Jump to content

Bellari EQ570 Analog EQ Measurements


Langston
 Share

Recommended Posts

The stimulus for this post came from here. I thought a separate topic would be more appropriate as others may be interested in what turned out to be a well-made, yet inexpensive little analog EQ for those who want to keep things simple. I haven't measured an analog EQ in years and it brings back memories of my favorite console before Yamaha almost single-handedly converted the concert production world to digital.

 

Component quality and PCB layout are very good, switches and knobs have excellent feel and alignment with no play, power adapter has a snug fit, designed and built in the USA by a company I've dealt with for over a decade in the installation industry. American quality at a Chinese price - I am impressed.

 

Front

1200961808_BellariEQ570Front.thumb.jpg.fac1e1d448c36c7c8f50f23de4d2752e.jpg

 

Rear

1428540569_BellariEQ570Rear.thumb.jpg.d7da57d7c5d0baa00734179fc5ea6af1.jpg

 

Interior 1

292296691_BellariEQ570Internal1.thumb.jpg.999c828789718ed59cea09edc098e4d4.jpg

 

Interior 2

798435781_BellariEQ570Internal2.thumb.jpg.7fa62164a1f6c99080575712cd95c85d.jpg

 

Unbalanced consumer audio equipment is generally standardized to a 2V RMS max output, whereas professional equipment is standardized at 12V RMS max. The Bellari EQ570 passes audio through its buffers whether or not the EQ is engaged and its max output is 3.5V RMS (without boost EQ) before distortion starts to rise. The unit has very little noise as well. The first two plots follow Amir's technique at ASR, which I think is an excellent snapshot of signal vs. noise and distortion at the standardized voltage level. SINAD is the reciprocal of THD+N and probably a bit more intuitive. It is a measure of how much higher a 1kHz test tone is than the noise plus distortion in a device (at 2V RMS in this case).

 

Dashboard with EQ Bypassed

766532950_Dashboard2VEQOut.thumb.PNG.886b3d009203fdd54b47a4b6e5fdc16f.PNG

 

Dashboard with EQ Engaged, but Flat

1646918432_Dashboard2VEQIn.thumb.PNG.9a1997b8ad4619ed3f3bdce0e2bf34ff.PNG

 

Magnitude vs. THD from 100mV to 2V with EQ Bypassed

812311478_BellariEQ570RMSvsTHDEQBypassed.thumb.png.dd642d6cf9ec87e92f24d2ca334738dd.png

 

Magnitude vs. THD from 100mV to 2V with EQ Engaged, but Flat

776425351_BellariEQ570RMSvsTHDEQFlat.thumb.png.2461e406734883bd708c9ca596b68cbd.png

 

60Hz Low Shelving Filter at 1V (see plot legend for knob positions)

902631946_BellariEQ570MaxMinEQ60Hz.thumb.png.0bc11b9ecd97d5d861db1aaf07edc61f.png

 

400Hz Peaking Filter at 1V (see plot legend for knob positions)

891864234_BellariEQ570MaxMinEQ400Hz.thumb.png.644f289b87b0cea0e760ba79f404ce51.png

 

2kHz Peaking Filter at 1V (see plot legend for knob positions)

833969592_BellariEQ570MaxMinEQ2kHz.thumb.png.fdc680f729d62e0b4f96c13862f7d4ef.png

 

7.5kHz High Shelving Filter at 1V (see plot legend for knob positions)

1056488521_BellariEQ570MaxMinEQ7_5kHz.thumb.png.229d58ca10edbb5cbfa4b3f9faa8d374.png

 

All Filters at Max and Min at 1V

1242021391_BellariEQ570MaxMinEQAll.thumb.png.612a240932d710d100073ec1c2934100.png

 

God bless you and your precious family - Langston

  • Like 5
  • Thanks 1
Link to comment
Share on other sites

I bought one of these little jewels to help with a peak in my room around 2khz and it's done a great job of it.  I'm sure I could use some room treatment for reflections but I'll get to that. This EQ has made listening more enjoyable.  Thanks for your measurements! 

  • Like 2
Link to comment
Share on other sites

On 10/5/2021 at 10:47 PM, Subway said:

 

I don't know. : ) I do know about the original product from Aphex called the Aural Exciter, and the Bellari gizmo is likely a simple distillation of it.

 

Don't read past this point if you value your time or sanity. : )

 

Background

The WWII battleships were some of the loudest places on earth occupied by people and they had a terrible time communicating with the crew. Thus they trained the heck out of them to minimize the need, but the need to communicate changes in the theater of action was still there. Sound system power was quite limited at the time, but they found that if they pushed the system pretty hard into distortion (10%+ 3rd harmonic), they could get the equivalent of about 10dB+ louder communication ability without additional voltage from the amplifier or excursion from the compression drivers in those reentrant horns. The clipping amplified the consonants in speech aurally (perception), and it is the consonants that provide understanding (articulation) in speech. It wouldn't be until the 70's that Peutz would develop a speech intelligibility metric he called the Articulation Loss of Consonants (ALCons), but the major ground work had been laid during the war.

