Jump to content

Height of impedance peaks for vented enclosures


lowerFE

Recommended Posts

I'm trying to better interpret the impedance results I'm measuring from my PR enclosure as there is barely a left impedance peak. 

 

I believe the height of the left and right impedance peak signifies the efficiency of the port. What does a low impedance peak height signify vs a high impedance peak height? What if the left is much shorter than the right peak? What if it is the other way around? 

 

Thanks

Link to comment
Share on other sites

The best advice I can give is to use a simulation program and use it to look at the way the system design changes the impedance curve. Start by settling on a tuning. Change the enclosure volume while keeping tuning the same and watch how the response and impedance change. Then look at how the tuning will change the impedance curve while keeping the rest constant. Lastly you can look at various drivers in the same alignment and see how that changes the curve. This is the best way to learn the tradeoffs and how the impedance curve reflects the performance.

  • Like 1
Link to comment
Share on other sites

Painting in broad strokes here,

 

The height of the impedance peaks are related to the height of the impedance peak if it was a sealed box. Higher BL product = higher impedance peaks, higher damping = lower impedance peaks.

 

The impedance minimum between the peaks corresponds to system tuning. This is an excursion minimum for the driver where most of the output is coming from your port/PR. This resonance splits the sealed box's impedance peak into two.

 

The relative heights of the two impedance peaks corresponds to how much of the sealed system's impedance peak is above or below system tuning. Tuning well below Fs results in a larger in-band impedance peak and a smaller below-tuning impedance peak. See Ricci's data on the XXX Ported system or the TC Sounds LMSR-12 at 13Hz for examples of extreme low tuning.

 

These are generalizations. Simulations will tell you a lot more.

  • Like 1
Link to comment
Share on other sites

We know that in sealed boxes, impedance peaks increase Xmax within their bandwith. See the Xmax Investigation thread for more.

 

There is some evidence that Xmax is increased at the impedance peaks in resonant enclosures.

 

Caveat: The site has very few data points for both sealed and resonant enclosures of the same driver, and most of those change other factors as well. Comparison to simulations has limited accuracy, especially at output extremes, so take the following conclusion with a grain of salt.

 

It appears that distortion is reduced and displacement-limited output is increased near the impedance peaks in resonant systems. Greater than predicted by the effect of the port alone.

Link to comment
Share on other sites

The impedance curve I measured is very different than simulation. In the simulation, I could never get the left impedance peak to be as small as what I measured. In fact, I can't really change the height of the impedance much at all. When I tune lower, the left peak moves further to the left and drops in height a bit, but nothing drastic. 

Link to comment
Share on other sites

What PR are you using? What driver are you using?

 

Are you using any padding or damping materials?

 

What is your intended tuning frequency? Does this correspond to the measured impedance minimum between the peaks?

 

When I tune lower, the left peak moves further to the left and drops in height a bit, but nothing drastic. 

This part is normal behavior.

Link to comment
Share on other sites

What PR are you using? What driver are you using?

 

Are you using any padding or damping materials?

 

What is your intended tuning frequency? Does this correspond to the measured impedance minimum between the peaks?

 

This part is normal behavior.

 

Currently I'm using 2 of the Peerless 830878 3.5'' passive radiator loading a Tang Band W4-1720 in a enclosure with about 2 liters of net volume. The passive radiators have additional mass added to further lower tuning to ~47Hz judging from the impedance sweep. I do have a good amount of damping material inside. 

 

I cannot verify simulated tuning vs measured tuning because I cannot measure the weight of the epoxy I added to the PR, and I don't have a sensitive enough scale to measure the weight of a small 10/32 bolt and nut added to the PR. I added more and more nuts until I hit my desired tuning. I actually wanted to hit a 45Hz tuning, but I couldn't add anymore nuts to the short bolt, so I settled with 47Hz since this is an experimental build to learn about what happens when I try to get bass out of an extremely undersized enclosure. Eventually I will do a "Ricci style" build where I go to extreme efforts to get as much >40Hz bass as I can from a tiny speaker. 

Link to comment
Share on other sites

Yeah, based on your application your results make a lot of sense.

 

The padding is what's killing your impedance peaks. It reduces system efficiency by reducing both impedance and system sensitivity. It's a necessary evil though because if you plan on using this driver above a few hundred Hz you'll need to tame box resonances. You can try seeing how much padding you can remove before box resonances become problematic.

