George Augspurger Double-Chamber
The Speaker Enclosure;" George Augspurger; Electronics World
I have a translation in Spanish in my Web.
http://www.geocities.com/dbr_audio/
regards Jaime
The Speaker Enclosure;" George Augspurger; Electronics World
I have a translation in Spanish in my Web.
http://www.geocities.com/dbr_audio/
regards Jaime
Back in 1984 I read David Weems' book available through Radio Shack not just once but many times. Studied those projects w/ported closures this type in particular. Using some upscale car audio drivers, for fun, I built a pair of 5 1/4" 2 way speakers. One sealed & one using this closure design some of you are wrestling with. If you read ALL the way to the bottom, you'll learn the little bit of extra work is worth it. Below in kelticwizard's example is how David Weems described the first stage of the design process. I love KW's illustration in post #11.
Some of you are overthinking this simple process.
1. Plug in your driver's TS data into WINISD as a bass reflex
2. Use WINISD's suggested** port*** dimensions*** (PD) & box volume (BV) allowing*add'l space for:
a. the divider plus any add'l bracing you feel necessary
b. 3 ports*** instead of one
** sized for no chuffing of course
Yup. Plan a little/lot bigger box, maybe needing more bracing but that's it. Then build it. If it sounds sloppy or thin, either wrong type driver (Oops!) or the TS data is a little off which means you'll hafta tweak BVS & the PDS. I purposely typed those upper case bolded "Ss" to stress both BVs & ALL three PDs. Argh!!!
The easy way around that nightmare. Build it still allowing space for the "extras" but install only one port. Put the extras inside during the tuning stage. When you're satisfied you have the PD/BV relationship right, install the extras, make it pretty for you're other half & ENJOY!
Way back then I did not have the TS data, WINISD or any real test equipment to help w/the hard stuff so I winged it!
Took me 2 tries to get the sealed BV right (about 7" X 1' X 2.5') W/a stout 75/WPC it sounded pretty darn good in my sub compact.
Then I built the ported closure allowing about 30% add'l net BV; PD ~4"(L) X 2". While they were ugly, they thumped! I questioned my need for a sub.
Listened to a lot all kinds of music during my commutes to work w/discounted access to high-end home & car stereo equipment. Stayed in touch w/my old job w/professional table & radial arm saws and the like.
Before I retired those drivers to my car doors, I did conduct limited testing & listening in a small room (~9' X 11' w/8' ceiling) using an Audiocontrol Ricterscale (RS), turntable fitted w/a decent MC cartridge playing a variety, and other home gear of higher quality & power* than the best available car stuff at the time.
* 220/WPC
Early listening revealed they had the same FR above 100hz. Below that remarkably similar to maybe an octave w/out "help." Did not notice a diff below cutoff to the tweeter. Would play quite loud, cleanly most anything.
Later I was ready to learn their limits so I applied some "help" below 70hz to that end via the RS, first in the form of EQ. Got them flat to 45hz*. Could have taken that further, but just getting there* took +3dB@60hz, +7dB@45hz & this wasn't a bass/dB contest. Then routed the signal through the RS's EOX so I could more easily evaluate them below 110hz.
Together & separately I tortured them w/tracks off the Telac label. Little by little w/the V knob and many balance knob twists later until they each in turn said enough! Carefully selected passages from Bach's Toccata & Fugue, Frederick Fenel, and the abuse of abuse, Tchaikovsky's 1812 overture w/digital cannons! In these tests their tonal quality were VERY similar. Ported would play slightly louder than the sealed.
Ok Tony but what about F3, output, power handling, etc? The ported could handle, though only a little, more than the sealed!
I've read this entire thread three times plus selected parts 2 or 3 times more.
Some of you are overthinking this simple process.
