Sonic T-Amp

[Old Guy]

If you guys are happy with the T-amps, fine. Subjectively they sound OK.

I did show the link to Tripath's spec sheet for the module.

If it's called a T amp it has to have a TriPath chip. And by Tripath'a own published specs there are problems. If people can't hear them, that's one thing. But to deny they are there when the manufacturer says they are doesn't work.

I heard things way before I saw the spec sheets.

5 or ten years from now this technology will rule the earth. But it ain't ready for prime time.

[cclark65]

The problem I've always had with digital amps is, inherent in the design, there's a lot of HF junk. This has to be filtered out. I can hear these filters- and I'm not a youngster.

My comments weren't just about that amp but digitals in general. I still haven't heard one I would buy for HF use. I'd apply the same criticism to many of the tube amps I've heard, and a lot of solids also, however.

The amps I use for a living ( mostly QSC) would never find their way onto my stereo. Pro power amps these days are optimized for full throttle operation. They actually have higher distortion figures at a few watts out- which of course is average listening levels.[/QB]

<font size="2" face="Verdana, Helvetica, sans-serif">I have the same problem with digital amps. I have 2 racks of QSC PLX amps driving subs, midbass, and midrange, but when it comes to the horns, forget it. I stick with some Altec 9442 amps to drive them.

[Old Guy]

Hi Tom- we may just be quibbling semantics.

I do hear stuff I don't like in these amps. It's getting fixed fast. The state of the art in just the last couple years has advanced tremendously.

Maybe if I put it this way?

How long have tubes been around? I'm sure there were tube amps in 1940 that everyone thought were wonderful. The state of the art advanced and there came much better ones.
Solid state amps, ditto.

Just wait till you see what the digital amps will be like in a few years. It's not a mature technology yet. It's coming. There are going to be ones that will blow you away.

Bear in mind I use amps hard. The distortion in digital amps rises rapidly at higher frequencies with increased power output, and yes that's very documented. It's getting fixed. But it won't be a quantum jump. It'll be millions of little ones.

[Tom Brennan]

"But to deny they are there when the manufacturer says they are doesn't work."

I deny nothing Old Guy, I'm just saying they sound good. A perfect technology we have nowhere---SS, tubes, chip-amps, whatever. Pick your poison.

[GM]

Originally posted by Old Guy:
The problem I've always had with digital amps is, inherent in the design, there's a lot of HF junk. This has to be filtered out. I can hear these filters- and I'm not a youngster.

My comments weren't just about that amp but digitals in general. I still haven't heard one I would buy for HF use. I'd apply the same criticism to many of the tube amps I've heard, and a lot of solids also, however.

The amps I use for a living ( mostly QSC) would never find their way onto my stereo. Pro power amps these days are optimized for full throttle operation. They actually have higher distortion figures at a few watts out- which of course is average listening levels.

<font size="2" face="Verdana, Helvetica, sans-serif">Greets!

Agreed and understood, the modded one was somewhat filtered and I had to further filter it with a damping resistor to suit myself, so when all was said and done, it only had a couple of 'clean' watts.

GM

[Zene]

Gentlemen ... BTW, Hi Old Guy, the 34's are 70% finished. (That's a special coded message).
Every time I read about the little T's I get frustrated at the comments; like, "It's cute for the money", or "It doesn't keep up with my $4000 tube amp". Well, DA!
It isn't even close to 5 or 6 watts. Here is a quote from Michael Mardis (hope he doesn?t mind?).
"Let us take the Tripath chip as an example. It can deliver about 8V peak before clipping. Let us take this 8V maximum level before clipping and call it 0dB. The average musical signal is 18dB below this. Thus: 8V-18dB = 1V.
That's 1V RMS, as the -18dB level is also RMS. 1V RMS into 8 ohms will give us? not much. Only 0.125 watts RMS. 1/8th watt or 0.25 watts RMS into 4 ohms. That's all.
So if you want to run this amp WITHOUT CLIPPING, average musical power will be only 1/8th watt. (More or less depending on the music you like ...."
Trying to run these on typical full range speakers is a worthless test. You guys using horns are heading in right direction. Bi or tri-amp them. I will be quad if not oct-amping (sp?) on my system.
Bi-amping my P.audio coaxes with them is proving to be a great combination.
Do the math, you need very efficient speakers.
The top end seems to support the raise in distortion shown on Michael's graphs. Do not expect it to do something it tells you in plain English that it cannot do well. Properly modded for hi freq is another story. Read all of his site and eMail him for help. He is a great guy.
Zene

[Panomaniac]

Aloha all,
just thought I'd crash the party as my name had been mentioned.