 

"Ricci style" isn't terribly concerned with efficiency, and sensitivity even less so. It's more about putting a crap-ton of power into the low corner where box size determines sensitivity, and using components with enough displacement and thermal durability to handle the juice.

 

Your lower impedance peak is further reduced by using such a small enclosure, which pushes system resonance upward. Again, unavoidable in such a small enclosure. Your lower impedance peak doesn't matter much anyways though, since a resonant system won't have any appreciable output that far below tuning anyways.

Link to comment
Share on other sites

"Ricci style" isn't terribly concerned with efficiency, and sensitivity even less so. It's more about putting a crap-ton of power into the low corner where box size determines sensitivity, and using components with enough displacement and thermal durability to handle the juice.

 

 

Hey now...I'd say that's not entirely accurate. :P

I definitely consider efficiency and sensitivity. However if the system runs out of displacement or thermal handling very early it's sort of pointless in this day and age of extremely cheap high output amplifiers. Also once things start to heat up and behave in a less than linear manner, those small signal efficiency advantages start to disappear. The less efficient design on paper may actually become the more efficient one once operating at realistic levels seen in use. Depending upon the application of course.

Link to comment
Share on other sites

Yeah, based on your application your results make a lot of sense.

 

The padding is what's killing your impedance peaks. It reduces system efficiency by reducing both impedance and system sensitivity. It's a necessary evil though because if you plan on using this driver above a few hundred Hz you'll need to tame box resonances. You can try seeing how much padding you can remove before box resonances become problematic.

 

"Ricci style" isn't terribly concerned with efficiency, and sensitivity even less so. It's more about putting a crap-ton of power into the low corner where box size determines sensitivity, and using components with enough displacement and thermal durability to handle the juice.

 

Your lower impedance peak is further reduced by using such a small enclosure, which pushes system resonance upward. Again, unavoidable in such a small enclosure. Your lower impedance peak doesn't matter much anyways though, since a resonant system won't have any appreciable output that far below tuning anyways.

 

Oh wow I did not know damping material kills system efficiency. I'm gonna remove it then. There is a small midrange chamber that is fully stuffed, so I might be able to get away with no damping material since the woofers only do 300Hz and below. I'll test it out if that works.

 

And yes, the "Ricci style" build is gonna stuff a lot of woofers and a few hundred watts into a 14'' x 6'' x 6'' speaker. I'm still considering whether to put 2 5'' woofers + 2 5'' PR in the back or 4 5'' woofers sealed. I've attempted a 4 woofer dual opposed build, but it failed because there's some sort of phase cancellation and I'm getting way less output than expected. I never figured out why there was a cancellation, but I suspect I need a divider to have only 2 woofers share an airspace to avoid cancellation. While 4 woofers will get me more overall output, I'm concerned about the 40-50Hz range most because that's arguably the most important range for music bass, and the PR system could be a lot cleaner in that range. 

 

That said, I would like to study up on enclosure design at a very deep level before I attempt something like this again. While these projects are unbelievably fun, they are rather time consuming and expensive, so I'd like to optimize as many things as I could before I buy and build. However, I'm not having much luck finding books on these. All the stuff I learned about CEA2010, or even stuff like power compression, vent compression, group delay, and woofer behaviour at high power levels are from forums, and I wish I can read a book or something to learn about speaker and enclosure design in great detail and have in depth knowledge like Josh and Contrasseur. 

Link to comment
Share on other sites

Oh wow I did not know damping material kills system efficiency. I'm gonna remove it then. There is a small midrange chamber that is fully stuffed, so I might be able to get away with no damping material since the woofers only do 300Hz and below. I'll test it out if that works.

 

 

Our Bass Overlord was kind enough to publish his findings on the subject.

 

The impedance peak gets lower as padding is added. Since voltage is held constant, and P=V^2/Z, more current flows and more power is drawn from the amplifier. Power in has increased.

 

You can see the sensitivity drops as well, so power out is reduced.

 

Efficiency is power in vs power out, so on both fronts efficiency has been reduced.

 

It's usually not that big of a deal though. Ricci's case study shows impedance peaks down by maybe 30%. This change reduces efficiency only in THE MOST EFFICIENT bandwidth for the system. Sensitivity in a sealed box is only down a couple dB, so in many cases, maximum output is not affected since most systems run out of excursion before thermal handling at these frequencies. As you can see in Ricci's conclusion, it's usually always a good idea to have some damping in the system. It's just about finding the right amount for your application. Remember, a resonance peak out of the intended bandwidth can be audible if the low pass filter doesn't attenuate it enough.

