Forums

Class D commercial audio amp

Started by N_Cook November 29, 2016
On 11/29/2016 02:29 PM, N_Cook wrote:
> On 29/11/2016 18:17, Phil Hobbs wrote: >> On 11/29/2016 12:39 PM, N_Cook wrote: >>> On 29/11/2016 16:45, bitrex wrote: >>>> On 11/29/2016 11:37 AM, Phil Hobbs wrote: >>>>> On 11/29/2016 11:03 AM, N_Cook wrote: >>>>>> >>>>>> Using PWM input to a Si8244BB predriver soic. After the inductor HF >>>>>> cut >>>>>> filter, on the 8 ohm speaker load there is about 1.5V pk-pk 200Kz >>>>>> constant level of switcher waveform, is that normal? what sort of >>>>>> background level would be normal ? >>>>> >>>>> You think you've got it bad. ;) "Inductorless" Class D amps use the >>>>> speaker inductance and rolloff to smooth out the waveform. No EMI >>>>> problems, no sir, never, of course not. >>>>> >>>>> Sounds like your filter is just knocking the sharp edges off the >>>>> switching waveform in order to pass EMI, rather than actually getting >>>>> rid of the fundamental. >>>>> >>>>> Cheers >>>>> >>>>> Phil Hobbs >>>>> >>>> >>>> If the OP were to give us the approximate reactances of the L and C at >>>> 200kHz and the supply voltages, we could probably actually tell him if >>>> 1.5V pk-pk was a reasonable residual amount of fundamental for a likely >>>> two-pole Butterworth having that break frequency...;) >>> >>> Certainly, the rail voltages for a pair of powerFETs is + and -80V, C is >>> .68uF 400V, inductor is about 30mm diameter toroid with about 45 turns >>> of about 1.5mm wire >> >> Sounds like something on the order of 100 uH, depending on the core >> material. In that case the LC rolloff starts at >> >> f_3dB = 1/(2 pi sqrt(LC)) ~ 19 kHz. It rolls off asymptotically like >> (f_3dB/f)**2, which at 200 kHz is 40 dB down (1/100 times the >> amplitude). So 160V p-p ought to roll off to, yeah, about a volt and a >> half. >> >> Cheers >> >> Phil Hobbs >> > > Thanks for that,just as well my hearing is not much good at 200KHz, nor > my speaker being that responsive at 200KHz either. > Other than winding some extra turns on the toroid and paralleling some > more C, anything that could have been added ? It would have been done by > the manufacturer I suppose, if possible . > There is one of those loop-through ferrite lumps on the internal speaker > wires, but that obviously does not filter much 200KHz
Well, you could cascade additional sections to get a steeper rolloff--three sections would get you about a factor of a million (120 dB). The group delay (phase nonlinearity) tends to get worse though. For extra credit, you could put a notch at the switching frequency by parallel-resonating one of the inductors at 200 kHz. Since the waveform is far from sinusoidal, I'd expect that to hurt rather than help--the parallel capacitor takes over at high frequency, so you lose the benefit of the inductor at the higher harmonics. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On 11/29/2016 01:47 PM, Phil Hobbs wrote:
