Is my Emotiva BasX A2 limiting the performance of my Polk R700s?

[cscs]

Hey everyone - I'm hoping someone can help me figure out if my Emotiva BasX A2 is limiting the performance of my Polk Audio R700s.

I run Wiim Ultra --> Audient ORIA Mini --> Emotiva BasX A2 --> Polk Audio R700s

I think the set up sounds great for a budget system but I think given the size of the speakers there should be more low-end punch. Using an electronic music kick drum as an example, I feel like the transients are a bit softer/rounder instead of snappy/punchy.

This past week I started going down the rabbit hole to find a solution. First I re-optimized the speaker placement to improve imaging and move the SBIR null up to 200Hz+. I re-calibrated with Sonarworks SoundID room correction to ensure the response is flat & phase corrected at the listening position. I then added a +3dB low shelf at 120Hz to make the low end a bit more exciting. All of this together made a massive difference and has made the speakers come alive but I still feel like the low-end punch still isn't quite there.

I've started to think the amp may not be a great match for the speakers.

- Polk Audio R700 are 88dB sensitivity. Impedance is 4 Ohms but dips to 3.6 Ohms at approx 80 - 200Hz range with difficult phase angles. Full measurements here.
- Emotiva BasX A2 is rated 250W RMS per channel @ 4 Ohms. Specs from the company here.

I've fed the specs into both ChatGPT and Gemini to help me calculate an answer to this. ChatGPT says the BasX A2 is more than enough and Gemini is saying the R700s need more 'grip' so I should upgrade to a higher current amp with much higher dampening factor like the Buckeye NC502MP. It's 500W @ 4 Ohms, 450W @ 2 Ohms (BasX A2 only lists 8 and 4 Ohm power), 27A current and a much higher dampening factor of 2666 (compared to BasX A2's 250 @ 4 Ohms).

Below is the 'best case' scenario calculation that Gemini helped me with factoring in the BasX A2 with the R700s. The hardest frequencies to drive also happen to be where the 'punch' from the kick drum sits so thats what this calculation is aimed at uncovering. Is it realistic that the BasX A2 can actually drive the R700s at 89dB? Are there missing factors that should be taken into account with the calculation?

If someone could please tell me if this is crazy or correct that would be very much appreciated!

 

[DVDdoug]

I doubt the amplifier is the problem. Most amplifiers have flat frequency response over the audio range.

I've fed the specs into both ChatGPT and Gemini to help me calculate an answer to this. ChatGPT says the BasX A2 is more than enough and Gemini is saying the R700s need more 'grip' so I should upgrade to a higher current amp with much higher dampening factor like the Buckeye NC502MP. It's 500W @ 4 Ohms, 450W @ 2 Ohms (BasX A2 only lists 8 and 4 Ohm power), 27A current and a much higher dampening factor of 2666 (compared to BasX A2's 250 @ 4 Ohms).

Mostly "AI confusion"...
If the amplifier can't put-out the necessary current, the voltage will drop and it will clip sooner than expected. But if it's honestly rated for 4-Ohms that won't happen. And if you're not hearing distortion from the bass that's probably not happening.

...We don't calculate "peak power" from the peak voltage like that. If its putting-out 31.6VRMS into 3.6 Ohms on the musical peaks, that's 250W and 7.9 Amps RMS. And with short-term program peaks the voltage probably won't "sag" even if it can't sustain 250W continuously. And in reality, you're probably only occasionally playing the "worst case" frequencies/notes where the impedance drops to 3.6 Ohms.

Can you turn it up to get the "punch" you're looking for? (Most live and dance-club music is VERY LOUD overall, not JUST in the bass.) If so, you might want a bigger shelf-boost, or more boost between about 50-100Hz when listening at more reasonable levels. +3dB is twice the wattage and "noticeably louder" but NOT "a lot louder"...

Or, although your speakers should be "full range" and a subwoofer shouldn't be necessary, you could add a "big" subwoofer to potentially give you more "bass power" and this makes it easier to adjust the bass separately. But without an AVR and "bass management" or other active crossover it can be tricky to integrate a sub.

 

[voodooless]

I then added a +3dB low shelf at 120Hz to make the low end a bit more exciting.

Maybe try extending the low shelf to 150 Hz, or add another slight PEQ around 120 Hz to add some punch?

It may just be your room. Can you post some measurements? Most likely it’s not your amp. Amps don’t have a sound unless broken or deliberately designed. Yours is not and has plenty of power, so not the issue.

