• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Amplifiers Which Test Well, Under $1,000 USD

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
The heatsinks in the shape of the company's name are very distinctive. Would love to have 4 or 6 of these on a rack driving an active setup.

Also, on the discussion about ultrasonics, there's a NC100HF dedicated for tweeter use. Anyone know anything about that? Rolls off below about 500Hz, but I wonder if there was any work done to it in the ultrasonics.

I use the NC100HF modules in my own DSP active speaker. They work well. They have to be used in conjunction with the NCxxxMP modules as thats where they get the PSU and control from.

https://www.audiosciencereview.com/forum/index.php?threads/3-way-dsp-xo.991/post-43650

Just to be clear these are the NC500MP modules which are different to the NC500. I use the NC502MP (2 channel version) in my P502

The NC100 response is 10Hz to 50kHz, the xo filtering needs to be done elsewhere.

n1.PNG
 
Last edited:
OP
Ron Texas

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,077
Likes
8,913
The heatsinks in the shape of the company's name are very distinctive. Would love to have 4 or 6 of these on a rack driving an active setup.

Also, on the discussion about ultrasonics, there's a NC100HF dedicated for tweeter use. Anyone know anything about that? Rolls off below about 500Hz, but I wonder if there was any work done to it in the ultrasonics.

I want heat sinks which say Go Texas, LOL. (University of Texas in Austin)
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
I have an old Acurus A250 gathering dust. It was $995 in 1994 or so when I bought it.

Here's someone who measured his: https://insta-stalker.com/post/BtrlP4QBEyP/

It says "modded" but still...

"My rebuild/moded Acurus A250, measurements: Pic1: Full load just before clipping, 2 x 320watts @ 1KHz, THD+N 0.013% with a 6.6ohm load (average load of an 8ohm speaker). 1 channel driven: 349watt. Voltage drop at PSU is only 3V between one channel load or both channels driven, so the transformer is doing it's job pretty good. Maximum THD+N measured were 0.076% @ 10KHz. Also look at the harmonic distortion after 1KHz and up, looks pretty good! Under normal operation the noise floor were close to the limit of our scope.. that's under THD+N 0.01%!! "

1550723144892.png


"Pic2: Noise floor; 50Hz hum from the transformer is now completely eliminated with the new starground. Also the 100Hz from the HexFred-bridges is at a level where it really can't be any better then this... we wouldn't be able to hear it anyway."

1550723272121.png


"Pic3: A 10KHz square wave: Yellow is the target curve, Green is the amplifiers output. To come closer to target I could go lower on C13/14, but they are 220pf now, it's minimal how much this would change selecting a smaller value, as the transistors has an internal C itself, open/close time... anyway! - this test was also OK."

1550723192096.png


---

I'd rate mine perfectly adequate. Somebody else's picture:

1550723648182.png
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
The old Acurus A250 has H2 at (105-25) -80 dB and beautiful harmonics profile, with monotonically decreasing harmonics.

index.php


How it sounds with complex music like orchestral? I ask because the H2, H3... are > -90 dB.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
I like very much the BVaudio PA300SSE measurements:

http://www.bvaudio.sk/merania pa300sse.htm

Usually we listen music at home with < 10 watts continuous. Measurements at 8-10 watts are very informative.

Very clean and monotonically decreasing without odd harmonics.

Spektrum pre 1kHz/10W/8ohm

1k10W.png


and very clean too.

Spektrum pre 19+20 kHz/10W/8ohm

CCIF10W.png


The question is how much NFB (negative feedback) to have these graphs and numbers? Maybe I must send a mail to Ladislav Bunta. http://www.bvaudio.sk/kontakt.htm

Numbers: http://www.bvaudio.sk/pa300sse.htm

BVaudio-PA300SSE-specs.png
 
Last edited:

b1daly

Active Member
Forum Donor
Joined
Dec 15, 2018
Messages
210
Likes
358
A couple of "dumb questions" for any of the smart folks here inclined to take a stab at them...