Link to comment
Share on other sites

And yes, the "Ricci style" build is gonna stuff a lot of woofers and a few hundred watts into a 14'' x 6'' x 6'' speaker. I'm still considering whether to put 2 5'' woofers + 2 5'' PR in the back or 4 5'' woofers sealed. I've attempted a 4 woofer dual opposed build, but it failed because there's some sort of phase cancellation and I'm getting way less output than expected. I never figured out why there was a cancellation, but I suspect I need a divider to have only 2 woofers share an airspace to avoid cancellation. While 4 woofers will get me more overall output, I'm concerned about the 40-50Hz range most because that's arguably the most important range for music bass, and the PR system could be a lot cleaner in that range. 

If your speakers are all playing in phase, you won't have any cancellation. Here's what you have to remember.

 

For sealed systems, efficiency below Fb is solely determined by the compliance. In Ricci style builds, i.e. undersized boxes, the speaker compliance is negligible compared to the small box's compliance. Since there's no impedance peak below Fb, motor strength plays virtually no part.

 

Near Fb, sensitivity is still determined by the tiny box size, but impedance is much higher for more efficient woofers. Less power is drawn, so efficiency is greater.

 

Above Fb, the sensitivity is determined by the reference efficiency of the woofer. This is the spec published by manufacturers. Note the 98dB sensitive 21IPAL doesn't reach this sensitivity until 100Hz.

 

For sealed systems, Ricci's mantra is maximum full-bandwidth output from a small size, so he optimizes the limiting region below Fb. He picks woofers with enough thermal handling and excursion to get as much as possible down there. That's why he chose the horribly inefficient RE XXX's, because they can handle absurd power levels and have more displacement than just about any 18". However, I'd never say he's not concerned with efficiency, which is why he upgraded to the RF T3-19s. Their ridiculously powerful motors gave them more efficiency at and above Fb. Efficiency was only icing on the cake though. He only made the swap because he wouldn't be sacrificing any below Fb performance, slightly improving it in fact.

 

Small caveat, the "inductance hump" will warp the shape of the sensitivity graphs. However (and you'll find the same answer from simulations), efficiency is entirely unaffected. Inductance warps the impedance in exactly the same way. Any observed increases and decreases in sensitivity are accompanied by equal increases and decreases in impedance. This is why Le is not in the sensitivity and reference efficiency formulas.

  • Like 1
Link to comment
Share on other sites

LowerFe,

If you are building an ultra small bass module and trying to squeeze as much output as possible there are a few things to keep in mind in general. You have mentioned drivers in the 2.5-5" range. These are small drivers. Displacement is limited as is the power handling of the voice coil/s and the sensitivity. You also seem to be very limited on the size of the enclosure. Even to reach normal acceptable play back levels these tiny drivers will be being pushed quite hard, so it would be best to assume that they will be operating in the upper half of the performance envelope much of the time.

 

45Hz extension from a 14x6x6" enclosure means it will be very inefficient. Is this your desired enclosure size? You want useful output to at least 50Hz correct? What about the budget? Where would you like to be on cost? What will the amplification be?

Link to comment
Share on other sites

What does this mean for your enclosure?
 

Below Fb, you will see no gains in sensitivity. Adding more woofers will give you more displacement and power handling. This is where the increased maximum output comes from, and can only be achieved with the application of proportionally more power. Comparing Ricci's 21SW152-4 single driver to double driver systems, sensitivity is only increased above 60Hz.

 

Fb will go up in frequency. Adding more woofers gives each woofer less air space, so each one's air spring is stiffer, yielding a higher resonance. Notice Ricci's 21SW152-4 single driver system has an Fb ~50Hz, but the dual opposed is 65Hz despite adding a little more volume. This means you won't see any of the Fb efficiency gains until a much higher frequency. If you look at the Xmax Investigation thread, you'll see that Xmax is also increased at Fb. A driver can often remain relatively linear all the way up to it's mechanical limits. If your single driver system was getting an Xmax boost at 40Hz, your quad driver system might not get the boost until 70Hz. Ricci's 21SW152-4 data show only a 2dB gain in CEA-2010 burst performance around 50Hz going from one driver to two. Everywhere else, a 5dB gain can be seen. At Fb your maximum performance will increase by adding more drivers, but it will not be double.

