Any interest in a budget ($100-200) subwoofer “shootout”?

[dasdoing]

yes

 

[KaiserSoze]

To clarify, when I say room correction, I mean so I can perform the max SPL tests indoors. Using Klippel’s ISC module, one just needs a proper “reference” (outdoor, NFS, anechoic) and then other measurements can be taken indoors. This is useful for various reasons. In this case it keeps me out of the summer heat and isn’t as annoying to my neighbors.

https://www.klippel.de/products/rd-system/modules/isc-in-situ-compensation.html

There is an enormous amount to digest in that Klippel page, but from what I was able to grasp from it and from your comments and pictures I now have a better appreciation for how it is possible for the Klippel instrument to take measurements of a speaker indoors and mathematically derive what the measurements would be without the effect of the room. The measurement taken outdoors is a ground plane measurement taken in a consistent, well-controlled manner, and as such it is not difficult to see how it is possible to derive, from this measurement, what the measurements would be if taken with the speaker and the microphone both standing on tall skinny poles. Out of kindness to your neighbor you take the high-volume measurements indoors, where the measurements will be affected by the reflections and the reverberation. But the Klippel apparatus is able to combine the outdoor and indoor measurements in a way that gives you high-volume measurements as though they had been taken outdoors.

So when you said "room correction", you meant something very different from what "room correction" ordinarily means. You meant "room correction" as it applies specifically in the context of the process for taking measurements using the Klippel apparatus.

After thinking about some of this, it did raise a couple of questions in my mind, about how subwoofers should ideally be measured. On the one hand, if you apply EQ to flatten their responses this will narrow the differences in their individual frequency responses, i.e., it will make them all sound nearly the same, at least from the standpoint of tonality in the bass. On the other hand, since most people nowadays would probably apply some form of EQ to flatten the inherent response of the speaker (if not to correct for the room modes), and since the distortion profile of the subwoofer will change when EQ is applied, measurements of distortion taken without any EQ applied would not give the true picture of what the distortion will be after EQ has been applied. This is particularly true with respect to the use of EQ to raise the response of a subwoofer at very low frequency below 30 - 40 Hz, to make the response flat all the way down to 20 or 25 Hz, because when this is done, the excursion of the driver increases dramatically, which produces a dramatic increase in distortion when a very low-frequency signal is applied or when whatever frequency is applied includes very low-frequency content. It seems to me that this is an important question but also a difficult question with no easy answer.

 

[hardisj]

Elac showed up and testing is now done! (Unless I decide to test the Klipsch a 5th time to try to see if I can get a 20Hz reading that isn’t distorted above my noise threshold; no other issues with the other woofers).

A few things I have learned today while testing some of these subwoofers:
1a) There's a whole lot of amplifier limitation (built in limiters) in this price range.
1b) If anyone ever decides to do their own testing, there is no reason to increase the voltage input past 2v if you run the gain wide open. All the limitations seem to be kicking in well before that. I assume the amps are all built with the expectation of relatively low voltage input from the AVR pre-out.
2) Room correction via Klippel's ISC does indeed work as well as I had hoped. I doubted some of the results so I packed everything up and brought it all outside. Three tests later and the results are the same as they were indoors, using the room correction. Oh, well.
3) I like the Elac. Not saying it is the best. But I like it.
4) Depending on your needs, there are different “winners”. Trade-offs. I have plans to approach the data with various viewpoints. IMHO, a single CEA-2010 curve isn’t a full representation of the best option for some cases.


Additionally, with all this “practice”, I can now knock out a subwoofer test in under an hour. That’s 1/10 the time it takes me to test a loudspeaker.


Oh, and when you are making 8 measurements (for averaging; drops the S/N considerably), these subwoofers sound like someone is framing a house next door at the low frequencies and like a woodpecker at the higher frequencies.

Now I just have to go through the data write the review (or skip straight to the YouTube video).

Stay tuned!

 

[KaiserSoze]

Elac showed up and testing is now done! (Unless I decide to test the Klipsch a 5th time to try to see if I can get a 20Hz reading that isn’t distorted above my noise threshold; no other issues with the other woofers).