- Why does the beat effect that arises between signals of steady frequency not occur in the US range? So if you had a 50khz and 51khz tone, wouldn't the voltages sum in the wire in the same fashion, resulting in a 1k artifact?

- From what getting by reading various posts here, some class D amps have low pass filters for the US signals and some don't? Is the fancy filter Amir bought from Audio Precision required only to measure amps that lack this LP filtering?

- I have to admit the thought of these ultrasonic artifacts getting into the signal makes me queasy, though I understand that most transducers can't reproduce them (nor can we hear them). But if that's the case, why is any LP filtering needed at all at the output of the amp?

This is all quite interesting.

The root of my unease I think is that I experience a sort of pain in my ears with certain recordings, and certain speakers. It's one of those things that I'm not sure if I'm imaging or not. Sometimes it's almost a pain in the jaw. But it's not created by obvious audible signals.

One theory I have is that it's caused by weird aharmonic distortion introduced in the recording and mastering process, and possibly also an artifact from listing to 320kbit mp3 streaming files.

It could also be that modern audio gear can actually capture and reproduce sounds up to 20khz, or higher. Since the experience of "bright" music can sound more clear, there is a tendency to use EQ to boost high frequencies past where they would naturally occur. For example, a violin will have a natural frequency spectrum (including ultrasonic components.) If you put a high frequency shelf on it, which people do all the time, you are significantly creasing the energy in a part of the spectrum where are ears are not that sensitive. Nonetheless, the energy is there, and could cause over-stress of the hearing system.

In any case, I've taken to listening to music with a low pass filter, dropping down significantly the frequencies from 17khz or so up to 20khz. I'll futz with the slope depending on the speaker, but overall I find this to "sound better" to me:)
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de
- Why does the beat effect that arises between signals of steady frequency not occur in the US range? So if you had a 50khz and 51khz tone, wouldn't the voltages sum in the wire in the same fashion, resulting in a 1k artifact?
What made you think that this wasn't the case? Of course this effect occurs in the ultrasonic range as well. It's a form of amplitude modulation. An AM detector could demodulate this into a 1 kHz tone.

However, a perfectly linear system would not create any additional components besides 50 and 51 kHz. For a real 1 kHz signal to appear, you need some form of nonlinearity. (From which it follows that yes, an AM detector is inherently nonlinear. It is most obvious with a classic diode detector, but also applies to mixer-based topologies, in which case additive mixers makes it the most obvious. That's mixers in the RF sense, i.e. multipliers, not the friendly adders you find in audio.)
- From what getting by reading various posts here, some class D amps have low pass filters for the US signals and some don't? Is the fancy filter Amir bought from Audio Precision required only to measure amps that lack this LP filtering?
Any Class D amp without any kind of output lowpass filtering is penny-pinching bargain basement crap (think 3$ amplifier boards shipped from China) that would cause substantial RF interference with any length of speaker wire. The hopes of passing emissions testing (as mandated e.g. by the FCC) like that are pretty much exactly zero. After all, what makes switch-mode amplifiers so efficient is their fast transients, but on the downside those contain components up to very high frequencies.

Years ago my parents bought a Panasonic all-in-one micro system that turned out to have pretty lousy FM reception with the supplied wire antenna. You may be able to guess what the problem was - the built-in Class D amplifiers were wreaking havoc on FM near the device, confirmed by placing another radio next to it. That's the 100 MHz range. Running coax up to the attic and placing the very same antenna there fixed the issue. I assume that the Class D amps are of the particularly problematic single-ended output variety (the bridged output variety makes it much easier to get rid of common-mode components, and I think just about any IC for more than a few watts is built like that nowadays).

AFAIK, even with filters in place, the remaining ultrasonic components may be strong enough to skew AP measurements. So that's what the extra filter is for.
 
