire in a world increasingly filled with RF/EMF noise. Be sure to keep your cables off the floor or there WILL be impedence changes. Also, the Frequency Response is largely not relevant as 95+% of audiophiles don't have effective bass traps or ways to remove excess energy from the room. Intelligibility is the better metric, and no one uses it because it requires bass trapping and that's just to ugly and expensive."

Are you ready to rip my head off yet for all these largely unknown but completely true scientific facts? :)


Of course science would also require a blind perception test ...


When talking about high-end audio it is more of The Science™ than about the scientific method. The former is also far more profitable so it is not surprising to find so little of the latter in this field.


Audiophool class-D amplifier. Now I've seen EVERYTHING :P


I assume you meant "audiophile", not audiophool, right? If so, I wonder what makes an amplifier to be an audiophile amplifier in your opinion?

Objective and measurable parameters of the amplifier in question are phenomenal.


I assumed it was intentional, as in “fool”.

It’s a shame because Class-D has come a long way, there seems to be a natural opposition to anything digital.

Class-A is great but is expensive, heavy and a big electricity bill.


> It’s a shame because Class-D has come a long way, there seems to be a natural opposition to anything digital.

The funny thing being that Class D has nothing to do with digital. A Class D amplifier is say to a class A one what a switching power supply is to a linear one, although the amplifier works on current rather than voltage, but the principle is essentially the same: they're 100% analog, and sound wonderful.


It's also free in the winter if used in place of a space heater


There is a picture of special, braided IEC power cable on there.


Geometry. Rigidity. It's POTENTIALLY legit. Here is a pretty science based primer on some of the things going on in the cable world.

https://www.gcaudio.com/tips-tricks/why-power-cords-are-impo...

One thing to understand about this niche high end cable market is that intellectual property just isn't a thing. There isn't enough money to fight lawsuits and wage war against copycats. Everyone loses throwing money away.

The end result is that only large companies like AudioQuest can afford to "give away" the science behind their products by patenting them. Other companies have to talk around and obscure what ACTUALLY makes their cable sound good. This is a pretty raw deal for consumers who are left exposed and vulnerable to bad designers, fraudsters, and tricking themselves into thinking something sounds better when it's not. And that's not even getting into value judgements about what a cable is worth based on how much closer to perfection it gets the user!

The system does seem to be working as intended by small niche cable makers who spend lifetimes accumulating arcane knowledge and handbuilding quality cables while keeping the bar for entry to the market high.

Another neat tidbit about cables is that the product lines and price points are not linear. They just can't be in reality from a design perspective. The people that say a cable is a cable are half correct. It's impossible to have a cable lineup of 7 levels and improve on the induction, capacitance and resistance all at the same time in each level. So there are tradeoffs, and occasionally a cheaper cable from the same company will sound better than a more expensive cable because one or two of the parameters were improved while another got worse. Depending on the interactions between the connected equipment and their deviation from perfect, if the parameter that took a step backwards in a more premium cable happens to be the weak point of the equipment as well, then an objectively better and more expensive cable can sound the same or worse.


_power cable_, the one you plug into wall socket - thats how you recognize audiophool scams. You maaaybe could claim some speaker cables are better in theory, but then no one cashed out Randi foundation 1Mil for coat hangar speaker cable challenge

https://gizmodo.com/updated-journalist-accepts-1-million-cha...

spoilers they didnt get the money, they didnt even participate :)


The problem with this is that it's way too easy to to tell the difference between power cables. Even non audiophiles can hear the difference. This is also 15 years ago. Manufacturing has taken such a leap between then and now that even if it was in doubt then, it's certainly not now. Now it's simply a matter of taste, and the better way to represent the reality of cables is that it's subjective, they do sound different, and even though everyone has their favorite cables, no one can agree on what sounds good, much less what makes them sound good. But in general, an better cable will sound "more agreeable" to more people.

I also agree there are a lot of garbage overpriced cables.

I think it's important to remember cables are not measured under appropriate loads, and they aren't measured connected to real equipment.

The measuring equipment falls far far short in the Audio world. Take for example the Audio Precision APx555B. It's the best audio analyzer in existence, and if you go to audiosciencereview, there are now DACs that have measurements that are below the measurable threshold for this machine in certain regards.

Power cables are the same. Not measurable in a meaningful way.


You wrote a lot of nonsense. Randi was willing to pay $1mil to Anybody able to tell the difference between normal speaker cable from Circuit City vs audiophool product. Nobody EVER took the challenge.

Saying Power cables are also better, but just not measurable puts you square with the rest of magical scam audio crowd.


I spent 10 minutes looking for how the challenge would be set up. I couldn't find the requirements. Seems like publicity stunt that is uncashable. Have you seen the terms and conditions?

Would be very interested in them. Seems like Randi is straying out of his lane on this one.


I understand the sentiment, but tech has improved a lot. Check out the NAD M22. I own one. It's not to be laughed at under any circumstances or in any comparison to other amplifiers.


Wow, that's quite a lot of capacitors. Why so many?


500W is a lot of juice. Consider that 120 times per second, the amp gets no power from the wall and has to rely solely on the power in capacitors. The capacitors have to supply the full 500W. Possibly more—something that’s nominally 500W “peak” but driving a speaker may see instantaneous loads much higher.

Modern circuitry has increased efficiency and reduced the heat dissipated by an amp, but the capacitors still need to supply the full amount of power between mains cycles.

It is sometimes easier to get the desired performance out of paralleled banks of capacitors rather than just use one large capacitor. This does look a bit excessive, though.


Perhaps it's a measure against failure? Do these capacitors fail open?

Or price, although that would raise new questions. Another is availability of components. Or perhaps the form factor (these capacitors seem slightly less tall than the ones on the other board in the picture).


Designing for failure seems unlikely here. I think they just wanted lots of capacitance and a low ESR. Lots of capacitors is a cheap way to get that.


I'd guess it's because they have a high quality switching power supply, a power factor corrector, and a digital amplifier. All those things require good capacitors -- probably more than if all the circuitry was linear.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: