Audio amplifier testing help wanted!

[cheesehead]

I just bought two pairs of good, old-fashioned audio amplifiers off a guy on a forum. They're homemade units - his own design - but use high-quality transistors, and are reputed to be quite good. (The ones I bought are the RTD60A and RTD100, if you're wondering.)

http://www.regaliatech.com/

He claims .05% THD . What's the best way to test said amplifiers' performance, assuming I don't just have a magic distortion measuring box? (Apparently, such things do exist. They also cost a LOT. )

 

[RaynorWolfcastle]

.05% THD at what power level and using what input tone frequency? The easiest thing to do would probably be to playback a single tone and measure the output using an oscilloscope. Honestly, it's not worth the trouble, any decent amp played at a reasonable volume will not have audible distortion (I doubt anything less than a couple of percent can be heard).

 

[Analog]

Your specs are lacking some specifics. Usually, distortion is rated at a power level over a set of frequencies at a specific load. For instance, 200W into 8 ohms from 20 - 20KHz, 0.05% THD. That tells me the load, the wattage, the bandwidth etc. To do that sort of measuring, a precise THD analyzer may be necessary. Otherwise, you'll have to do some comparisons on your own.

THD is really only one measure, there is IMD, phase distortion to name a few. I've tested some tube amps in the 1% THD levels and they are the best sounding amps I've ever heard. Caveat: don't let the number on distortion fool you. Low distortion doesn't mean a good sounding amp. If you don't have the test equipment, give them a listen and see for yourself. Test them at their lowest impedance handling to see if they oscillate. That's what I would do.

 

[f95toli]

0.05% distortion for a transistor amp isn't very good assuming it is a value measure well below clipping into 8 ohms; but it isn't terrible either.
"Normal" distortion figures for a transistor amp are between 0.001% and maybe 0.2% and shouldn't really be frequency dependent in the 20 Hz-20 kHz band.

However, THD doesn't really tell you much about how a good an amplifier is, unless of course THD is several percent which would sounds terrible coming from a transistor amp (odd harmonics).

 

[Howard]

Strange that you tell subjectivists to suck it when the final word in audio is completely subjective? ESPECIALLY when we don't have a universally-accepted concrete metric for how good something sounds?

 

[cheesehead]

Originally posted by: Howard
Strange that you tell subjectivists to suck it when the final word in audio is completely subjective? ESPECIALLY when we don't have a universally-accepted concrete metric for how good something sounds?

Well, we do.

Ever take a look at the Nelson Pass amplifiers? Or perhaps the Halcros? They're considered the best amplifiers in the world for a good reason: intermodulation distortion (IMD), harmonic distortion (THD), and noise are all so low you can't measure them.

Of course, they also cost about as much as a small car.

Furthermore, the Halcros are about as complex as your average Tokamak, with more gain stages and feedback loops than a whole recording studio. On the other other hand, the designs of Nelson Pass, using a more minimalist approach, simply use a dozen output transistors per channel biased so that, while highly linear, the whole amp is maybe 20% efficient on a good day. (They also make great space heaters - turn up the volume, and you can warm your house!)

Admittedly, some people do like the small amount of harmonic distortion created by most tube amplifiers. However, it drives me absolutely nuts anywhere outside a guitar amplifier. Perhaps the best stereo system I have ever heard consisted of a nice pair of Paradigm Reference series towers and a Pass Aleph amp - products with the numbers to back up their claims.

Originally posted by: Analog
Your specs are lacking some specifics. Usually, distortion is rated at a power level over a set of frequencies at a specific load. For instance, 200W into 8 ohms from 20 - 20KHz, 0.05% THD. That tells me the load, the wattage, the bandwidth etc. To do that sort of measuring, a precise THD analyzer may be necessary. Otherwise, you'll have to do some comparisons on your own.

I looked through his design notes on the DIYaudio forum. It looks to be a pretty solid design, with low distortion shown on the scope even at 100w into 4 ohms at 10khz.

To be honest, these amplifiers won't be used anywhere near even 100 watts per channel. In fact, I'd estimate a maximum of 20 watts into 8 ohms from the smaller amp, and 40 watts from the larger ones. These will be used for a pair of bi-amped speakers with an active crossover.

Phase distortion is also something of a non-issue. I'll also have phase distortion from the active crossover, and by mucking around with the active filters, one can cancel out the other.

 

[Howard]

Originally posted by: Cheesehead

Originally posted by: Howard
Strange that you tell subjectivists to suck it when the final word in audio is completely subjective? ESPECIALLY when we don't have a universally-accepted concrete metric for how good something sounds?

Well, we do.

Ever take a look at the Nelson Pass amplifiers? Or perhaps the Halcros? They're considered the best amplifiers in the world for a good reason: intermodulation distortion (IMD), harmonic distortion (THD), and noise are all so low you can't measure them.

Haaa... the Halcro specs are complete bullshit, to say the least, and Pass' amps aren't known for ultra-low distortion. You're talking out of your ass here.

THD, as an indicator for general sound quality, is absolutely meaningless. IMD might be a little better but good luck finding that spec on everything.

 

[mjrtoo]

If you base sound quality upon pure specsmanship, you are doing yourself a great disservice. Some of the very best sounding amplifiers are tube amplifiers, and they are pretty much based upon distortion. You could have an amp with very low THD, but sound like pure crap...

Put down the spec book, plug in the amp and listen to how it sounds. Howard is correct, audio is and always will be purely subjective, and I would have it no other way.