Site Links
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
Categories
shrinking sealed box experiment
Thanks to everyone’s ‘encouragement’ (how many drivers do you have) I’m feeling the need to step it up a bit… Actually, I have been thinking of ordering a couple of the DAudio RSS210HO-4 ohms and play around with a smallish sealed enclosure. I actively bi-amp between woofer and mid/tw via miniDSP, and employ passive for mid/tw.
While in the past I’ve combined the mid/tw (or broad range Audio Nirvana) with; a bass horn, an open back (pair of 12in Beymas) and most recently a woofer/ PR design, I’ve been hankering to try a sealed set-up, and focus more on bass quality and signal integrity; prioritizing the subjective, the bass extension in-room -6dB results (over the -3dB) and leaving the bottom octave, if necessary, to a sub. And, via a sealed enclosure skip the constructive and deconstructive phase interference of ports and PRs, their resulting steep drop-offs and distortions. To date my fav low ends have been via TL’s and eq’d open backs. Can’t say I have tried a sealed woofer design, but I tend to favor sealed subs vs their ported brethren.
I’ve found the Meadowlark technical discussions (as posted in another thread) in combo with Vance Dickason’s conclusions / generalizations around QtcS of sealed enclosures and their subjective sound quality (Loudspeaker Design Cookbook, 5th 1.30 Driver Q and enclosure response pg 12) of interest, and worth playing around with.
Sealed box calculators (that I’ve seen) assume a passive and non- aperiodic design. I want to take a look at using as additional tools mild active eq with or w/o aperiodic venting, and target controlled bass in a small(ish) design.
Interestingly, looking at ‘results’ from online calculators, VituixCad, and the PE recommended sealed box volume for this driver, I see a range from 0.4 to~ 0.7 cu ft. These results assume passive and non-aperiodic systems. I’m thinking of starting with the larger ‘calculated’ enclosure and moving smaller, looking at the net effects of judicially implementing additional tools (active eq and aperiodic venting).
Any guidance / comments you guys may have in designing / sizing sealed woofer designs, objective and subjective results, would be of interest- thanks
Comments
Q>=1.0 has always been my favorite sealed bass, provided it can deliver solid output tickling 50Hz.
A lot of builders that favor sealed boxes like or prefer the critically damped alignment with Qtc of 0.5, but of note is that you cannot achieve a Qtc lower than the Qts of the driver itself, nor can you really get close to Qts value. A little bump in the bass with a Qtc of 0.8-0.9 is not really bad sounding subjectively, provided low cut off. Qtc of 0.7 is seen as the lowest F3 possible sealed without help or EQ, and is seen as (IIRC) transient perfect. Qtc around 0.6 is also accepted by those wanting the lowest Q (0.5) possible. Above 1.0 tends to be not damped enough and will sound like vintage speaker warmth.
Aperiodic is modeled with Ql of 3.0, and a low alpha or highly stuffed.
Aperiodic multichamber, 2 sealed volumes with an aperiodic vent between or sometimes known as MAPD, usually ends up with volumes from 1:1 ratio to 1:3 ratio. Stuff the front vol, line the rear vol, and measure results. It is fairly easy to reduce impedance at Fs this way, and yields very good subjective bass or midrange results.
InDIYana Event Website
Hey, it's Mr. "I have too many drivers" here. I have a pair of RSS265HO-4 that I would be happy to shed at my usual low prices!
From my basement to yours! No seriously I do have a pair of these drivers. Drop me a PM if you want them for your project. The HO is great in small enclosures. The 10" RSS will be much better for bass than the 8".
Mr. Charlie "I have too many drivers"
Thanks, but the 8's are pushing my size target as is.
w/o eq, I’d lean towards Qtc btwn 0.5 to 0.7.
What I’m trying to get my head around is this: with eq I can pull down a hump in it’s smaller box with a higher Qtc (say towards 1.0). But, by doing so (smaller box with negative dB eq) am I limiting the woofer in a way that reduces quality of reproduction?
