Class D amplifiers and "ringing" (pre-purchase questions)

I am considering the purchase of a Class D integrated amplifier costing about $600. I have been doing some research, but because I have no training in electronics I cannot quite understand the phenomenon of ringing.

Ringing, evidently(?), is inherent in all Class D amplifiers, regardless of quality and/or cost. My questions are:

1. Is ringing audible, or is it simply an electronic measurement event?
2. If audible, what does it sound like? A "whine" behind the music, like a distant siren? "Ringing" like the sound decay of a bell being struck? Is it continuous or episodic?
3. Is ringing associated with the volume at which the the amplifier is playing?
4. Is ringing a viable reason to avoid Class D amplifiers?

Thanks!

 

Can't say I've ever heard of this problem with a good class-D amp. I have a Jeff Rowland Design Group Continuum S2 integrated and all it sounds to my ears is musical.

 

have never heard ringing in any class D amp. my main gripes include weaker bass at low volumes and a slightly clinical sound. however they can be very smooth and transparent sounding with powerful dynamics. system matching is important.

 

No ringing I've ever heard in any class D amp. Put 12 engineers in a room and you'll likely get 12 totally different versions of the 'facts'.

 

Do you have a link to an article that uses the term 'ringing' in relation to Class-D?

 

OK, thanks, guys. Sounds like my concerns may have been unwarranted. @avanti1960, the integrated Class D amp I am planning to purchase has a built-in bass boost (can be turned off), so maybe it is to help lower frequencies as low listening levels.

FWIW, this is from an article discussing Class D "ringing" quoted here:

"Even with the best filter components, there are several more ways to burn the pudding...The voltage on the power stage should stabilize immediately after each switching edge. Ringing corresponds to an enormous boost in emissions that even the best output filter can't block. When left unchecked, the parasitic circuit elements present in all power components will produce up to 50% overshoot and ringing that can last up to a microsecond. As luck will have it, this ringing is usually smack in the middle of the FM band. Getting a nice, clean ringing-free square wave requires detailed attention to gate control, circuit layout and damping. Too many designers think that a high-speed H-bridge driver and four MOSFETs is all one needs to build an amplifier. It's all you need to build a radio transmitter, so much is certain. If you see a gate driver capable of running a 10kW UPS, chuckle."​

And also, PS Audio has a well-reviewed product, the Stellar Gain Cell DAC. It is sort of a DAC-preamplifier designed to be used with their Stellar series of Class D amps. In the owner's manual, instructions are provided on how to use the filter option to optimize the SQ vis-a-via the "ringing."

"The next menu item is the filter section. there are 3 selectable filters in the Stellar Gain Cell DAC. There filters have no effect on the analog sources. The filters only operate on PCM digital audio sources and do not affect DSD source material. Digital filters generally have a trade-off between how sharp the filter is and how much ringing results. A sharp filter allows the most high frequency signal to pass through, but at the expense of greater ringing...

Filter 1: Slow Roll-off Linear Phase
This filter is the least sharp and has the least amount of ringing. There is a slight loss in high frequencies that may be noticeable with CDs and other 44.1KHz material. This should not be detectable at all with higher sample rate sources. Because it has very little ringing, we have found it to be the most musical sounding filter choice and we have made it the default filter for the Stellar.

Filter 2: Fast Roll-off Minimum Phase
This filter has better high frequency response than Filter 1, but more ringing as well. There is no pre-ringing, so this is still a very musical sounding filter. Some people may prefer this filter for 44.1KHz source material.

Filter 3: Fast roll-off Linear Phase
This filter also has very good high frequency response. It has slightly less ringing than Filter 2, but it does exhibit pre-ringing. While this filter may actually measure the best in a laboratory, we found it to be the most analytical sounding."​

 

THIS is the article I quoted from above. Also, just to note that Texas Instruments (for example) make circuitry items specifically designed to combat Class D amplifier ringing, called an OUTPUT SNUBBER.

 

Thanks for the article, I'll take a look.

The second part of your post (regarding PS Audio) refers to the response of digital filters in the DAC. Especially at CD sampling rates, engineers have to make decisions about how much of one thing they want compared to another. There are no ideal anti-alias/reconstruction filters, and digital filters have a phenomenon called ringing, and there can be pre- and post-ringing depending on the filter design. As the name suggests, pre-ringing is the appearance of signal before it was meant to happen. Some designers/listeners consider this unnatural, so filter designs are sought that minimize or eliminate pre-ringing at the expense of post-ringing (which is considered less of a problem). Meridian, for one, are big on apodizing filters which remove all the pre-ringing. However, you get nothing for nothing, and have to compromise performance somewhere else (e.g., frequency or phase response).

Here are some articles on the subject:

Ringing False: Digital Audio's Ubiquitous Filter

Archimago's Musings: HOWTO / MUSINGS: Playing with Digital Filtering - Impulse Responses and Frequency Effects resembling Chord, Old-Skool Meridian, MQA, NOS...

I can tell you from my own listening to different filter types that I agree with PS Audio, and prefer slow roll-off linear phase responses.