1. Plug in your driver's TS data into WINISD as a bass reflex
2. Use WINISD's suggested** port*** dimensions*** (PD) & box volume (BV) allowing*add'l space for:
a. the divider plus any add'l bracing you feel necessary
b. 3 ports*** instead of one
** sized for no chuffing of course
Yup. Plan a little/lot bigger box, maybe needing more bracing but that's it. Then build it. If it sounds sloppy or thin, either wrong type driver (Oops!) or the TS data is a little off which means you'll hafta tweak BVS & the PDS. I purposely typed those upper case bolded "Ss" to stress both BVs & ALL three PDs. Argh!!!
The easy way around that nightmare. Build it still allowing space for the "extras" but install only one port. Put the extras inside during the tuning stage. When you're satisfied you have the PD/BV relationship right, install the extras, make it pretty for you're other half & ENJOY!
Yup! Don't over think it guys. I've played w/the idea of "creative tuning" but can't get past the "it ain't broke don't fix it" mantra. If you can wait, I plan to retire in 5 years, I'll get back you then! Might take it to the next level w/a triple chambered closure w/5 ports!
But as KW also said and I've alluded to the same, they're NOT separate tunings so don't get your knickers in a knot over that. 3 ports all the same size. Two chambers, 1 twice the net size of the other.
Way back then I did not have the TS data, WINISD or any real test equipment to help w/the hard stuff so I winged it!
Took me 2 tries to get the sealed BV right (about 7" X 1' X 2.5') W/a stout 75/WPC it sounded pretty darn good in my sub compact.
Then I built the ported closure allowing about 30% add'l net BV; PD ~4"(L) X 2". While they were ugly, they thumped! I questioned my need for a sub.
Listened to a lot all kinds of music during my commutes to work w/discounted access to high-end home & car stereo equipment. Stayed in touch w/my old job w/professional table & radial arm saws and the like.
Before I retired those drivers to my car doors, I did conduct limited testing & listening in a small room (~9' X 11' w/8' ceiling) using an Audiocontrol Ricterscale (RS), turntable fitted w/a decent MC cartridge playing a variety, and other home gear of higher quality & power* than the best available car stuff at the time.
* 220/WPC
Early listening revealed they had the same FR above 100hz. Below that remarkably similar to maybe an octave w/out "help." Did not notice a diff below cutoff to the tweeter. Would play quite loud, cleanly most anything.
Later I was ready to learn their limits so I applied some "help" below 70hz to that end via the RS, first in the form of EQ. Got them flat to 45hz*. Could have taken that further, but just getting there* took +3dB@60hz, +7dB@45hz & this wasn't a bass/dB contest. Then routed the signal through the RS's EOX so I could more easily evaluate them below 110hz.
Together & separately I tortured them w/tracks off the Telac label. Little by little w/the V knob and many balance knob twists later until they each in turn said enough! Carefully selected passages from Bach's Toccata & Fugue, Frederick Fenel, and the abuse of abuse, Tchaikovsky's 1812 overture w/digital cannons! In these tests their tonal quality were VERY similar. Ported would play slightly louder than the sealed.
Ok Tony but what about F3, output, power handling, etc? The ported could handle, though only a little, more than the sealed!
I did that already...
3CR Photos by wolf_teeth_speaker | Photobucket
...and this site models these DCR/3CR very well:
Advanced Speaker Box Calculator
Later,
Wolf
Greets!
FYI, Claudio's investigations [dead] linked in post #39 that folks used to design these till recently is now here: Claudio Negro's home page, how to project and test a loudspeaker using Speaker Workshop
GM
I may build, instead of buy, my next sub because of the revival of this thread. But first... ...how does one model anything but a sealed, vented or a multi chamber BP closure on Advanced Speaker Box Calculator? I love it because it's runs on my iPad and the graphs.
But when switched to the multi band pass page, regardless of what I plugged in, I got the same results - results I couldn't begin to interpret. I first modeled both drivers'* on the other page**. Then I took the volume, plugged it in at 2:1 ratio for Vr1 & Vr2 and the port dimensions for 2, 3, & 5. I made Vf1 ginormous & the other port diameters .0001 nanometers. Later I even tried going wacko w/all kinda different volumes & I may build, instead of buy, my next sub because of the revival of this thread. But first... ...how does one model anything but a sealed, vented or a multi chamber BP closure on Advanced Speaker Box Calculator? I love it because it's runs on my iPad and the graphs.