A few things I have learned today while testing some of these subwoofers:
1a) There's a whole lot of amplifier limitation (built in limiters) in this price range.
1b) If anyone ever decides to do their own testing, there is no reason to increase the voltage input past 2v if you run the gain wide open. All the limitations seem to be kicking in well before that. I assume the amps are all built with the expectation of relatively low voltage input from the AVR pre-out.
2) Room correction via Klippel's ISC does indeed work as well as I had hoped. I doubted some of the results so I packed everything up and brought it all outside. Three tests later and the results are the same as they were indoors, using the room correction. Oh, well.
3) I like the Elac. Not saying it is the best. But I like it.
4) Depending on your needs, there are different “winners”. Trade-offs. I have plans to approach the data with various viewpoints. IMHO, a single CEA-2010 curve isn’t a full representation of the best option for some cases.


Additionally, with all this “practice”, I can now knock out a subwoofer test in under an hour. That’s 1/10 the time it takes me to test a loudspeaker.


Oh, and when you are making 8 measurements (for averaging; drops the S/N considerably), these subwoofers sound like someone is framing a house next door at the low frequencies and like a woodpecker at the higher frequencies.

Now I just have to go through the data write the review (or skip straight to the YouTube video).

Stay tuned!

So, you can do them very quickly now, eh? Not such a big deal now, eh? Hmmm. Did you ever wonder how much better the 3010 sounds compared to the 1010? And for that matter how the 3030 compares to both of them? And speaking of these Elac subs, I've been wondering how they compare to the SVS SB-1000 and the SB-2000 Pro, and I'll just bet that you've been wondering the very same thing!

 

[hardisj]

I can do them quick enough, compared to the 10+ hours it takes me to test a regular loudspeaker.

I actually haven't' been wondering about any of those subs. Probably because they're so expensive and I can't just "order and test" them. But, if someone wanted to accommodate a way for that to happen then they know how to contact me. Alternatively, if we can raise enough funds through my contribution page then I'd be game. https://www.erinsaudiocorner.com/contribute/

But shipping alone is going to be expensive.

 

[raistlin65]

I can do them quick enough, compared to the 10+ hours it takes me to test a regular loudspeaker.

I actually haven't' been wondering about any of those subs. Probably because they're so expensive and I can't just "order and test" them. But, if someone wanted to accommodate a way for that to happen then they know how to contact me. Alternatively, if we can raise enough funds through my contribution page then I'd be game. https://www.erinsaudiocorner.com/contribute/

But shipping alone is going to be expensive.

Plus, because you have tested some subs that Brent Butterworth has, we can interpolate from his data if it's different and use it to compare to your data for subs that you haven't both tested. So should be possible to kind of see how SB-1000 compares to the ELAC sub.

 

[Colonel7]

Elac showed up and testing is now done! (Unless I decide to test the Klipsch a 5th time to try to see if I can get a 20Hz reading that isn’t distorted above my noise threshold; no other issues with the other woofers).

A few things I have learned today while testing some of these subwoofers:
1a) There's a whole lot of amplifier limitation (built in limiters) in this price range.
1b) If anyone ever decides to do their own testing, there is no reason to increase the voltage input past 2v if you run the gain wide open. All the limitations seem to be kicking in well before that. I assume the amps are all built with the expectation of relatively low voltage input from the AVR pre-out.
2) Room correction via Klippel's ISC does indeed work as well as I had hoped. I doubted some of the results so I packed everything up and brought it all outside. Three tests later and the results are the same as they were indoors, using the room correction. Oh, well.
3) I like the Elac. Not saying it is the best. But I like it.
4) Depending on your needs, there are different “winners”. Trade-offs. I have plans to approach the data with various viewpoints. IMHO, a single CEA-2010 curve isn’t a full representation of the best option for some cases.


Additionally, with all this “practice”, I can now knock out a subwoofer test in under an hour. That’s 1/10 the time it takes me to test a loudspeaker.


Oh, and when you are making 8 measurements (for averaging; drops the S/N considerably), these subwoofers sound like someone is framing a house next door at the low frequencies and like a woodpecker at the higher frequencies.