 

[witwald]

Hey everyone - I'm hoping someone can help me figure out if my Emotiva BasX A2 is limiting the performance of my Polk Audio R700s.

It is highly doubtful that a correctly operating BasX A2 amplifier is limiting the performance of your loudspeakers. The BasX A2 has a very high damping factor, and anything greater than 200 is enough to make an amplifier's frequency response very independent of the loudspeaker's frequency-dependent impedance curve.

The low minimum impedance of the R700, when coupled with the low sensitivity of the R700 loudspeaker means that it is drawing quite a lot of power at higher listening levels. As there is still quite a bit of energy content in the 100Hz frequency region, which is where the impedance minimum occurs, the power demands can be high. Maybe you are simply overdriving the amplifier in an attempt to reach high SPLs? If so, then a more powerful amplifier will give you more headroom to reproduce low-frequency transients cleanly.

I then added a +3dB low shelf at 120Hz to make the low end a bit more exciting.

By adding that +3dB shelf, you've just doubled the demands placed on the amplifier at low frequencies, so maybe it is overloading?

Instead of changing the amplifier, you might consider adding a 12-inch or 15-inch subwoofer to reproduce the low frequencies. This should easily provide more low-end punch. You could also go for 2x 12-inch stereo subwoofers. Either approach would reduce the power demands being placed on the R700 loudspeakers and their respective amplifier.

 

[cscs]

Maybe try extending the low shelf to 150 Hz, or add another slight PEQ around 120 Hz to add some punch?

It may just be your room. Can you post some measurements? Most likely it’s not your amp. Amps don’t have a sound unless broken or deliberately designed. Yours is not and has plenty of power, so not the issue.

I've experimented with a few EQ settings. Unfortunately I found boosting the 120Hz range alone sounded a bit unnatural compared to the low shelf and extending it higher made some male vocals sound a bit off.

Yellow is old speaker placement
Blue is new speaker placement
Red is new speaker placement with room correction
Green is new speaker placement with room correction and +3dB 120Hz low shelf added

 

[voodooless]

The blue line shows some clear drop off. I assume this is EQ’ed flat on the green graph? Maybe try not to do that?

There also is a null at around 65 Hz. That may also have some effect on the perceived punchiness.

Ultimately, the reality is that these things are not party speakers. You simply can’t fight physics, just bend it until just before the breaking point

 

[cscs]

I doubt the amplifier is the problem. Most amplifiers have flat frequency response over the audio range.


Mostly "AI confusion"...
If the amplifier can't put-out the necessary current, the voltage will drop and it will clip sooner than expected. But if it's honestly rated for 4-Ohms that won't happen. And if you're not hearing distortion from the bass that's probably not happening.

...We don't calculate "peak power" from the peak voltage like that. If its putting-out 31.6VRMS into 3.6 Ohms on the musical peaks, that's 250W and 7.9 Amps RMS. And with short-term program peaks the voltage probably won't "sag" even if it can't sustain 250W continuously. And in reality, you're probably only occasionally playing the "worst case" frequencies/notes where the impedance drops to 3.6 Ohms.

Can you turn it up to get the "punch" you're looking for? (Most live and dance-club music is VERY LOUD overall, not JUST in the bass.) If so, you might want a bigger shelf-boost, or more boost between about 50-100Hz when listening at more reasonable levels. +3dB is twice the wattage and "noticeably louder" but NOT "a lot louder"...

Or, although your speakers should be "full range" and a subwoofer shouldn't be necessary, you could add a "big" subwoofer to potentially give you more "bass power" and this makes it easier to adjust the bass separately. But without an AVR and "bass management" or other active crossover it can be tricky to integrate a sub.

I'm not able to get the punch even when cranked up loud. If anything I think it gets slightly softer. Today I added a high pass filter at 50Hz and kicks sounded noticeably snappier, however it removes the weight.

I would love to add subs but I live in a loft/condo so its going to cause issues with my neighbours.

 

[voodooless]

Instead of changing the amplifier, you might consider adding a 12-inch or 15-inch subwoofer to reproduce the low frequencies. This should easily provide more low-end punch

Punch is really the domain that typically lives above where hifi subwoofer are crossed: > 120 Hz. So unless the OP misinterpreted the meaning of “punch”, this won’t really help much. But for sure a subwoofer will give a fuller more powerful sound.

 

[voodooless]

Today I added a high pass filter at 50Hz and kicks sounded noticeably snappier, however it removes the weight.