But when switched to the multi band pass page, regardless of what I plugged in, I got the same results - results I couldn't begin to interpret. I first modeled both drivers'* on the other page**. Then I took the volume, plugged it in at 2:1 ratio for Vr1 & Vr2 and the port dimensions for 2, 3, & 5. I made Vf1 ginormous & the other port diameters .0001 nanometers. Later I even tried going wacko w/all sorts of volumes & port dimensions... ...always the same graphs.
* see below
I had hoped I could model my "creative tuning" idea that I've long considered but doubted that it would be easy to accomplish. That said, after basic modeling & tuning, I've wondered if acoustic equalization instead of electronic were possible by...
1. the smaller chamber uses a passive radiator - practical but not innovative.
2. the larger chamber - one (or both) port dimensions purposely longer than modeled to reduce output in the upper portion of its range. Regardless...
I think either of these two Dayton Audio drivers from Parts Express could get down to the mid 20s in 3-3.5 cu ft w/little narrow boost near the bottom but I'm open to suggestion.
Dayton's popular Reference 12" sub. Description states F3 of 20hz in a vented 4.6 cu ft closure - kinda big. Model it smaller w/boost per above
12" part # 295-445
Resonant Frequency (Fs). 25 Hz
DC Resistance (Re)6.45 ohms
Voice Coil Inductance (Le) 1.6 mH
Mechanical Q (Qms) 2.94
Electromagnetic Q (Qes) 0.64
Total Q (Qts) 0.53
Compliance Equivalent Volume (Vas) 2.67 ft.³
Mechanical Compliance of Suspension (Cms) 0.21 mm/N
BL Product (BL)17.5 Tm
Diaphragm Mass Inc. Airload (Mms)195.2g
Maximum Linear Excursion (Xmax) 14 mm
Surface Area of Cone (Sd) 506
Dayton's new "ultimate" 10 inch sub. They claim F3 of 22hz in a vented 1.55 cu ft closure? I doubt it, though "doubling the recipe" (twin drivers) in series & some boost per above, maybe
Part# 295-510 Impedance 4 ohms
Sensitivity 83.4 dB 1W/1m
Voice Coil Diameter 2.5"
TS data
Resonant Frequency (Fs) 26.2 Hz
DC Resistance (Re) 3.3 ohms
Voice Coil Inductance (Le) 1.27 mH
Mechanical Q (Qms) 3.03
Electromagnetic Q (Qes) 0.54
Total Q (Qts) 0.46
Compliance Equivalent Volume (Vas) 1.24 ft.³
Mechanical Compliance of Suspension (Cms)0.23 mm/N
BL Product (BL)12.7 Tm
Diaphragm Mass Inc. Airload (Mms)160.4g
Maximum Linear Excursion (Xmax)19 mm
Surface Area of Cone (Sd) 327.8 cm²
Sorry for long-winded ness. Thnx for reading this far & anyone's help & ideas.
But when switched to the multi band pass page, regardless of what I plugged in, I got the same results - results I couldn't begin to interpret. I first modeled both drivers'* on the other page**. Then I took the volume, plugged it in at 2:1 ratio for Vr1 & Vr2 and the port dimensions for 2, 3, & 5. I made Vf1 ginormous & the other port diameters .0001 nanometers. Later I even tried going wacko w/all kinda different volumes & I may build, instead of buy, my next sub because of the revival of this thread. But first... ...how does one model anything but a sealed, vented or a multi chamber BP closure on Advanced Speaker Box Calculator? I love it because it's runs on my iPad and the graphs.
But when switched to the multi band pass page, regardless of what I plugged in, I got the same results - results I couldn't begin to interpret. I first modeled both drivers'* on the other page**. Then I took the volume, plugged it in at 2:1 ratio for Vr1 & Vr2 and the port dimensions for 2, 3, & 5. I made Vf1 ginormous & the other port diameters .0001 nanometers. Later I even tried going wacko w/all sorts of volumes & port dimensions... ...always the same graphs.