Now I just have to go through the data write the review (or skip straight to the YouTube video).

Stay tuned!

Have you ever tested any of the Dayton Reference subwoofer drivers? I've toyed with building a sub from a flat pack and a plate amp of my choice. It wouldn't be in this budget territory but would be significantly cheaper than an SVS-1000 sealed or ported.

 

[hardisj]

I am working on creating a google sheet to capture the metrics, some extras I think are worth noting and the basic specs. I'll update as I create more tests.

Some things I have taken the liberty to provide:

• Average SPL from 20-31.5Hz, 40-80Hz and an overall average SPL from 25-100Hz. Using 20Hz skews the results because some speakers (the Klipsch in this case) don't give me a result, so those cases have a higher average if I use 20Hz. That's why I chose 25 - 100Hz for the overall.
• I created three "Value" columns where I take the price and divide it by the above average SPL bands. That gets you a "$/dB". I think it's an interesting metric. It would be better to have that broken down by frequency but I think that would make a very "busy" chart.
• I've also created an "Output vs Size (dB/ft3)" column. Pretty self-explanatory there.

So, for example, the Polk is the best value in the 2 small bandwidths I chose. But, from 25-100Hz the Elac seems to be the best value. Either of these two, however, appear to have the best overall "value". But when factoring in space consumed, the Elac clearly bests the Polk.

And then, the Klipsch has the most output from 25-32Hz.

So, you can get a feel for how the numbers may influence a decision one way or another when you consider the various factors as opposed to just being plopped down a number and that's it.

I'm not ready to share the doc yet. But I think you guys will appreciate the work I'm putting in to this test. This is something I want to build on going forward so just tossing the data out in to the world without some context doesn't serve my goal.

 

[KaiserSoze]

Have you ever tested any of the Dayton Reference subwoofer drivers? I've toyed with building a sub from a flat pack and a plate amp of my choice. It wouldn't be in this budget territory but would be significantly cheaper than an SVS-1000 sealed or ported.

I've looked at that option several times but each time I add it all up, including obviously a comparable amp, I end up with something that costs very nearly as much but that requires me to put in a lot of work. I don't mind the assembly so much but the finish is a big deal. I don't want something that looks like it's been spray painted. To get a proper, clean-looking gloss black is a ton of work and based on what I've seen others do it is very, very difficult to get it to look half-decent.

Which particular combination of driver, knock-down box and amp are you considering?

 

[ta240]

I don't know if the overall product or performance is better but with the flat pack build you end up with a much more solid subwoofer. In the same price range they just don't come with that sort of cross bracing and structure.

But I do agree that the finishing the MDF is the main challenge. I built one and ended up with a good smooth finish but it took a lot of bondo, sanding primer and sanding to get there. People with more talent in that area may be able to do it faster but I was happy when my wife said "that looks like you bought it in a store" at the end.

 

[Colonel7]

I've looked at that option several times and each time I add it all up, including obviously a comparable amp, I end up with something that costs very nearly as much but that requires me to put in a lot of work. I don't much the assembly so much but the finish is a big deal. I don't something that looks like it's been spray painted. To get a proper, clean-looking gloss black is a ton of work and based on what I've seen others do it is very, very difficult to get it to look half-decent.

Which particular combination of driver, knock-down box and amp are you considering?

Actually I was asking Erin @hardisj . But since you asked, the 10" HF and any of the cheap 300-500w plate amps (could even do a dsp amp on sale). And yes, I know the SVS are 12" but there ain't no way black ash or white or black gloss will be in my house, which is why I'm interested in the driver testing. I tend to be a patient finisher and have the materials for that. Even have some left over veneer that is probably about enough and I always have some partsexpress coupons. It would just need to be close enough or similar in performance, not equal to SVS but at least much better than the budget boxes. I've been impressed by the SVS teardown but they don't go on sale either except for factory seconds or refurbished only a time or two per year (plus the WAF is real).

 

[KaiserSoze]

I don't know if the overall product or performance is better but with the flat pack build you end up with a much more solid subwoofer. In the same price range they just don't come with that sort of cross bracing and structure.