Last edited:

Ilkless

Major Contributor
Forum Donor
Joined
Jan 26, 2019
Messages
1,757
Likes
3,437
Location
Singapore
Wow, that is a beautifull case. How is that possible for such price o_O

Surely someone on ASR could bear to try them out. ;) I can at least tell for sure the company is legit and actually shipping out products (not vapourware), because there is a reseller of their preamps local to me in Asia.
 

KNielsen

Member
Joined
Feb 22, 2019
Messages
16
Likes
5
I have an old Acurus A250 gathering dust. It was $995 in 1994 or so when I bought it.

Here's someone who measured his: https://insta-stalker.com/post/BtrlP4QBEyP/

It says "modded" but still...

"My rebuild/moded Acurus A250, measurements: Pic1: Full load just before clipping, 2 x 320watts @ 1KHz, THD+N 0.013% with a 6.6ohm load (average load of an 8ohm speaker). 1 channel driven: 349watt. Voltage drop at PSU is only 3V between one channel load or both channels driven, so the transformer is doing it's job pretty good. Maximum THD+N measured were 0.076% @ 10KHz. Also look at the harmonic distortion after 1KHz and up, looks pretty good! Under normal operation the noise floor were close to the limit of our scope.. that's under THD+N 0.01%!! "

View attachment 22291

"Pic2: Noise floor; 50Hz hum from the transformer is now completely eliminated with the new starground. Also the 100Hz from the HexFred-bridges is at a level where it really can't be any better then this... we wouldn't be able to hear it anyway."

View attachment 22293

"Pic3: A 10KHz square wave: Yellow is the target curve, Green is the amplifiers output. To come closer to target I could go lower on C13/14, but they are 220pf now, it's minimal how much this would change selecting a smaller value, as the transistors has an internal C itself, open/close time... anyway! - this test was also OK."

View attachment 22292

---

I'd rate mine perfectly adequate. Somebody else's picture:

View attachment 22294
Not to hijack this thread.

Stumpled across my measurements via google;)

Yes it's heavily mooded.
Sound absolutely great and plays orchestra (and anything else) with no problem.

Regards
IMG_20190125_150208-896x664.jpg
 

KNielsen

Member
Joined
Feb 22, 2019
Messages
16
Likes
5
Wow, that is some serious wiring going on in there. :)
Yes. A new groundtrace (starground) has been made to separate the noise from "power gnd" and "signal gnd". Works amazingly well! Also AC input has been moved to the middle to get the caps as close to the PCB as possible.
The wirring is all 10awg throughout.
- I could go on, long list ;)

Not quite finished yet.

And new measurements must be taken in the near future.

Regards
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL

KNielsen

Member
Joined
Feb 22, 2019
Messages
16
Likes
5
Do you happen to have measurements from its un-modded state?
No.. but I have two Acurus A250's.... One standard and the one shown.
I could if time allows it, collect some data from the original amplifier. Would be fun (I hope) to see what has happened.

THEY DEFINITELY DO NOT SOUND THE SAME :)

Edit: I'll share here when the modded and standard amplifier has med measured
 
Last edited:

ryanmh1

Member
Joined
Jan 11, 2019
Messages
93
Likes
117

That is just a TI TPA3116 based Class D amplifier. The best-case measurements can be found in the application notes from Texas Instruments. There isn't going to be some magic implementation of the product that beats them. Most likely, they will do worse, if anything. Best-case measurements are all over the internet for these things, and they are nothing to get excited about. Well, other than the fact that for $25 you can get a 25W power amplifier into 8 ohms.

The problem is that while the parts and circuitry that go into a high performing DAC are quite cheap, and the potential results spectacular for $10 worth of parts, no one has yet designed a cheap Class D amp that has exemplary performance for a low price. Even an ideal implementation will measure horribly, even if it sounds acceptable.
 

KNielsen

Member
Joined
Feb 22, 2019
Messages
16
Likes
5
Top Bottom