Or, if I go with a larger box to target 0.5 to 0.7 is the driver more prone to not being controlled (quickness ?) relative to its signal input?
I’d rather have a woofer that is well controlled and in line with the signal and then judiciously eq to supplement its output (ie +1 to 2 dB) to a rolloff target (ie between a sealed and a ported design, maybe 9dB?) than to have a woofer that is less controlled (ie not tight?). Hope that makes some sense- I’m looking for a best quality at sacrifice of lowest response biz I can use a bit of eq to push the -6dB level (and hence -3dB) up.
Or put a bit differently, I can pull down a slight hump resulting from a smaller box (with eq) , but does this sacrifice quality?
Thx
I would concern myself with how it sounds which means you have try a prototype box and then EQ/Shape it.
I agree on the 0.5 ~ 0.7 Qtc but sometimes a song weak in the bass dept gets a Qtc of 0.8 from the DSP to make it sound "right".
But then for almost all music a 0.7 is right on the money.
If you notch out a peak, it will increase overhead on the amp.
Passively = yes, DSP = No.
Yes?
DSP yes. It is a well known trick to increase amp overhead.
I see where I screwed up; "Overhead" = additional power AVAILABLE HEADROOM, not additional load.
"Overhead" implies additional load/cost/energy waste.
Got it.
I think the aperiodic vent is a pretty good idea to get a less resonant sounding bass in a smaller box. Hopefully you have a DATS or some other method of measuring impedance curves. The basic recipe is this:
Now you are back to one peak like a sealed box, but it is acoustically damped by the resistive port and has a much lower amplitude. In theory, the aperiodic enclosure should have better bass control than either a ported or sealed box. In practice, let your ears be the judge.
Bookmarked for Billet's Recipe!
So, doing some math, I’m looking to build a test box with a width of 10 inches and a depth of 8.5 inches (what I’d like to ~end up with for my project for the woofer section) and start with a height of approx. 21 inches (using a removable end-cap bottom plate). Then after measurements (remove the end-cap and) systematically ‘whittle’ the box down, with a little help from my table saw.
An online calculator for a QTC .707 requires a Vb of 0.5 ft^3, per the driver’s params. Parts Express ‘recommended’ sealed enclosure volume is listed at 0.4ft^3. To get a broader range for testing, using my abacus provides the following; Vb 0.65 ft^3 requires height of 21.125 inches, Vb 0.50 ft^3 (with a QTC .707) 17 inches, Vb 0.45 ft^3 15.625 inches, and Vb 0.40 14.25 inches (these are net of est displacements of driver and cross supports). While I’m not certain what I’ll end up finding, from the outset I’d like to find a Vb that creates a minimal peak in a completely sealed enclosure, then with an aperiodic vent potentially addressing a portion of the peak. And finally, gingerly EQ-ing to extend and shape (level out the balance of any peak) the low end targeting the -6dB level, while maintaining a slow-ish roll-off (somewhat closer to the attributes of a sealed, than per the steeper ported or PR designs). If you see an error in my math, plz let me know.
Now if I can just get lucky with the FedEx delivery when the better half is out, and do this build with NOS ‘basement’ parts- ha.
This ^ is going to be interesting and I like your plan, Sir.
Thx UPS, timely delivery of my drivers and a few parts…
As per the above, I will be adding a test Vb of .35, to give +/- 30% volume from the targeted .707 QTC of .5 ft^3. So a range of +30% .65 ft^3, .50 ft^3 tgt Vb, .45 ft^3 down 10%, .40 ft^3 down 20%, and .35 ft^3 down 30%.
I will be looking at frequency measurements with and w/o aperiodic venting, with and w/o eq, and subjective listening. While the above suggestion of designing with a port and monitoring impedance to a single peak has merit, knowing that an AVent will bring down impedance, which may, among other things, be beneficial to the amp > I’m running with a Legacy Powerbloc 4 that has 650 w @ 4ohms / 325 @ 8ohms per channel, so impedance will be secondary (although thinking rhyming with) to frequency.