But I do agree that the finishing the MDF is the main challenge. I built one and ended up with a good smooth finish but it took a lot of bondo, sanding primer and sanding to get there. People with more talent in that area may be able to do it faster but I was happy when my wife said "that looks like you bought it in a store" at the end.

Hey, if the wife says it looks like something you bought in a store, you must have done an outstanding job. The mistake that I think that most people who attempt a mirror finish make is in not realizing just how perfectly flat and smooth the surfaces need to be before applying the shiny stuff. I'm not the least bit confident that I could get it right no matter how much time I spent on it, which is why I would likely not attempt a mirror finish if I built one. I've thought about covering the enclosure with thin sheets of acrylic. If the covering sheets are cut very exactly, with mitered edges that meet up properly, it should be possible to spray the back sides with black and stick them to the enclosure. But of course anyone who has ever tried to build any sort of cabinet with perfectly mitered corners is shaking their head and saying no, don't even try that.

 

[KaiserSoze]

Actually I was asking Erin @hardisj . But since you asked, the 10" HF and any of the cheap 300-500w plate amps (could even do a dsp amp on sale). And yes, I know the SVS are 12" but there ain't no way black ash or white or black gloss will be in my house, which is why I'm interested in the driver testing. I tend to be a patient finisher and have the materials for that. Even have some left over veneer that is probably about enough and I always have some partsexpress coupons. It would just need to be close enough or similar in performance, not equal to SVS but at least much better than the budget boxes. I've been impressed by the SVS teardown but they don't go on sale either except for factory seconds or refurbished only a time or two per year (plus the WAF is real).

Sorry if I butted in, but my curiosity got the best of me because I have looked into this option carefully, several times. I was really only wondering if you had found a solution that I had overlooked.

The SB-1000 is $500 in ugly fake ash wood in black. I lean toward acoustic suspension, and the RSS265HF-8 is particularly good for use in a sealed enclosure (because the -3dB point is determined primarily by how low the Fs/Qts ratio is for the driver, and for this driver it is 25/.49, indicating an F3 around 35 Hz). Unfortunately there do not seem to be any full kits using this particular driver, only the "starter kits" which have the baffle and the internal bracing for the cabinet. The knock-down MDF cabinet that would be needed is $120, and the driver itself is $145. The least expensive amp option is $130, which means it comes to $400 before adding for miscellaneous parts. I don't know what veneer would cost, and I'd most likely mess that up and have to pay for it twice. But even if the cost isn't much lower, the money is much better spent to get something that you actually looks attractive rather than something that is ugly as sin. I strongly dislike ugly fake ash wood in black. Now you've got me thinking again that I should build. One thing nice about building is that if you build two, the savings are greater because the total effort and total cost isn't twice as great as for one. And for me, building two and using the speaker lever inputs is advantageous is certain respects.

 

[Colonel7]

Now you've got me thinking again that I should build. One thing nice about building is that if you build two, the savings are greater because the total effort and total cost isn't twice as great as for one. And for me, building two and using the speaker lever inputs is advantageous is certain respects.

That's the driver.

You're right that DIY is usually for self-satisfaction and not savings esp. for speakers, but subs could be a different case and may even get a custom finish for the price bracket. This Covid business has changed my perspective on DIY because I'm likely to be homebound for the duration.

 

[dddenis]

Thank you for this very interesting test, Erin! I am confused as the results show very high -3dB frequencies of around 50 Hz.
It looks like cheap drivers packed in a too-small enclosure without much EQ. Although those are budged woofers, I expected at least -3dB frequencies of around 30...40 Hz. This looks also like that a loud bass sells better than a deep bass? There are bookshelf speakers which go deeper.

View attachment 82062

 

[hardisj]

Plus, because you have tested some subs that Brent Butterworth has, we can interpolate from his data if it's different and use it to compare to your data for subs that you haven't both tested. So should be possible to kind of see how SB-1000 compares to the ELAC sub.

I found his master list of subs but I can't find any overlap. But, maybe I'm just overlooking it or not looking at the right list. You have a link to any in particular?