 

Above Fb, you will see great gains in performance and exactly the kind of scaling you'd hoped for. Josh has said many times that this punchy midbass can be perceived as low sub-bass. In reality, it's just extra midbass. Without EQ an undersized system will be very punchy and top-heavy. After flattening the response, your sub-bass will no longer be overshadowed by the mids and you'll notice all the rumble your system is capable of. 

Link to comment
Share on other sites

And yes, the "Ricci style" build is gonna stuff a lot of woofers and a few hundred watts into a 14'' x 6'' x 6'' speaker. I'm still considering whether to put 2 5'' woofers + 2 5'' PR in the back or 4 5'' woofers sealed. I've attempted a 4 woofer dual opposed build, but it failed because there's some sort of phase cancellation and I'm getting way less output than expected. I never figured out why there was a cancellation, but I suspect I need a divider to have only 2 woofers share an airspace to avoid cancellation. While 4 woofers will get me more overall output, I'm concerned about the 40-50Hz range most because that's arguably the most important range for music bass, and the PR system could be a lot cleaner in that range.

 

Will you have electronic DSP to equalize this system?  IMO, this question is crucial to answer because if you can use DSP you can get a lot more bass without having to give up a lot of high frequency sensitivity.  This approach is taken by a growing number of consumer devices that offer great bass extension in very small packages.  Of course, bass output is ultimately compromised, but that's just life with small drivers and small enclosures.

 

An older example of this kind of implementation which still lives in my bedroom is the Logitech Squeezebox Boom.  The Boom was designed as a somewhat oversized bedroom alarm clock / music player.  It relied on a stereo pair of 3" drivers in very small sealed enclosures powered by a total of 30W Class D, IIRC.  Because the bass output was very limited, the unit applied a bass boost filter with full gain with the master volume set low.  As the master volume was increased, the filter gain was lessened so that the unit could play louder without being overloaded by bass.  This design strategy worked very well because of differences in perceived tonal balance vs. SPL by boosting the bass the most at lower levels where it's needed most.  While it's hardly a subwoofer, it produces acceptable bass down to 40 Hz or so (maybe lower?) in a small room with corner loading for low-to-moderate listening levels.

 

Anyway, there are other designs out there that deliver 40 Hz extension in very small packages, but the vast majority of them use some sort of DSP to boost the bass and probably strategies similar to those used for the Squeezebox Boom to allow for clean playback at higher levels, albeit with diminished bass.

Link to comment
Share on other sites

Thanks for the responses everyone. 

 

If your speakers are all playing in phase, you won't have any cancellation. Here's what you have to remember.

 

For sealed systems, efficiency below Fb is solely determined by the compliance. In Ricci style builds, i.e. undersized boxes, the speaker compliance is negligible compared to the small box's compliance. Since there's no impedance peak below Fb, motor strength plays virtually no part.

 

Near Fb, sensitivity is still determined by the tiny box size, but impedance is much higher for more efficient woofers. Less power is drawn, so efficiency is greater.

 

Above Fb, the sensitivity is determined by the reference efficiency of the woofer. This is the spec published by manufacturers. Note the 98dB sensitive 21IPAL doesn't reach this sensitivity until 100Hz.

 

For sealed systems, Ricci's mantra is maximum full-bandwidth output from a small size, so he optimizes the limiting region below Fb. He picks woofers with enough thermal handling and excursion to get as much as possible down there. That's why he chose the horribly inefficient RE XXX's, because they can handle absurd power levels and have more displacement than just about any 18". However, I'd never say he's not concerned with efficiency, which is why he upgraded to the RF T3-19s. Their ridiculously powerful motors gave them more efficiency at and above Fb. Efficiency was only icing on the cake though. He only made the swap because he wouldn't be sacrificing any below Fb performance, slightly improving it in fact.

 

Small caveat, the "inductance hump" will warp the shape of the sensitivity graphs. However (and you'll find the same answer from simulations), efficiency is entirely unaffected. Inductance warps the impedance in exactly the same way. Any observed increases and decreases in sensitivity are accompanied by equal increases and decreases in impedance. This is why Le is not in the sensitivity and reference efficiency formulas.

 

That's some great information there. I need some time to digest this. I won't have time until a few more days where I can really think about this and ask questions. 

 

Are you sure a dual opposed configuration like this will not cause cancellations?