 

OK, the first article is by Bruno Putzeys, one of the elite digital amplifier designers in the world. His article is largely about the pitfalls of designing digital amplifiers to a (low) price. Bruno himself designs no compromise solutions. He is the brains behind Nypex, Ncore, Kii and Mola-Mola. It seems to me that the article is more about the pitfalls of Class-D done wrong (or cheaply) than about inherent problems. Bruno had a great blog section on the Mola-Mola site, but it was taken down some time ago, however, he points out in interviews that doing digital amplification right is not trivial:

Bruno Putzeys: Head of the Class (D, That is)

If you are considering buying an amplifier based on a Hypex, Ncore or ICE module, you can be assured that the issues identified in your quoted article will have been addressed.

Here is a post I made in another thread that contains some of the blog material by Bruno:

---
The Mola Mola blog page has (sadly) been removed, but I had quoted a section by main designer Bruno Putzeys regarding a panel session he was on with some other big name traditional amp designers in another thread, so, here it is:

Probably the most thought-provoking question that got asked was to give a run-down of advantages and disadvantages of our respective technologies. It's an interesting one because it's clearly framed to fit the quintessentially American preoccupation for "balance". You line up pros and cons so that nobody comes out looking silly regardless of which way they choose.

That kind of balance is often illusory. What are the pros and cons of class D? For once in a live discussion I didn't fumble the answer, which went something like this: whom are you asking? If you're asking the user, keep in mind that they've just bought a small, light amplifier that wastes little energy, has lots of power reserve, sounds great and generally does what it says on the tin (clearly I'm assuming a good class D amp, I believe I make those). I can't think of any disadvantages from a user perspective. Provided the class D amp is fully mature, there is literally nothing on the "con" side of the balance.

Any and all disadvantages land squarely on the designer's shoulders. Class D is an astoundingly complex technology. If you want to do it well, prepare for a lot of math. There are virtually no rules of thumb of the sort "tweak x and y will improve". When you move to class D you get good efficiency straight off the bat. The rest are problems: "how do I make it bullet-proof?" or "how do I prevent RFI" and most importantly "how do I get great sound?". In many linear amps, especially FET and valve based ones, the distortion is to a large degree euphonic so if the design of a good error control system gets overwhelming there's always the escape route of leaving the distortion in and claiming that this sonic footprint was what you were after all the time anyway. Class D offers no such cop-out. The open-loop distortion of a class D amp is not particularly interesting sonically so your only option is to eliminate it completely. That's right: the only good sound you can get from a class D happens to be "identical to the input signal". Well isn't that what high end audio should be about anyway?
---

Some more links for you:

Hypex Electronics B.V.

Mola Mola

 

I've never heard Class-D amps having weaker bass at lower volumes.

Of course, our ears are less sensitive to lower frequencies at lower listening levels, but that isn't an amplifier "problem."

 

of course it is a hearing thing but every class D amplifier I have heard (except for the Rogue Audio Pharaoh) needs to be louder than usual to kick in the bass.

 

This info is based on a complete misunderstanding of basic audio theory.

First of all, you don't want a "clean ring-free square wave" in audio. There is no such thing as a square wave except in the imagination. Switching the power on and off always takes time to transition from "0" to "1" - even with a light switch, and certainly with semiconductor devices. There is no device or even wire that is truly free of inductance or capacitance, as it is a fundamental property of current flow.

A digital amplifier is based on switching the full amplifier power on and off super-fast to approximate the musical waveform. Imagine it like turning a light switch on and off super fast to dim the light (which is how incandescent wall dimmers work), or even how an oven turns on and off to achieve the desired heat. Even a microwave pulses the power on and off at 50% to approximate the output.

A light bulb or oven range heater element have a low-pass filter built-in - it takes time to heat or cool the element and air or pan, so fast switching is not necessary. You can watch a radiant heat stove top turn its element on and off. Likewise, a loudspeaker has a built-in lowpass filter - at frequencies above the audio range, the mass of the tweeter diaphragm starts becoming cumbersome, and the signal cannot be reproduced any more - much like how you can shake your fist, but can't shake it that fast while holding a dumbbell.

What is being described as "ringing" above is a natural effect of the switching electronics not being able to provide infinitely fast signal transitions. A real "square wave" instead has a shape described as the "Gibbs phenomenon", because the frequency bandpass characteristics of electronic devices don't allow the infinite number of harmonics to describe such a wave:


The wiggles, "ringing", are a natural part of signals. This is actually how you might expect to see a 1kHz square wave on a 50MHz oscilloscope, where you calibrate the probe capacitance for the least overshoot or undershoot. The more Gibbs effect there is, the less "RF emission" present, because we know the signal is being lowpass filtered - the complete opposite of the allegations in the marketing technical fluff.

The digital amplifier makes audio by turning the amplifier power rail on and off very fast, similar to 1-bit DSD or sigma-delta digital converters at 2.8MHz or above. When the pulse-width or density signal switches closer to frequency limit, you get more Gibbs effect, which is still an accurate signal:



But you know what? We don't want that square wave in our signal at all! We want to filter out the fast switching of the square wave so we don't burn out tweeters. We need to recover the original audio waveform that is made of hundreds or thousands of faster pulses. The square wave ringing, and even the square wave itself is smoothed out by a low-pass analog filter network, so it is never seen at the output terminals.