I’m thinking best to measure driver close range, maybe very close to driver (maybe just beyond Xmax) and/or 4 inches to focus on the box size (and sealed vs. AVented) affect on the driver’s frequency output / distortion and exclude baffle / room affects (which will come into play with final speaker and DSP settings).
Per the measure twice and cut once philosophy, any suggestions on measuring distance would be appreciated.
Love your plan!
Use a tape measure for distance - 4" sounds good to me and less trouble to set the mic-up each time without accidently poking a hole in the driver . . .
The three most useful measurements are
1) anechoic
2) full space
3) in-room response (not covered here)
Anechoic gives you the measurement of the driver’s response only, without the effects from the cabinet shape.
Full space shows measurement of the loudspeaker system including the cabinet it’s in.
In-room response shows what’s happening at the listening position due to reflections due from the ground, walls and furnishings etc.
1) measure with the mic (as close as possible) to the centre point of the driver, with the mic as close as possible eg. In millimetres from the cone. Be careful not hit the mic/cone, or overload the mic. So a good distance is 5mm or 1/4” from cone; using drive level of 0.283V (-20dB from 1W).
Note that the accuracy of this measurements is good from 10 Hz, but the upper soft limit is around 4311/(diameter of cone). So for a 12” woofer with an effective cone diameter of 10”; then the upper limit is 4311/10= 431Hz. At higher frequencies than this the measurement loses accuracy.
2) Free space. Make sure your mic is back 3 times the diameter of the cone. So if you’re measuring a 6.5” midwoofer, which has an effective cone diameter of 5”; then measuring at 18” is just enough to get the far field measurements. Moreover, if you you measure back at 3 times the longest dimension of the cabinet ie. Diagonal of the longest height/width/depth; then you will pickup the measurements complete with the effects of the cabinet shape.
The downside of this is that you are further away, and you’ll get reflections from the ground that will contaminate your frequency response. To remove these reflections, you need to gate your measurements.
References and further reading:
How to Achieve Accurate In-Room Quasi-Anechoic Free-Field Frequency Response Measurements Down to 10 Hz. By Jeff Bagby
http://audio.claub.net/software/FRD_Blender/White Paper - Accurate In-Room Frequency Response to 10Hz.pdf
Application Note: Loudspeaker Electroacoustic Measurements by Audio Precision:
https://drive.google.com/file/d/1ICLgcjqGtyvNaSva4jPQ8Z3vYwFN9_Ex/view?usp=share_link
TBC…
Thanks for this info, and the links.
For this experiment, it's primarily all about anechoic to get the impact of differing sealed Vb and 'tools' of employing aperiodic venting, and then a bit of eq. Objective measurement at the cone (per your input distance and noted upper frequency limits- thx), and subjective listening. I remember a now very old Speaker Builder Magazine article (split over two issues) where a ~15 or 18in woofer was housed in a very tight box (top was a horn if I remember) and managed via eq. circuit (but not in combo with an aperiodic vent). I've looked in the archives but no luck in finding. If anyone has an old copy- plz let me know.
Less so the impact of the cab shape (the length/height of which will vary), but I will additionally measure at 3x effective cone diameter (~ 7in, so 21-24inches) so I'll take a few measurements there as well to see if that adds any info.
I may have a copy of that Speaker Builder article. I'll dig through my back issues and see.
Sorry, I went through all my SB back issues from 1980 to 1996 and could not find the article that you are looking for. I also checked my SB 2000 CD disc. I am missing SB back issues for 1997 to 1999, so perhaps it was in one of those years. After 2000, Ed Dell merged Speaker Builder, Audio Amateur, and Glass Audio into a single magazine called Audioxpress.
Thanks for looking. I had kept many back issues over the years, until I moved- the box must have been trashed. If I remember the front cover was yellowish / light tan and the speaker was on the cover. So, it is as you have narrowed down, probably 97-99.
Thanks again.
If you can come up with the specific issue you're looking for, your local library may be able to get you a copy. That's how I got a copy of an article on doppler distortion from a December1977 Audio magazine in 2009. If you google
Speaker Builder magazine back issues and look at images you may be able to find the front cover you remember.