 

So. We learned we could make huge apparent increases in communication based power without actual increases in electrical or acoustic power.

 

Radio and the Loudness War

The 80's saw a distinct change in the way music was broadcast over FM. Whereas the earlier decades of radio worked from fairly fixed definitions in broadcast technology and end-user reproduction systems, the 80's saw wild advances in both domains. While the earlier decades had to compete based on content, the 80's had the advantage (?) of maturing processing capabilities that the Beatles pretty much started in the 60's with multitrack and large scale concerts. Thus compression and hard limiting was used by radio to increase the apparent loudness of their transmission, but this initially brought complaints from both record producers and end users. A smart guy at Aphex used the clues in Peutz's work and started clipping audio on purpose and found a range of adding distortion that mimicked the natural harmonics of the bass and midrange in such as way that it highlighted them without actually increasing electrical power dissipation. Thus we tend to perceive more bass and vocal output/clarity even though the actual volume is unchanged. This is exactly what 80's radio wanted because it subjectively offset the heavy handed early compression dullness by "restoring" the liveness of the sound and (perceptually) making it even louder!

 

Of course recording studios applied it to vinyl and tape reproduction formats as both have the same hard limits as radio transmission. Concert sound began experimenting with it as an effect even though they weren't faced with the same headroom limitations. It's in use today via plug-in's in digital recording and concert production, though Aphex still offers the rack mount version. Dave Levine (aka Rat) used it for the Chili Peppers* - he's one of the last of a dying breed that still mixes on analog consoles with outboard racks.

 

2021

What's old is new again.

 

God bless you and your precious family - Langston

 

* Dave became the first and only (AFAIK) FOH engineer for the Red Hot Chili Peppers decades ago when he rented a modest concert system to them in a small venue before they were known. The bassist (Flea) stripped naked during the show and spray painted the mic stands and a bunch of Dave's other stuff yellow! Afterwards, a very not-rich Dave pleads to the group that he can't afford to fix/replace the stuff and the group paid for all the damage. They fell in love with each other that night and have stuck together since then.

  • Like 3
  • Thanks 1
Link to comment
Share on other sites

Thanks Langston. 

 

"The first "big" album to use the Aural Exciter was Fleetwood Mac, Rumors."

Soundoctor - Aphex

 

"What do the mid-’70s albums of Fleetwood Mac, Jackson Browne, and Led Zeppelin all have in common?

A slice of legendary outboard gear: the Aphex Aural Exciter."

How the Aphex Aural Exciter breathed new life into the studio (happymag.tv)

 

 

  • Like 2
Link to comment
Share on other sites

On 10/7/2021 at 12:37 PM, Moosh Bronsun said:

is it me or is there a decent amount of deviation between the left and right channels at each stage?

 

Good observation - we're been spoiled by a digital world where everything is spot-on. With most analog stuff, variations between L/R like this are common, though the EQ570 is probably better than most, maybe better than all near this price range.

 

Still it doesn't matter. Worse case is less than 0.5dB and loudspeakers with L/R differences less than 2dB are rare. Then you have room modes and reflections and non-symmetrical furniture arrangements and imperfect listening positions and...

 

I think it's reasonable to consider the EQ570's channel balance flawless considering all that other stuff. : )

 

God bless you and your precious family - Langston

Link to comment
Share on other sites

20 minutes ago, Langston said:

 

Good observation - we're been spoiled by a digital world where everything is spot-on. With most analog stuff, variations between L/R like this are common, though the EQ570 is probably better than most, maybe better than all near this price range.

 

Still it doesn't matter. Worse case is about 1/3rd of a decibel and loudspeakers with L/R differences less than 2dB are rare. Then you have room modes and reflections and non-symmetrical furniture arrangements and imperfect listening positions and...

 

I think it's reasonable to consider the EQ570's channel balance flawless considering all that other stuff. : )

 

God bless you and your precious family - Langston

Figured as much but didn't recall that on the SBAF Loki measurements (though the Loki is 2x the price).

 

Strongly considering this move.

 

P.S. I find it interesting that the curve relaxes with lower amounts of cut/boost and moves lower in the frequency range. Its as if the 2k pole could double its boost/cut it would end up at 3k+. Wild.

 

Thanks again for the exhaustive measurements, I'd also be curious as to what happens when say: cutting the 2k pole a quarter turn but then boosting the 7.5k pole by a quarter turn. Would it make the Q factor steeper on the upper end (the 2k pole)? The mind boggles...

Link to comment
Share on other sites

On 10/7/2021 at 2:04 PM, Moosh Bronsun said:

Figured as much but didn't recall that on the SBAF Loki measurements (though the Loki is 2x the price).

 

A few things of note re: the SBAF Loki measurements:

 

1. The ARTA measurements are single channel (the second is used in loopback to remove the sound device transfer function errors), so you can't judge channel balance with them. They were performed nicely - I can tell that even the max. boost traces are undistorted and noise-free. The guy makes no mention of the drive voltage(s) used, thus we have no idea if the Loki can output undistorted EQ boosts in the neighborhood of the 2V RMS standard for home audio.