 

W    W

W    W

 

LowerFe,

If you are building an ultra small bass module and trying to squeeze as much output as possible there are a few things to keep in mind in general. You have mentioned drivers in the 2.5-5" range. These are small drivers. Displacement is limited as is the power handling of the voice coil/s and the sensitivity. You also seem to be very limited on the size of the enclosure. Even to reach normal acceptable play back levels these tiny drivers will be being pushed quite hard, so it would be best to assume that they will be operating in the upper half of the performance envelope much of the time.

 

45Hz extension from a 14x6x6" enclosure means it will be very inefficient. Is this your desired enclosure size? You want useful output to at least 50Hz correct? What about the budget? Where would you like to be on cost? What will the amplification be?

 

Josh, you're a bad influence! I'm strongly leaning towards putting four 5'' woofers in that box to get as much Sd and VC to absorb power. 

 

With the PR system I actually want 40Hz extension. With the four woofer sealed system I plan on achieving 30Hz extension. Yes, it is going to be horribly inefficient, but it will make bass (hopefully).

 

14x6x6'' is the desired enclosure size. I'm not too concerned about the cost, maybe around $500 for bass drivers per speaker. The budget is flexible (curiosity is an expensive thing). I will be using multiple little palm sized amps such as the SMSL SA-98. If that's not enough during testing, I always have a iNuke 6000DSP  :D

 

 

 

Will you have electronic DSP to equalize this system?  IMO, this question is crucial to answer because if you can use DSP you can get a lot more bass without having to give up a lot of high frequency sensitivity.  This approach is taken by a growing number of consumer devices that offer great bass extension in very small packages.  Of course, bass output is ultimately compromised, but that's just life with small drivers and small enclosures.

 

Yes I will be using the miniDSP 2x4 HD for the speakers. I agree, it is absolutely essential for speaker design. The box will also have a high end tweeter and midrange inside (eating up 1 L of precious internal volume), and I got those right and sounding really good already. I'm pretty much trying to satisfy my curiosity of seeing how good can a small speaker possibly sound and how much big (Hi-Fi) speaker performance I can possibly squeeze. I can even do the output limiting feature on many commercial speakers. The speaker will be EQ'd flat to ~30Hz and I'll use the compressor to limit distortion and protect the speaker. 

Link to comment
Share on other sites

 

Are you sure a dual opposed configuration like this will not cause cancellations?

 

W    W

W    W

Not at the size enclosure you're using, and the frequencies you intend to use them at. The distance between woofers is acoustically small. Even at 300Hz, the woofers are just about quarter wavelength of each other. The lower you go, especially down to 40 and 30Hz, the distance between sources, even wrapping around the cabinet, is negligible compared to the wavelength they produce. They will be pushing in phase unless you wired one wrong.

 

Again, each woofer will eat up a little internal volume. Your proposed enclosure is ABSOLUTELY TINY so even going neodymium might have a measurable effect on the interior dimensions. Still, maximum output will be increased for the reasons previously discussed.

Link to comment
Share on other sites

I've thought about doing something similar before except as a slightly bigger, active studio monitor type deal. Now I can experience it through your build. :D

 

As Contrasseur said you won't have any cancellation issues until up above 250Hz with those dimensions so if something was cancelling the bass it was likely a wiring issue or something else.  What frequency are you planning to cross the woofers to the mid range? Have you decided on a midrange and tweeter?

 

You will need dsp to increase the low end extension like SME mentioned. Really no way around it with a cab that tiny. On the other hand it might be a good opportunity to get your hands dirty with limiter and compressor use. 30Hz extension is probably a bridge too far for something so small.

Link to comment
Share on other sites

I've thought about doing something similar before except as a slightly bigger, active studio monitor type deal. Now I can experience it through your build. :D

 

As Contrasseur said you won't have any cancellation issues until up above 250Hz with those dimensions so if something was cancelling the bass it was likely a wiring issue or something else.  What frequency are you planning to cross the woofers to the mid range? Have you decided on a midrange and tweeter?

 

Can I hear what you had in mind for your active studio monitor? I'm surprised it is only a little bit bigger than mine. 