Ron
First time working with MDF, cuts nice and glues up square, but routing is a sport best done outside..
(I’ll be using a cap for the vent to get sealed measurements)
I finished the box, the one end cap comes off to uncover the vent for comps. Something just seems wrong with taking the box to the table saw-ha.
I started taking a few measurements. So far only at 0.6Vb and 0.5Vb (ft^3). I’ll post measurements when I get the Vb below the 0.5ft^3 (0.707 QTC) level.
The top, from an old audio nirvana build
This is a great project. I did something similar a number of years ago but playing with vented alignments. Just did it to kill time, I think - nothing as empirical as what you are doing.
If I can make a comment about the SS aperiodic vents, mine buzzed under operation. I found that regular Oatey plastic floor drain ends stuffed with thick open cell foam work better for less money. YMMV, my 2c.
InDIYana Event Website
Thx, Yes. I had these from an old horn bass project, and would probably go with your idea, and/or in combo with a slot. As per my following post, the vent needs, and can be, tuned.
Tests parameters:
At .707 QTC Vb 0.50ft^3. I ran sweeps at 0.60, 0.50, 0.40 (PE’s recommended sealed Vb), and 0.35ft^3.
I compared sealed, aperiodic vented, and at and below 0.40ft^3 applied eq (relative to the 0.60ft^3 being used as a target). Measured with subjective listening. Application in a 3-way, or two-way where LP on the woofer at or below 400Hz.
A few observations from my measurements:
As would be expected, as Vb reduced the sealed low end became truncated.
At 0.40ft^3 a rise/ bump in frequency started becoming apparent. Chalk one up the online Vb calculators targeting 0.50ft^3 vs PE’s recommended enclosure 0.40ft^3 (perhaps fine for a sub / auto- target audience?).
Aperiodic vents do look to provide a bit of relief from the bump (constraining Vb impacting frequency). They also exhibit greater roll-off, thinking as a result of the vented sound propagation. I had used an additional layer of foam in addition to the Scan-Speak vent, and in a final application I’d play around with finding an optimum level of venting, and absorbing/attenuating higher frequencies. Like EQ (and TL stuffing) it’s a tool that can be overdone.
At the smallest Vb in the test ( 0.35ft^3), looking to meet or exceed the result from the .6ft^3 sealed box (the initial target), along with the vent I applied the following eq: 100Hz / -1.0 gain / Q1.2, 50Hz / 0.6 gain / Q 0.8, and 35Hz 2.0 gain / Q 0.8. Probably far from the optimum adjustments, but the smaller box provided better measured results. And subjectively, I didn’t sense any limitations on the sound quality. I initially was thinking I’d hear some impact from the ‘tuned’ smaller box, but at least in this low-end frequency range I didn’t. Up into the higher ear-sensitivity ranges 1,000 to 4,000Hz enclosures can be problematic.
Best, IMO, for 3ways where LP expected to be below 400Hz. My preference is for mids to be open or vented (i.e. to reduce back waves) but for woofers small sealed enclosures with limited aperiodic venting and judicious EQ, works nicely with a focus /preference on slower and extended roll-offs vs -3dB cliffs.
Additional playing around I’ll be looking at further shaping with eq (there is ample room yet with these drivers), as I compare the -6dB level of my current 3-ways (using the 8ohm version of this driver in combo with a PR). While they may end up being a bit shy as compared to the PR design at the -6dB level, I’m thinking the absence of the -3dB cliff (which often seems to draw attention to a narrow bass note) in combo with a smother in-room roll-off may be preferable.
I’ve attached a couple of graphs (and a couple more pics). One showing sealed enclosures over the Vb range, and the other comparing the 0.60ft^3 sealed ‘target’ vs the 0.35ft^3 ‘tuned’ little box.
Nice results! Based solely on the graphs, I like the 0.35 cu ft with the boost from 50 to 120 hertz. Overall, there appears to be a surprisingly small difference between the 0.35 and the 0.60 enclosures.