 

[KaiserSoze]

Dynamic drivers all have a resonance, the severity of which is different for different drivers and is described by the driver's Q, or Qts. The greater the driver's intrinsic damping, the less severe the resonance and the lower the Q. There is both mechanical and electrical damping, and as such the total Q of a driver is customarily divided into two partials, which sum like conductances in series (or resistors in parallel) to yield the total Q of the driver. In practice the electrical damping is typically about 10 x greater than the mechanical damping but this varies greatly among different drivers. In any case it is generally true that a driver with a low intrinsic Q is a driver with a big magnet whereas a driver with a high intrinsic Q is a driver with a small magnet. (The electrical damping depends on both the length of voice coil wire within the gap and on the total strength of the magnetic field in the gap, flux density multiplied by the gap height.) Thus, low-cost drivers with small magnets tend to have high Q values (poor damping), whereas more costly drivers with bigger magnets tend to have lower Q values (better damping). With both sealed and ported enclosures the spring effect of the air within the enclosure effectively increases the strength of the driver's intrinsic mechanical spring, thereby increasing the effective Q of the driver once it is installed in the enclosure. As such, an inexpensive driver with a small magnet generally requires a larger enclosure so that the spring effect of the enclosure will be weaker, such that the effective system Q will not be excessively high. An excessively high system Q will lead to a pronounced peak in the response at the system resonance point, which is essentially the driver's free air resonance shifted higher in frequency.

But subwoofers are also different from big woofers in regular speakers because big woofers in regular speakers are expected to operate over a wider frequency range, which means that Qtc (the Q of the driver once installed in the enclosure) needs to be in that rough range of .6 to .9. With subwoofers the required frequency range is not nearly as wide, and because of this, subwoofers generally have high Q values compared to big woofers in regular speakers. One obvious reason why this is desirable within limits is that in order for Qtc to be lower, the enclosure would need to be bigger, sometimes a great deal bigger. But this isn't the only advantage of high Qtc. When the Q of a driver is high, the weak damping means that less power is required to make the driver oscillate, as desired to produce sound, i.e., the speaker will be more efficient and will have higher sensitivity. This is especially desirable with a subwoofer because of the amount of power that is fundamentally required to produce deep bass at high volume. A high-Q subwoofer requires a less-powerful amplifier compared to a low-Q subwoofer.

The strategy of using a Qtc value somewhat higher than would be useable for the woofer in a regular speaker is used to varying degrees in subwoofers generally. In general, an inexpensive subwoofer will make strong use of this strategy and as a consequence its frequency response will resemble the hump of a dromedary. More expensive subwoofers will generally use this strategy to a much milder degree, and consequently they have much flatter frequency responses that extend in both directions, into the deep bass but also into higher frequency, which enhances their ability to interface properly with the main stereo speakers without leaving a broad dip in the mid-bass region.

 

[hardisj]

Have you ever tested any of the Dayton Reference subwoofer drivers? I've toyed with building a sub from a flat pack and a plate amp of my choice. It wouldn't be in this budget territory but would be significantly cheaper than an SVS-1000 sealed or ported.

I have not.

I did purchase (2) of the Ultimax 18" subs a whiiiile back. But I wasn't too fond of them. So I sold them and went with (2) Stereo Integrity 18" subs (don't recall the exact model number) in a 4ft3 enclosure, tuned to 18Hz, IIRC. Still have 'em in my home theater. They go wubba-wubba quite well.

 

[hardisj]

FWIW, I have a buddy who works at PE. I messaged him last weekend to see if he might be able to put me in touch with someone there to see if I could get some test samples of their subwoofers. He was selling his house and forgot to check for me. I pinged him again this afternoon and he said he'll ask his coworkers tomorrow.

I also just emailed Monoprice the same thing. Basically just saying "people are asking if I would test your stuff, but I can't afford it, so would you be willing to send me a test sample that I would send back when the testing is complete".

Fingers crossed.

- Erin

 

[hardisj]

Also, if you guys know anyone or happen to represent a company that produces subwoofers, feel free to have them get in touch with me so we can try to work something. I'm honest. I can be brutally honest. But I only ever do that if it's something that's way overhyped via YouTubers and I want to shame them for recommending crap.