> On 11/29/2016 01:28 PM, bitrex wrote: >> On 11/29/2016 01:17 PM, Phil Hobbs wrote: >>> On 11/29/2016 12:39 PM, N_Cook wrote: >>>> On 29/11/2016 16:45, bitrex wrote: >>>>> On 11/29/2016 11:37 AM, Phil Hobbs wrote: >>>>>> On 11/29/2016 11:03 AM, N_Cook wrote: >>>>>>> >>>>>>> Using PWM input to a Si8244BB predriver soic. After the inductor HF >>>>>>> cut >>>>>>> filter, on the 8 ohm speaker load there is about 1.5V pk-pk 200Kz >>>>>>> constant level of switcher waveform, is that normal? what sort of >>>>>>> background level would be normal ? >>>>>> >>>>>> You think you've got it bad. ;) "Inductorless" Class D amps use the >>>>>> speaker inductance and rolloff to smooth out the waveform. No EMI >>>>>> problems, no sir, never, of course not. >>>>>> >>>>>> Sounds like your filter is just knocking the sharp edges off the >>>>>> switching waveform in order to pass EMI, rather than actually getting >>>>>> rid of the fundamental. >>>>>> >>>>>> Cheers >>>>>> >>>>>> Phil Hobbs >>>>>> >>>>> >>>>> If the OP were to give us the approximate reactances of the L and C at >>>>> 200kHz and the supply voltages, we could probably actually tell him if >>>>> 1.5V pk-pk was a reasonable residual amount of fundamental for a >>>>> likely >>>>> two-pole Butterworth having that break frequency...;) >>>> >>>> Certainly, the rail voltages for a pair of powerFETs is + and -80V, >>>> C is >>>> .68uF 400V, inductor is about 30mm diameter toroid with about 45 turns >>>> of about 1.5mm wire >>> >>> Sounds like something on the order of 100 uH, depending on the core >>> material. In that case the LC rolloff starts at >>> >>> f_3dB = 1/(2 pi sqrt(LC)) ~ 19 kHz. It rolls off asymptotically like >>> (f_3dB/f)**2, which at 200 kHz is 40 dB down (1/100 times the >>> amplitude). So 160V p-p ought to roll off to, yeah, about a volt and a >>> half. >>> >>> Cheers >>> >>> Phil Hobbs >>> >> >> Drat, I'm too slow! ;-) > > Your philosophy has problems too. ;) > > Cheers > > Phil Hobbs >
That's just like, your opinion, man... https://www.youtube.com/watch?v=pWdd6_ZxX8c
On Tue, 29 Nov 2016 12:41:36 -0700, Jim Thompson
<To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote:

>On Tue, 29 Nov 2016 13:58:06 -0500, Phil Hobbs ><pcdhSpamMeSenseless@electrooptical.net> wrote: > >>On 11/29/2016 01:52 PM, George Herold wrote: >>> On Tuesday, November 29, 2016 at 1:18:54 PM UTC-5, Phil Hobbs wrote: > >[snip] > >>>> Well, if you stick in a cap multiplier with a 10-ms time constant, you >>>> won't have to be bothered by it. ;) >>> >>> Hah, I've got this picture of Jim with a hacked hearing aid. to-93 pac >>> transistors hanging out of his ear. (I'm sure some geek makes transistor >>> ear rings.) >> >>That is a visual the world is not ready for. ;) >> >>It's probably possible to do DSP with a spread-spectrum sampling clock >>to get rid of birdies. The math would be a bit of a challenge. >> >>Cheers >> >>Phil Hobbs > >You may laugh, but I've considered, for non-conversation situations, >such as TV viewing or going to the movies, rolling my own fully-analog >hearing aid, with good muffs... could care less about power >consumption, a daily recharge cycle no problem. Could even have >balance, and adjustable equalization ;-) > >I designed several analog class-A and one class-AB aid(s) in the >mid-70's. > >Sometimes miniaturization leads to crap performance. > > ...Jim Thompson
Maybe just buy this... <https://www.amazon.com/Sennheiser-RS120-Wireless-Headphones-Charging/dp/B0001FTVEK/ref=sr_1_1?ie=UTF8&qid=1480448617&sr=8-1&keywords=sennheiser+headset> and modify between the TV and the Sennheiser transmitter to shape the frequency response? ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | I'm looking for work... see my website.