Okay, so then do not cut the lows, but shelve them down, let’s say -2 dB below 80Hz or so?

 

[cscs]

The blue line shows some clear drop off. I assume this is EQ’ed flat on the green graph? Maybe try not to do that?

There also is a null at around 65 Hz. That may also have some effect on the perceived punchiness.

Ultimately, the reality is that these things are not party speakers. You simply can’t fight physics, just bend it until just before the breaking point

Those 4 were measured by REW but it doesn't tell the full story because its just 1 mic measurement from the listening position. Here is the L+R speaker measurement from SoundID (38 measurements around the listening position) that was taken right after the blue REW measurement. So the real response is actually much better than how it appears on REW. After room correction is applied those peaks and dips are EQ'd flat (second image).

 

[cscs]

Okay, so then do not cut the lows, but shelve them down, let’s say -2 dB below 80Hz or so?

Thanks for the idea. I'm going to give this a try.

 

[witwald]

I'm not able to get the punch even when cranked up loud. If anything I think it gets slightly softer.

Two effects may be coming into play. Firstly, at high SPLs, the small-diameter woofers may be running out of linear excursion. Secondly, the room is being energized a lot more by the low-frequency energy, and this may linger around and be relatively audible after the input transient has actually died away.

 

[witwald]

... or add another slight PEQ around 120 Hz to add some punch?

Do you mean setting the PEQ around 120 Hz to reduce the signal (i.e., negative gain) so as to add some punch? That seems to be the general way to do it. Wouldn't boosting the level around 120 Hz just make things sound warmer and/or muddier?

 

[witwald]

...I found boosting the 120Hz range alone sounded a bit unnatural compared to the low shelf and extending it higher made some male vocals sound a bit off.

Welcome to the world of the mixing engineer, as this seems to be what's happening here: the EQ is being added here and there to try to "improve" the sound quality, above and beyond what's in the original recording. The audible effects of these types of changes will, of course, vary from one recording to another.

It is good that the anechoic frequency response of the Polk Audio Reserve R700 loudspeakers is nicely within ±2 dB between 40 Hz and 20 kHz. Over a lot of that range the tolerance is ±1 dB. From that perspective, these are relatively neutral loudspeakers. They seem to be well suited to EQ to make them even flatter.

 

[witwald]

Today I added a high pass filter at 50Hz and kicks sounded noticeably snappier, however it removes the weight.

So, reducing the low-frequency power at 50 Hz by 50% removes the "weight" but makes things sound "snappier". This simple modification effectively turns the R700 loudspeakers into compact bookshelf loudspeakers.

I would love to add subs but I live in a loft/condo so its going to cause issues with my neighbours.

Understood. In any case, your loudspeakers seem to have a −3dB point of 39 Hz or so, which is quite low and well-suited to the accurate reproduction of a wide variety of music. If sound leakage affecting your neighbours is a concern, then it might be reasonable to expect that the BasX A2 amplifier won't be being overloaded, as the maximum SPLs might need to be kept in check to some extent.

 

[witwald]

Yellow is old speaker placement
Blue is new speaker placement
Red is new speaker placement with room correction
Green is new speaker placement with room correction and +3dB 120Hz low shelf added

@cscs Can you provide a diagram of the current speaker placement in your room, together with the various room dimensions, distances of speakers to the walls, coordinates of the listening position in the room, etc.? As your speakers are quite flat on-axis, then they should require a minimal degree of DSP equalisation below 250 Hz in order to tame the effects of low-frequency room modes. That is what the room tuning software should be quite good at.

After dealing with the room modes, you may still want to further smooth the already quite flat on-axis response of your loudspeakers. It would be sensible to ensure that you are only equalising the direct sound of your loudspeakers by using a short gated time window of not more than 4 milliseconds or so, in order to remove the effects of early sound reflections from the floor and ceiling. The microphone should be placed at the listening position at tweeter height with the speaker directly on-axis, with the seating removed. It would be expected that the EQ curves derived in this manner for the left and right speakers would be as close to identical as experimental tolerances allow.

 

[witwald]

Those 4 were measured by REW but it doesn't tell the full story because its just 1 mic measurement from the listening position.

But with your setup, isn't that what you need to be aiming for: a good measurement at the listening position?

The "full story" will simply be the usual jumble of effects from room modes and reflected sound. The former can be addressed with good success, but the latter really can't be. "Correcting" the reflected sound cannot be done without deleterious effects on the sound quality, irrespective of the smoothness of the incorrect "corrected" curve. DSP correction needs to be applied judiciously. Even though the 38 in-room measurements have been averaged together (the easy part) and then DSPed to get a flatter response, the room reflections are still doing their thing, and they can usually be ameliorated with some thoughtful passive room treatment (the less easy part) if they are a major problem.

