Class T amp for 23$, a revolution?

[Boke]

For the ultimate in audiophile snake oil have a look at power cord prices. I was in disbeleif when I first saw them. ( Most likely they are plugged into a spliced up nightmare from the electric company or at best some pennies per foot romex )

I won't name company names. But a two meter power cord will cost you $2,500, and each additional 1-meter increment is $500.

 

[PC-Engine]

Perceived loudness depends a lot on frequency not the amount of WATTS an amp can dish out or the amount of WATTS a speaker can absorb without overheating and ultimately blowing.

You can dedicate 10 GigaWatts driving a subsonic signal, but you ain't gonna hear it...

 

[MPI]

Perceived loudness depends a lot on frequency not the amount of WATTS an amp can dish out or the amount of WATTS a speaker can absorb without overheating and ultimately blowing.

You can dedicate 10 GigaWatts driving a subsonic signal, but you ain't gonna hear it...

Eh... strange post. :?

Obviously when talking about hifi it obviously means 20Hz-20kHz(well, even down to 10-12Hz, infrasound can sometimes be 'felt' rather than heard), and when compensating for percieved loudness we use so called 'A-weighted' dB. And the SPL(Sound Pressure Level, i.e. volume) isn't primarily decided by the wattage "absorbed" by the speaker, but the efficiency of the speaker, aka sensitivity, measured in dB/W@1m. One speaker can deliver 100 times(+20dB) higher SPL than another at the same input wattage. So the wattage the speaker can handle doesn't tell a whole lot.

But wattage in and of itself is overrated anyhow, usually you use less than 1W for casual listening anyway, and the increase in wattage doesn't yield quite as much "oomph" as people imagine. A doubling-up in wattage only yields a 3dB increase in SPL. And a doubling of percieved volume(10dB) requires roughly 10 times the wattage input. One of the main reason for oversized (proper)amps is that you get less distortion by keeping the usage range of the output stage narrower(=more linear, especially for FET transistors).

 

[_xxx_]

Actually, musicians' equipment always has the right ratings. A 50W guitar amp usually _is_ 50W. A 1800W power amp is enough for a 2000 people gig and will never fit in a package weighing less then 10 kg, like what I mentioned above

The power amp surely shouldn't be able to output more power than what the speakers can handle, otherwise it would blow these every time you fire it up...

 

[MPI]

Actually, musicians' equipment always has the right ratings. A 50W guitar amp usually _is_ 50W. A 1800W power amp is enough for a 2000 people gig and will never fit in a package weighing less then 10 kg, like what I mentioned above

Well, actually Class D amps can output kW+ power, with minimal heat dissipation. They've been mainly used in subwoofers(Sunfire Subs had 2,8kW amps, and they were 28x28x28cm!) up until recently, but hifi-grade Class D are coming strongly, a 200W can easily be made without heatsinks at all, the only limiting component is the DC supply.

The power amp surely shouldn't be able to output more power than what the speakers can handle, otherwise it would blow these every time you fire it up...

Actually you can easily regulate the amount of power going into the speakers so they don't blow them out every time... It's called a volume dial.

Having an insufficient amp is MUCH more dangerous to speakers.

 

[_xxx_]

Well, actually Class D amps can output kW+ power, with minimal heat dissipation. They've been mainly used in subwoofers(Sunfire Subs had 2,8kW amps, and they were 28x28x28cm!) up until recently, but hifi-grade Class D are coming strongly, a 200W can easily be made without heatsinks at all, the only limiting component is the DC supply.

Class D is not in any musicians equipment, since these can't handle the dynamic range needed, hence the use for woofers only

Actually you can easily regulate the amount of power going into the speakers so they don't blow them out every time... It's called a volume dial.

If a company builds an amp+speaker combo that should make sense, they'll build a system where you can set your volume to max without blowing anything out. Everything else would be unacceptible and certainly could never pass the QA in any company on this planet. And surely it shouldn't.

Having an insufficient amp is MUCH more dangerous to speakers.

Where did you get this crap? Whoever told you that deserves a good asskick and a muzzle.

 

[MuFu]

Having an insufficient amp is MUCH more dangerous to speakers.

Where did you get this crap? Whoever told you that deserves a good asskick and a muzzle.

That's not crap, it's been one of the cornerstones of good audio system design for years. A maxed out power section will clip and the periodic DC will fry your speakers - tweeters go first usually. Besides, it's always a good idea to have large amounts of headroom for large transient swings, even if your drivers can't handle an equivalent sustained load (which is usually what the handling rating pertains to).

For speaker + amp combos it's a bit different since they're probably probably output-capped at an appropriate level. Still... most cheap hifis start to square out if you turn them all the way up.

Actually, musicians' equipment always has the right ratings.

That is far from the truth, IMHO.

 

[_xxx_]

That's not the point. You're talking about crappy equipment, that's all.

If your power amp has higher output than what the speaker(s) can handle, you'll blow them out EACH AND EVERY TIME you turn up the volume above certain level. That's not the way it should be and certainly not good in any way. Whoever tells you something different deserves what?

Getting his ass kicked and a muzzle


Back to a class D amplifier, here's a nice link for you guys. If you have the slightest idea about electronics, you'll get it why it's not good for making music: Here.

 

[MuFu]

Overload clipping is a very simple concept. I would read up on it before you start on Class D amps.

Well, actually Class D amps can output kW+ power, with minimal heat dissipation. They've been mainly used in subwoofers(Sunfire Subs had 2,8kW amps, and they were 28x28x28cm!) up until recently, but hifi-grade Class D are coming strongly, a 200W can easily be made without heatsinks at all, the only limiting component is the DC supply.