On 11/29/2016 02:41 PM, Jim Thompson wrote:
> On Tue, 29 Nov 2016 13:58:06 -0500, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> On 11/29/2016 01:52 PM, George Herold wrote: >>> On Tuesday, November 29, 2016 at 1:18:54 PM UTC-5, Phil Hobbs wrote: > > [snip] > >>>> Well, if you stick in a cap multiplier with a 10-ms time constant, you >>>> won't have to be bothered by it. ;) >>> >>> Hah, I've got this picture of Jim with a hacked hearing aid. to-93 pac >>> transistors hanging out of his ear. (I'm sure some geek makes transistor >>> ear rings.) >> >> That is a visual the world is not ready for. ;) >> >> It's probably possible to do DSP with a spread-spectrum sampling clock >> to get rid of birdies. The math would be a bit of a challenge. >> >> Cheers >> >> Phil Hobbs > > You may laugh, but I've considered, for non-conversation situations, > such as TV viewing or going to the movies, rolling my own fully-analog > hearing aid, with good muffs... could care less about power > consumption, a daily recharge cycle no problem. Could even have > balance, and adjustable equalization ;-) > > I designed several analog class-A and one class-AB aid(s) in the > mid-70's. > > Sometimes miniaturization leads to crap performance. > > ...Jim Thompson >
Sure, why not? A TCA0372 would give you all the volume you could need. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On 30/11/2016 04:12, Jim Thompson wrote:
> On 29 Nov 2016 17:02:55 GMT, Rob <nomail@example.com> wrote: > >> Jim Thompson <To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote: >>> On Tue, 29 Nov 2016 16:03:40 +0000, N_Cook <diverse@tcp.co.uk> wrote: >>> >>>> >>>> Using PWM input to a Si8244BB predriver soic. After the inductor HF cut >>>> filter, on the 8 ohm speaker load there is about 1.5V pk-pk 200Kz >>>> constant level of switcher waveform, is that normal? what sort of >>>> background level would be normal ? >>> >>> "Class-D" isn't "audio"... the "D" stands for "distortion" >:-} >> >> Why? Of course a Class-D amplifier could have better linearity than >> a Class-A or Class-AB can achieve. > > (1) The damping factor sucks the most sour of lemons ;-) > > (2) And your "better linearity" claim has the same engineering fact > level as would be put forth by a used car salesman. > > (3) Class-D pros: low heat, small size, CHEAP, thus manufacturers of > consumer grade (aka junk) audio products love it. > > ...Jim Thompson >
I challenge you to design a power amplifier with better output impedance (damping factor), or linearity (distortion) than this one: http://www.hypex.nl/docs/NC400_datasheet.pdf You get extra points if your design is more than half as efficient as that one, with a full-scale two-tone input signal. Even if you can't design one that good, I challenge you to point to one that someone else has designed, with those features. Go on, I dare you. Or stop spouting nonsense. Chris
Jim Thompson <To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote:
> On 29 Nov 2016 17:02:55 GMT, Rob <nomail@example.com> wrote: > >>Jim Thompson <To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote: >>> On Tue, 29 Nov 2016 16:03:40 +0000, N_Cook <diverse@tcp.co.uk> wrote: >>> >>>> >>>>Using PWM input to a Si8244BB predriver soic. After the inductor HF cut >>>>filter, on the 8 ohm speaker load there is about 1.5V pk-pk 200Kz >>>>constant level of switcher waveform, is that normal? what sort of >>>>background level would be normal ? >>> >>> "Class-D" isn't "audio"... the "D" stands for "distortion" >:-} >> >>Why? Of course a Class-D amplifier could have better linearity than >>a Class-A or Class-AB can achieve. > > (1) The damping factor sucks the most sour of lemons ;-)
The output impedance can, as with any amplifier, be decreased using feedback. In this case the feedback point has to be behind the filter.
> (2) And your "better linearity" claim has the same engineering fact > level as would be put forth by a used car salesman.
Of course not. It is much easier to design a very linear voltage-to- dutycycle converter than a linear power amplifier.
> (3) Class-D pros: low heat, small size, CHEAP, thus manufacturers of > consumer grade (aka junk) audio products love it.
That does not have anything to do with the achievable quality, doesn't it? Today, Class-D is not only used for small amps in a consumer device, it is used in high-power stage amplfiers as well, and the high-end audio people love them.
Should 1.5v remnant 200KHz stay constant with whatever frequency or 
amplitude of audio going in? Even at zero audio input there is this 1.5V 
200KHz.