 

[witwald]

Here is the L+R speaker measurement from SoundID (38 measurements around the listening position) that was taken right after the blue REW measurement. ... After room correction is applied those peaks and dips are EQ'd flat (second image).

But what does the on-axis response of the loudspeaker look like when the combination of direct and reflected sound has been EQed flat? It is somewhat concerning that the nominally quite flat and neutral on-axis response of the R700 has been EQed away from that naturally accurate response that was originally engineered into the design.

Is the difference in sound an actual improvement in sound quality, or merely another variation of sound reproduction by the system, which now happens to include considerable DSP EQ in a frequency region where it would seem that it needn't be applied?

Although the textbook flatness of the DSPed response above 200 Hz that is plotted below may be alluring, it may be hiding changes to the sound quality that significantly detract from an accurate, natural-sounding reproduction of the music source material.

 

[cscs]

But what does the on-axis response of the loudspeaker look like when the combination of direct and reflected sound has been EQed flat? It is somewhat concerning that the nominally quite flat and neutral on-axis response of the R700 has been EQed away from that naturally accurate response that was originally engineered into the design.

Is the difference in sound an actual improvement in sound quality, or merely another variation of sound reproduction by the system, which now happens to include considerable DSP EQ in a frequency region where it would seem that it needn't be applied?

Although the textbook flatness of the DSPed response above 200 Hz that is plotted below may be alluring, it may be hiding changes to the sound quality that significantly detract from an accurate, natural-sounding reproduction of the music source material.

View attachment 500436

These are set up in my living room and not in a purpose built listening room so things like SBIR and room gain are big issues that I couldn't solve with speaker placement alone. In order to move the SBIR null to a higher frequency the speakers needed to go closer to the wall. This caused very boomy low end. As seen in the measurement it was 6-9dB louder than everything over 2kHz.

Every speaker and room sounds different and they modify the reproduction of the music in some way. So in my opinion the only way to achieve accuracy is to use room correction. A ton of studios & mix engineers use room correction to help ensure the mix translates in the intended way. This is the one I use: https://www.sonarworks.com/soundid-reference and run it on an ORIA Mini.

 

[witwald]

These are set up in my living room and not in a purpose built listening room so things like SBIR and room gain are big issues that I couldn't solve with speaker placement alone. In order to move the SBIR null to a higher frequency the speakers needed to go closer to the wall. This caused very boomy low end. As seen in the measurement it was 6-9dB louder than everything over 2kHz.

That boost in the low frequencies is a well-known effect, which was described by Allison in his 1974 JAES paper, "The Influence of Room Boundaries on Loudspeaker Power Output". Your experimentation has confirmed its existence. Placing the R700 loudspeaker up against the wall raises the frequency at which the first null in response occurs at approximately 150 Hz or so. At a 3 metre listening distance, sound reflection from the floor will add another null occurring at about the same frequency. The sound reflection from a 2.4 metre high ceiling will produce a null at about 50 Hz.

Every speaker and room sounds different and they modify the reproduction of the music in some way. So in my opinion the only way to achieve accuracy is to use room correction.

Using room correction below 250 Hz or so will definitely pay dividends. At a stretch, it might also be useful to apply it below 500 Hz. Of course, treating deep nulls is not really possible. However, there seems to be no disputing the fact that correction/equalisation of the low-frequency Allison room boost, as well as the peaks introduced by modal room resonances, will improve the reproduction of music. Controlling the low-frequency behavior of sound reproduction in a room is very helpful for enhancing the perceived sound quality.

A ton of studios & mix engineers use room correction to help ensure the mix translates in the intended way. This is the one I use: https://www.sonarworks.com/soundid-reference and run it on an ORIA Mini.

The ORIA Mini seems like a very capable unit. In your experience, is it possible to get it to perform room correction only below 250–500 Hz? Also, can it be used to independently correct for a non-flat anechoic loudspeaker response so as to get the direct sound to be flatter on axis (say above 500 Hz)? The two corrections could then be added together to produce a combined correction curve, one that fixes the low-frequency in-room behavior and the on-axis response at frequencies that lie above the room's Schroeder frequency.

@cscs I'm very curious to learn of a dozen or so studios and mix engineers that use room correction. Can you provide a list?