Some of you here know me from the other forums and email. You may also know that my favorite speaker of ALL time is a modified A7 with a 15 cell horn on top.

But to the subject of T-amps, about which I know some things.

In 10 points:

1) They are NOT digital. It's a common mistake. Class-D just means "switching amp". D was the next letter for class after C, no digital implied.
There are some amps that might be considered "digital" as they are a sort of power DAC. Must of them are said to sound pretty bad.

2) Class-T. This is just Tripath marketing. The Tripath chips do take a slightly different approach to switching, but not enough to warrant a new class. Tripath amps switch up around 1MHz, much higher than other Class-D amps.

3) Power. The small Tripath chips (here after "T-Amps) are limited by the low voltage rail. Usually 12V, but ranging as high as 14.6V. Most can give a pretty clean sine wave within a volt of the rail. All the small chips are bridged, for more effective power.

4) The famous Sonic Impact may be the cheapest of the T-Amps, but it also sounds that way. Modified, it can sound pretty good, but there are better T-amps out there - for more $$, natch.

5) Tripath do make bigger, more powerful chips.

6) We have yet to pull all the magic out of these chips. Not only is the technology young, so is the field practice. As was posted above, tubes have a long head start! In my work with the T-Amps I have made huge improvements in the sound, and I've only just begun.

7) Frequency response IS load dependant. The output filter must be designed for a certain impedance load. High impendence loads tend to peak on the high end, lower impedance loads droop at the top.

8) The output filter is VERY important to sound quality. The OP filter is where the rubber hits the road. Where RF becomes AF. Not easy to find much written about this, other than the basics. Much research remains to be done in this area. Think of how much the OP tranny of a tube amp can influence the sound. A lot of the harsh high end of the T-amps is simply bad filters.

9) Are they as good as tube amps? Especially DH Triodes? Yes and no. They do some things better, some things worse. Many guys who have heard my amps won't go back to their SETs. Some are a bit more cautious.

10) Class-D amps are the wave of the future. In the mainstream they will be promoted for efficiency and low cost. Fortunately for music lovers, they can also sound darn good. We are to a point, I believe, that it is now useless to build Solid State Class AB amps.

And that's enough for now.

Aloha
Panomaniac. AKA Michael Mardis

[Old Guy]

PWM is essentially on off pulses. While technically not digital, it's pretty much the same thing. I think we're just quibbling semantics. Much stuff labeled digital isn't, as far as that goes. I think the battle's already over on this one- PWM amps will continue to be called digital. Time will tell. I am certainly not always right.

You are correct about the higher sample rates being better, and the importance of the output filter. I'm in 100% agreement there.

While you're here- what's your opinion of the stuff CAD Audio in Denmark is doing? They are geared more towards Pro Audio than the Tripath stuff.

[Panomaniac]

Originally posted by Old Guy:
I think we're just quibbling semantics.

<font size="2" face="Verdana, Helvetica, sans-serif">Pretty much, yeah. But I do want folks to know that most switch mode amps are controlled in a completey analog manner. The outputs are fully on and off - a square wave that changes duty cycle - but that duty cycle is modulated in an analog fashion. There [i]are[/] some chips that modulate the square wave output from a standard didgtal signal. That is probably the future of Class-D

I think the battle's already over on this one- PWM amps will continue to be called digital.

<font size="2" face="Verdana, Helvetica, sans-serif">Agreed. A lot of terminology is like that. It does lead to a desire for a "digital" input to the "digital" amp. I have to explain that the amps don't take a digital input, and there is no way around that, they don't work that way. In a few years, I suspect, many of them will. For better or for worse.

I know the CAD stuff only from what I've read on the web. I'm no longer in pro sound, so a bit out of touch with that stuff. A buddy at QSC works on Class-D and says there will be more and more of it.

[Old Guy]

I agree that many PWM amps will go completely digital. A major driving force is going to be increased use of DSP. The future of amps is going to be the DSP functions built right into the amps. A few are already doing this.

Unfortunately. most manufacturers will use **** speakers and use DSP to make them sound decent, rather than use good speakers and make them sound great.