 Rob wrote:

> > That does not have anything to do with the achievable quality, doesn't it? > > Today, Class-D is not only used for small amps in a consumer device, > it is used in high-power stage amplfiers as well, and the high-end > audio people love them. >
** That has been true for over 20 years now. The really ground breaking class D amps were the K1 and K2 models from Crown - released in the mid 1990s. When bench tested as a "black box" it was genuinely hard to tell you were dealing with anything unusual. The only give away was the low heat production so the amp needed no fan. http://adn.harmanpro.com/site_elements/resources/995_1425481007/K1-K2-Data-Sheet-136713_original.pdf THD measures about 0.03% at rated power, no switching noise is evident at the output terminals and the DF is over 3000 at low and mid frequencies. Plus the case is hermetically sealed so no dust ever gets inside. Widely used and still used for pro-audio and home theatre applications. .... Phil
>"Today, Class-D is not only used for small amps in a consumer device,
it is used in high-power stage amplfiers as well," Class D with SMPS lends itself to that application very well because of the light weight. They are moved around alot. People are getting weaker. Being able to bench press 400 does not help you get one of those racks out of the van. In fact working out instead of doing actual hard work could make you more prone to back problems. An Icepower amp made by B&O of about fifty real watts (FTC and IHF do not apply to "professional equipment") is smaller footprint than a CD, and totally self contained. Put the AC in, the signal and connect the speaker. It is an attractive solution for engineers who don't want to do the actual work.
>" and the high-end
audio people love them. " Not all of them. Some shun them like the plague and I bet some can actually tell the difference. but then these are the same type of people who could tell the difference between regular AB or A amps anyway, as demonstrated by Bob Carver a long time ago. Some say the high end sounds shrill, or some similar adjective and I can see why. That output pi filter must have some effect they can hear. Some of these people are the same type who change their output transistors to ones with a higher ft and say they can hear the difference. Some ears are different than others. Generally 20 KHz is considered the upper limit, but is it really ? How steep is the rolloff ? Plus some people could be specifically listening for certain things just like alot of people can hold a conversation with one other in a room with 100 people talking. Put it this way, if you were to match RMS levels of a 10 KHz sine wave and a 10 KHz square wave and can hear the difference, you might just hear the difference in a class D amp. Bob Carver found a bunch of these golden ears ad challenged them that he could duplicate the sound of each and they would not be able to tell the difference. It had already been established that they could hear the difference between for example a Mark Levinson and a Crown or whatever. Well he did it, and did it by just measurements. He had to treat the amp as a black box and measured phase shift, damping and whatever else. But he did this all in the A/AB domain. I am not so sure it is even possible to match it up very closely in class D unless you get the switching speed way up there. there are people who prefer the sound of vinyl and that ight not all be that they like the noise and distortion. Vinyl has a natural rolloff at the high end, whereas a CD drops like a rock. It has to because of the low sampling rate. Oversampling and digital filtering may have improved things but there are still only 44,100 pieces of information every second. Now they have formats that are much better and some claim to be able to hear the difference and I a sure there have been some double blinds done on it. If I were the manufacturer I certainly would do it. Of course I might find one out of a thousand people who can really tell the difference but that would be in all my ads. The type of people Carver found all those years ago. My hearing is to the point where it simply doesn't matter. You could run a class D at 40 KHz and it would probably sound fine to me. On certain material I can barely hear the difference between the 10 KHz on a ten band EQ being all the way down and all the way up. That is at least 20 dB. Supporting my assertion of course is the fact that people with good hearing do not go in for hearing tests. We don't know if they can hear higher than 20 KHz. Some audiophiles do, and even though they sometimes do not test all that good, they can still tell the difference in different amps etc. What is the rolloff ? If they are down 10 dB at 20 KHz, what is it at 30 KHz ? Down 18 dB maybe ? Or is there some steadfast rule that says that absolutely nobody can hear that, perhaps based on the anatomy of the human ear ? Even that is not really provable because not everyone has the same anatomy.
On 11/30/2016 07:00 AM, N_Cook wrote:
> > Should 1.5v remnant 200KHz stay constant with whatever frequency or > amplitude of audio going in? Even at zero audio input there is this 1.5V > 200KHz.
You know how PWM works, yeah? AFAIK unless the amp has some fashion of automatic shutdown after the audio signal is removed for a time, the modulator and output stage will always be active. With no input signal the output waveform will just be a 50% duty cycle square wave at 200kHz, so yes I imagine you'd still be seeing the residual at the load terminal.