The PWM encoding uses techniques that keep the quantization noise (created by inaccuracies from pulse rate and width, much like quantization noise from low bit rate) out of the audio band, so they also can be reduced by the low-pass filtering. You can see how the noise increases at frequencies above 20kHz, so gradual lowpass filters are used that barely affect the audio range. These have very little bandpass ripple or ringing. Whatever argument the author of the quote above is having with himself is also a non-issue.





You can see an interesting phenomenon in the Gibbs effect plots - there are wiggles in the signal even before the square wave transition. Of course signals can't time travel - a filter imparts a phase delay. Very sharp filters, such as those used to anti-alias CD audio, can introduce this effect, "pre-ringing", where a sharp transient has audible effects. Different filter design can instead move the ringing to after the transition, or different implementation, such as oversampling or high-frequency conversion, can move the effects above our hearing. Again, digital amps don't need such steep filters.

The challenge in digital amp design is in other areas - fast enough switching devices, accurate time domain switching, voltage supply regulation or droop compensation in ADC modulation method, low quantization etc. Audio-only filtering of the digital amp output is a requirement, but a very low complexity problem in overall amp design.

 

I have a couple of Class "D" Crown amps that I have had for years. I used one for the stereo and the other powers the passive commercial subwoofer.

I really don't know what "ringing" is, but I have never experienced anything abnormal at all with either amp.

They both sound fine.

The one that was connected to the commercial sub gets quite a workout for years without having any issues.

 

Thanks for all the information! The main takeaway for me is that ringing is NOT an audible whining or a sustained noise of any type. It is simply the name of an electronic phenomenon that Class D amp designers have have to deal with. How they deal with it can have effects on the music produced by the amp, which is why PS Audio gives the listener some discretion in determining the type of filtering used. I will definitely read the articles, thanks for the links!

 

Crown for home use, how do you deal with the fan noise, have you modified it or it doesn't bother you?

 

Thanks for all the information. For me, this is a pretty dense field, with lots of pre-knowledge needed to even begin to really understand all the implications. But I think I did grasp the concept that ringing is a "natural" characteristic of the class D "process" and is not really anything that an end user needs to know much about, as it is handled by well designed Class D amps.

What threw me, more than anything, was just the term "ringing," which I took to be an acoustic phenomenon. I kept seeing it mentioned here and there without explanation. If they had called it something like Wave Flux I don't think I would have been as concerned.

 

Thanks for the response. I have a couple of Class D powered subwoofers and I haven't heard any unexpected sounds or noises. This will be my first Class D amplifier as a component, however, and "ringing" cropped up several times in my reading. Of course, I'm now learning not only what it is, but that it is shouldn't be of very much concern to the user.

 

The Audioholics article is dead-on accurate, but it deals with basic amplifier class definitions and the (original) engineering issues that amp designers had to deal with in each class. Anybody choosing Class D amplification products today will benefit from a couple of decades of excellent development and a wide range of superior-sounding products. No ringing.

 

Hi no you won't hear the fan noises I was asking about. My question was specific to this user as he is using professional amplifiers that use fans to could down transistors and the fans can be quite noisy.

 

Yeah, that's news to me. My mi-fi Pioneer receiver with their D3 amps certainly isn't bass shy. I auditioned it against receivers costing 4-5 times as much and thought it sounded as good or better. Glassy smooth but not harsh. No "grunginess".

 

No problems here with my Marantz PM-10S1 Reference integrated amp.

 

The Crown's are at the bottom of a portable 19" audio rack behind the TV.

I've been back there while they are running and the fan's are never really on.



The Altec lansing theater speaker's that I used to use with the one Crown amp are so sensitive that it takes virtually no power or strain on the amp to operate them. For years now, I have been running them strictly with tube amps, currently a small 3.9-Watt Decware single ended amp.

I have two other sets of Altec Lansing A7, Voice Of The Theater speaker's and I purposely set up this audio rack with thc Crown's so I could do a portable PA, if I wanted.

On the sub, the Crown runs in bridged mono mode and the commercial horn loaded sub is also highly sensitive, so I doubt that I am taxing the either of the amps at all.

Neither are ever more than room temperature to the touch. If the fan ever does come on, you would never hear it.

The volume indicator lights hardly ever move.

 

Interesting to know, must be the newer ones that are quieter, I have an older Crown, it sounds like a DC3 and the fan is always on at maximun, then I have a Couple of Audio Crest that while the fan noise is lower and varies with volume it is still audible. I am planning to make some modifications but not yet as I am not using them until I have a proper room. I wish to use them with an active crossover, I like the sound but the fans were always a concern and I am trying to gather all possible info from people who use Professional amps in a home set up. Thanks

 

The class D amp “

I wonder what kind of SPL would be achieved before those fans would have to spin up...