 

I'm currently using the Scanspeak D3004/6020 tweeter and the Scanspeak 10F/8424G midrange, and I plan on using these in the new speaker too because they just sound so darn good. With a special (read: extreme) design I can actually make it act like a point source speaker, and this combo is the second best sounding driver combo that can achieve this. (The best sounding combo involves upgrading to a $280 Beryllium version of that Scanspeak tweeter, and as a basshead I can't bring myself to spend that much on a tweeter)

 

I plan on using a 300Hz crossover, although I may raise it to 400Hz to relieve the small midrange a bit more. Say if there is no midrange, and I want to cross the woofers to the tweeter at 2000Hz, does that mean in order to avoid cancellation, the rear woofers have to be low passed at say 200Hz while only the front woofers get low passed at 2000Hz?

 

 

You will need dsp to increase the low end extension like SME mentioned. Really no way around it with a cab that tiny. On the other hand it might be a good opportunity to get your hands dirty with limiter and compressor use.

 

 

This is definitely not gonna be a passive build. I will be using miniDSP(s) as an active crossover for this speaker. I've been thinking about the compressor a lot, and I can't think of a nice way to do it with the miniDSP. The miniDSP only has a single band compressor, which will kill a lot of upper bass output with real content given the amount of boost I will be applying. While it is possible to make a DIY multiband compressor by buying multiple miniDSP's and using a mixer to combine the signal back into one, that's a lot of additional hardware. 

 

Are you aware of any DSP's that natively implements a multiband compressor, or another type of compressor?

 

 

30Hz extension is probably a bridge too far for something so small.

 

Challenge accepted!  :D

Link to comment
Share on other sites

The MiniDSP is a good start, but as you seem to have realized, a simple compressor is not going to give you much flexibility.  Unfortunately, I don't know of a good readily available DIY solution to this problem.  Such capability might be available in certain plate amps such as from SpeakerPower, but the advanced DSP capability may not be available to non-OEMs.

 

At the same time, maximum bass extension from tiny speakers is big part of the commercial market for Bluetooth speakers and the like.  The different implementations I've heard vary considerably, but all are ultimately limited by the displacement of the usually tiny drivers.  Ricci's point about 30 Hz isn't so much that you can't build a tiny speaker to play 30 Hz but that it probably won't be worth the effort because it won't play 30 Hz loud enough for you to appreciate it without overloading.

 

Anyone seen the JBL Charge 2?  I chose it for my mom for her birthday, splitting the price with siblings, because Mom wanted a Bluetooth speaker and it reviewed a lot better than most of the competition at the price point.  I didn't expect much, but admittedly I was quite disappointed by its sound.  I just had a good laugh when I looked it up just now and read the product description:

 

 

The new JBL Charge 2, a portable stereo speaker offering best-in-class sound with dual drivers and twin passive radiators that deliver deep, accurate bass so powerful that you can actually see them trembling – even as their sound hits your ears.

...

 

This is no joke, except that the word trembling may be an understatement.  The PRs on the sides are probably a bit under 2" or so and have a pretty insane amount of excursion.  The specs state a frequency response of 75Hz to 20kHz.  The low end figure is legit, but try playing pop music at moderate background level and those PRs start swinging like crazy.  A higher level or simply music that's heavier in the bass caused it to overload.  That wasn't the worst of it though.  Considering all the positive reviews on this thing ("it sounds like big speakers"), I figured it would at least have a balanced sound but it sounded very mid-bass boomy and rolled off at the top.  It's disturbing to contemplate what the average person regards as good sound these days.

Link to comment
Share on other sites

At the same time, maximum bass extension from tiny speakers is big part of the commercial market for Bluetooth speakers and the like.  The different implementations I've heard vary considerably, but all are ultimately limited by the displacement of the usually tiny drivers.  Ricci's point about 30 Hz isn't so much that you can't build a tiny speaker to play 30 Hz but that it probably won't be worth the effort because it won't play 30 Hz loud enough for you to appreciate it without overloading.