Class D is not in any musicians equipment, since these can't handle the dynamic range needed, hence the use for woofers only

The main reason they are used in subs is because of their efficiency. You can get high output for quite a low cost, small form factor and relatively low heat output.

They are not used in most high end audio equipment because they are very difficult to design with low distortion, their output being directly proportional to the PSU voltage and more subject to push-pull timing errors than conventional designs. High quality PSUs are prohibitively expensive. You can work around these problems by supplementing the PWM stage with analog feedback, but again that adds $$$s.

 

[_xxx_]

They are not used because of the signal quality they deliver, which is rather miserable. Class D amp is essentially no amp at all, but a switch with a low-pass behind it. You lose helluvalot HF, harmonics etc. As you said, to make it sound even acceptably good, it would cost too much and doesn't make sense.

 

[MPI]

Right... that used to be the accepted truth ...in 1997. You're a little behind the times, my friend!

Class D amps have made great strides in the last 2-3 years, particularily when the purpose built digital integrating filters came. The THD has gone from 10% to 0.1-0.01% and the best rivals Class A amps for quality.

Here's a couple of links for ya to get you up to speed since you hail from the mid-90's:

http://www.tactaudio.dk/ - the pioneer of hifi Class D

http://www.lcaudio.com/ - fantastic DIY Class D

http://www.audioholics.com/productreviews/avhardware/IntegraDTC-9.4Review_01.html

http://www.harmankardon.com/product...ENG&cat=REC&prod=DPR 2005&sType=C - Starting to come from big-brand consumer products mfg.

http://www.dbaudio.com/en/systems/amps/d12/ - for PA use

A friend of mine has a ICE Power DIY Class D amp, and it's FAN-f**n-TASTIC. And those 500W hasn't fried his speakers yet.

 

[Guden Oden]

Why would a few ms of DC fry speaker coils I wonder?

 

[MPI]

It's actually not the DC itself that's the problem, even if that CAN hurt the coils if prolonged and of enough current... think "doppvärmare"(cup heater?).

The danger of clipping is that the output practically becomes a square-wave, instead of pure sine. Square wave means a LOT of harmonic overtones(100kHz and up), which the poor tweeters don't stand a chance converting into sound. Which means that it all gets converted into heat. Fzzt!

 

[Dr. Ffreeze]

All,

I am no audio expert (designed a few subwoofer enclosures, working on a few 3-way and a line array) but I would have to agree with MPI.

EDIT: Sorry, I should clarify. I agree with MPI that an underpowered amp can easily kill a set of speakers, even if they came as a package together. If you disagree with this, then you believe that every packaged system can handle turning up the volume to max with zero damage to the speakers? I also agree that class D amps in the past were not of sufficient quality for audio use but they are getting much better and can be used in subwoofer applications.

Dr. Ffreeze

 

[_xxx_]

Guys, you're talking about failures and bad designs. A good amp doesn't EVER get overdriven when in normal use (meaning, not if you don't connect the inputs to the power line of your dishwasher or such ). Which means, It'll always produce the signal which is withing the area of the operating point of the transistor/tube without going into saturation.

EVERYTHING else is bad design.

That's all nice with Class D and all, but do you realize that it doesn't represent the original signal? Do you know what PWM is? Follow the link I provided above and read it. What this produces is not the original analog signal that went in but an approximation through a serial stream of square signals which are than being put through a low-pass.

THAT IS NOT HI-FI (I'll spell it: high fidelity).


EDIT:
Actually, a guitar amp is beautifull for explaining these things with amp/speaker relationships to people. The overdriven guitar sound is a perfect example of the signal driving the amp into saturation, and you can clearly hear the point where it starts happening.

 

[Bolloxoid]

If a company builds an amp+speaker combo that should make sense, they'll build a system where you can set your volume to max without blowing anything out. Everything else would be unacceptible and certainly could never pass the QA in any company on this planet. And surely it shouldn't.

You are not making sense. Setting your volume to the max with an underpowered amplifier will result in clipping noise that will damage the tweeter cones, as already explained here. Underpowered amplifiers are the practical danger to your speakers if you are careless with the volume knob. Blowing your woofer cones with an overpowered amplifier is a much less likely occurence.

 

[_xxx_]

You don't realize that there is no such thing like a "weak" amplifier. There's only a badly designed amplifier, where the pre-amplifier produces a signal which can overdrive the power-amplifier. That is just a very poor design. A cheap-o bad amplifier may be built like that, but that's what you get for the money. All that blah about stuff like "class D hi-fi amps" and "weak amps" is just a pile of crap, no pun intended. That kind of stuff is what I've been doing for the last 15 years or so.


EDIT: overdrive should of course happen with guitar amps, but definitely NOT in any kind of hi-fi devices. All signal ranges, impendances, true RMS etc. are standardized things which must be fullfiled in order to be sold as real hi-fi.

 

[MuFu]

That's all nice with Class D and all, but do you realize that it doesn't represent the original signal? Do you know what PWM is? Follow the link I provided above and read it. What this produces is not the original analog signal that went in but an approximation through a serial stream of square signals which are than being put through a low-pass.

THAT IS NOT HI-FI (I'll spell it: high fidelity).

By that logic, digitally sampled audio isn't "hi fi" either.

You don't realize that there is no such thing like a "weak" amplifier. There's only a badly designed amplifier, where the pre-amplifier produces a signal which can overdrive the power-amplifier. That is just a very poor design. A cheap-o bad amplifier may be built like that, but that's what you get for the money. All that blah about stuff like "class D hi-fi amps" and "weak amps" is just a pile of crap, no pun intended. That kind of stuff is what I've been doing for the last 15 years or so.

:?

 

[_xxx_]

It's an order of magnitude better than a class D amp.

 

[MuFu]

What have you been doing for the past 15 years?