 

2. The Audio Precision measurements have the advantage of showing both channels, but the guy made several mistakes that make the results much less useful. The first mistake is that he selected 100dB Y-axis scales for a device that can affect gain by +/-12dB at most, thus it's very difficult to see channel imbalances or anything else. The second mistake is that while the Audio Precision system is an integrated (software and hardware) platform, the guy doesn't bother to tell us the drive voltage(s) used. Thus the Loki might not be able to get anywhere near the 2V RMS standard for home audio equipment, or maybe it can exceed it like the EQ570, we can't tell. The third mistake is that he didn't provide results (as seen in the ARTA plots) of each knob position an end-user is likely to use, but instead shows a few useful results along with some simultaneous mixing of the EQ's which is interesting, but far less useful. The forth mistake he makes is he drives the Loki significantly beyond its linear range into distortion and doesn't seem to realize it (post #29). The heavy fuzz on Ch1 is one example as are the wiggles in the midrange traces (500Hz - 2kHz). Both are tell-tales of non-linearity caused by distortion and/or high noise levels and/or inappropriate FFT parameter settings in the measurement.

 

3. The ARTA measurements are the most useful because they show how the EQ behaves at various knob positions. From them I can see:

 

Like the Bellari EQ570, the Loki has a low frequency shelving filter. Unlike the EQ570, the remaining (3) filters are peaking types. While I slightly prefer the Loki's high frequency filter bandwidth, the lower (3) filter designs are so bad that it makes the unit a non-starter regardless of any other specifications it may boast. The low shelf filter affects a bit too much of the spectrum above 100Hz at likely in-use settings, but the real deal breakers are the critical mid-band EQ's around 400Hz and 2kHz. Those two bands show useful shapes only at their max. and min. gain settings. At the lower settings where they are much more likely to be used, the curves flatten out to the point of uselessness and affect way too much of the spectrum.

 

IMO, the Loki is useful for its bass and treble knobs alone. I'm amazed this thing ever got off the bench. After reading several reviews with measurements of other Schiit products, I am at a loss figuring out what they're up to. Some stuff appears to be quite good, while other stuff is mediocre or worse. Fascinating.

 

God bless you and your precious family - Langston

Link to comment
Share on other sites

7 minutes ago, Langston said:

 

A few things of note re: the SBAF Loki measurements:

 

1. The ARTA measurements are single channel (the second is used in loopback to remove the sound device transfer function errors), so there's no way to judge channel balance with them. They were performed nicely - I can tell that even the max. boost traces are undistorted and noise-free. The guy makes no mention of the drive voltage(s) used, thus we have no idea if the Loki can output undistorted EQ boosts in the neighborhood of the 2V RMS standard for home audio.

 

2. The Audio Precision measurements have the advantage of showing both channels, but the guy made several mistakes that make the results much less useful. The first mistake is that he selected 100dB Y-axis scales for a device that can affect gain by +/-12dB at most, thus it's very difficult to see channel imbalances or anything else. The second mistake is that while the Audio Precision system is an integrated (software and hardware) platform, the guy doesn't bother to tell us the drive voltages used. Thus the Loki might not be able to get anywhere near the 2V RMS standard for home audio equipment, or maybe it can exceed it like the EQ570, we can't tell. The third mistake is that didn't provide results (as seen in the ARTA plots) of each knob position an end-user is likely to use, but instead shows a few useful results along with some simultaneous mixing of the EQ's which is interesting, but far less useful. The forth mistake he makes is he drives the Loki significantly beyond its linear range into distortion and doesn't seem to realize it (post #29). The heavy fuzz on Ch1 is one example as are the wiggles in the midrange traces (500Hz - 2kHz). Both are tell-tales of non-linearity caused by distortion and/or high noise levels and/or inappropriate FFT parameter settings in the measurement.

 

3. The ARTA measurements are the most useful because they show how the EQ behaves at various knob positions. From them I can see:

 

Like the Bellari EQ570, the Loki has low frequency shelving filter. Unlike the EQ570, the remaining (3) filters are peaking types. While I slightly prefer the Loki's high frequency filter bandwidth, the lower (3) filter designs are so bad that it makes the unit a non-starter regardless of any other specifications it may boast. The low shelf filter affects a bit too much of the spectrum above 100Hz at likely in-use settings, but the real deal breakers are the critical mid-band EQ's around 400Hz and 2kHz. Those two bands show useful shapes only at their max. and min. gain settings. At the lower settings that are much more likely to be used, the curves flatten out to the point of uselessness and affect way too much of the spectrum.

 

IMO, the Loki is useful for its bass and treble knobs alone. I'm amazed this thing ever got off the bench. After reading several reviews with measurements of other Schiit products, I am at a loss figuring out what they're up to. Some stuff appears to be quite good, while other stuff is mediocre and worse. Fascinating.

 

God bless you and your precious family - Langston

Ahh OK. Makes sense (I am not super up on the science but do under most of the numbers, I have worked in recording studios for a while).

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...