 

I did understand him. For one speaker, he has a point. However, with a pair (after all, this is a speaker), I'm looking at around ~95dB @ 1 meter at xmax. With some room gain, which is not insignificant since this is being used in a small room like a bedroom or hotel room, the output is definitely usable. Even at 90dB the additional depth gives a more fullness to the sound even if 30Hz is not the fundamental note. You guys and your 8x18'' subs make you forget what it's like to listen at normal volumes, you know, like at weak sauce level of 70-80dB?  :P

 

The Sonos Play 1 uses a single 3.5'' woofer to achieve a F3 of 40Hz (I personally measured this outdoors). The Play 5 Gen 2 uses triple 4'' woofers and achieves a 30Hz extension with a F3 of 27Hz. I'm using quad 5'' woofers, by Sonos standards I should really be going for 20Hz extension!  :P

 

 

Anyone seen the JBL Charge 2?  I chose it for my mom for her birthday, splitting the price with siblings, because Mom wanted a Bluetooth speaker and it reviewed a lot better than most of the competition at the price point.  I didn't expect much, but admittedly I was quite disappointed by its sound.  I just had a good laugh when I looked it up just now and read the product description:

 

 

This is no joke, except that the word trembling may be an understatement.  The PRs on the sides are probably a bit under 2" or so and have a pretty insane amount of excursion.  The specs state a frequency response of 75Hz to 20kHz.  The low end figure is legit, but try playing pop music at moderate background level and those PRs start swinging like crazy.  A higher level or simply music that's heavier in the bass caused it to overload.  That wasn't the worst of it though.  Considering all the positive reviews on this thing ("it sounds like big speakers"), I figured it would at least have a balanced sound but it sounded very mid-bass boomy and rolled off at the top.  It's disturbing to contemplate what the average person regards as good sound these days.

 

While I've not heard of the JBL Charge 2, I definitely feel what you're talking about. I tried many small wireless/Bluetooth speakers before building my own. While I was generally very disappointed by the sound quality of most (which is what drove me to build this), the recent Sonos speakers stood out as being head and shoulders better than anything I've tried and actually sound comparable to real speakers. Before I had a tough time answering the "what speaker should I buy" question often asked by my friends because all the small lifestyle speakers sounds terrible compared to a bookshelf + receiver. Now, the answer is unequivocally a Sonos Play 1 or 5 (not the other models, those are still old model where Sonos wasn't crazy about sound quality like they are now).

 

In fact, if I heard a Sonos during my initial search for a small speaker, I don't think I would have attempted this project. In fact, when I heard the Play 5 and studied its design, I almost quit because it was so well designed and sounded so good I didn't know how I could DIY something better as a whole package. A big problem back then was that I could not use the compressor to limit bass without limiting the lower midrange that the woofer also covers, which essentially meant I could not use the compressor and the speaker could not be safely turned up beyond a bit above average volumes, and that was a big and annoying limitation that the Sonos didn't have. Fortunately, someone showed me a light at the end of a tunnel by showing me a way to build a DIY multiband compressor from multiple single band compressors, and that renewed the fire in me to see through this project to the end. 

Link to comment
Share on other sites

Can I hear what you had in mind for your active studio monitor? I'm surprised it is only a little bit bigger than mine. 

 

This is definitely not gonna be a passive build. I will be using miniDSP(s) as an active crossover for this speaker. I've been thinking about the compressor a lot, and I can't think of a nice way to do it with the miniDSP. The miniDSP only has a single band compressor, which will kill a lot of upper bass output with real content given the amount of boost I will be applying. While it is possible to make a DIY multiband compressor by buying multiple miniDSP's and using a mixer to combine the signal back into one, that's a lot of additional hardware. 

 

Are you aware of any DSP's that natively implements a multiband compressor, or another type of compressor?

 

 

 

Challenge accepted!  :D

 

I didn't have anything set in stone for active monitors. Mostly some general ideas and goals. They would've been a bit larger than yours. Plate amps would've been built in etc. As far as compressors a lot of the pro audio stuff has advanced limiting features.

 

Glad to see you started a dedicated thread.

Link to comment
Share on other sites

The DIY multiband compressor sounds like a great idea.  I bet a lot of people could benefit from this sort of thing, not just with miniature designs like what is being suggested here but also in sub-systems using Linkwitz Transform to bring up ULF or a variety of situations involving a low Qtc alignments for that matter.

Link to comment
Share on other sites

  • 3 years later...

Hi....the box is the Hegelian antithesis of the driver at resonance. So equal bumps is natural and means at the box tuning is right on the money. But the output is not "symmetrical" in that the basis falls away fast below the resonance and you (and lots of manufacturers) may not care for that result.
But what sounds bests at your seat is a complicated mix of driver, box, room, and your taste in bass sound since the odd reinforcements and cuts the box produces in the bass band is its own colouration.Something similar could be said about all driver housings to some degree but the BR box is complicated and has a tuning.

order pcb with components

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.

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.

Loading...
×
×
  • Create New...