* see below
I had hoped I could model my "creative tuning" idea that I've long considered but doubted that it would be easy to accomplish. That said, after basic modeling & tuning, I've wondered if acoustic equalization instead of electronic were possible by...
1. the smaller chamber uses a passive radiator - practical but not innovative.
2. the larger chamber - one (or both) port dimensions purposely longer than modeled to reduce output in the upper portion of its range. Regardless...
I think either of these two Dayton Audio drivers from Parts Express could get down to the mid 20s in 3-3.5 cu ft w/little narrow boost near the bottom but I'm open to suggestion.
Dayton's popular Reference 12" sub. Description states F3 of 20hz in a vented 4.6 cu ft closure - kinda big. Model it smaller w/boost per above
12" part # 295-445
Resonant Frequency (Fs). 25 Hz
DC Resistance (Re)6.45 ohms
Voice Coil Inductance (Le) 1.6 mH
Mechanical Q (Qms) 2.94
Electromagnetic Q (Qes) 0.64
Total Q (Qts) 0.53
Compliance Equivalent Volume (Vas) 2.67 ft.³
Mechanical Compliance of Suspension (Cms) 0.21 mm/N
BL Product (BL)17.5 Tm
Diaphragm Mass Inc. Airload (Mms)195.2g
Maximum Linear Excursion (Xmax) 14 mm
Surface Area of Cone (Sd) 506
Dayton's new "ultimate" 10 inch sub. They claim F3 of 22hz in a vented 1.55 cu ft closure? I doubt it, though "doubling the recipe" (twin drivers) in series & some boost per above, maybe
Part# 295-510 Impedance 4 ohms
Sensitivity 83.4 dB 1W/1m
Voice Coil Diameter 2.5"
TS data
Resonant Frequency (Fs) 26.2 Hz
DC Resistance (Re) 3.3 ohms
Voice Coil Inductance (Le) 1.27 mH
Mechanical Q (Qms) 3.03
Electromagnetic Q (Qes) 0.54
Total Q (Qts) 0.46
Compliance Equivalent Volume (Vas) 1.24 ft.³
Mechanical Compliance of Suspension (Cms)0.23 mm/N
BL Product (BL)12.7 Tm
Diaphragm Mass Inc. Airload (Mms)160.4g
Maximum Linear Excursion (Xmax)19 mm
Surface Area of Cone (Sd) 327.8 cm²
Sorry for long-winded ness. Thnx for reading this far & anyone's help & ideas.
It has to do with how you manipulate the volumes. I believe it states how to fill the cells, like no driver- use a zero. That kind of thing. Immense volume would be into a room as opposed to sealed off at zero.
Make sense?
Wolf
The question is OT 'cos the loudspeaker is supposed to radiate freely into the ambient, not into another (tuned) box.
hi gp4Jesus
A DCR is basically a bass reflex. In the measurements below, the Red plot is nearfield. The Green is the port tuning of the main chamber. The Violet is what separates a DCR different from a BR. This response is taken from the port of the second chamber. Notice it is a bandpass. Port responses normalized to nearfield.
A DCR is basically a bass reflex. In the measurements below, the Red plot is nearfield. The Green is the port tuning of the main chamber. The Violet is what separates a DCR different from a BR. This response is taken from the port of the second chamber. Notice it is a bandpass. Port responses normalized to nearfield.
Attachments
I missed something
I tried about 100 million liters in the Vf1 cell - about twice that of my family room.
Not enough sense to power the 1/2 watt LED bulb in my feeble brain.
Forgive me for being verbose:
I modeled the drivers mentioned in my previous post on this page...
Loudspeaker enclosure calculating with Thiele Small parameter.
Then I plugged in that volume & port info into the page below...
Advanced Speaker Box Calculator
...only to see identical graphs for both drivers
Elsewhere on the site I saw something to the effect: "...making port diameters VERY small (.0000000001 cm) will make the calculator model "sealed." I figured the reverse would work for the Vf1 cell.
It didn't. In fact NOTHING I tried, even extreme data produced anything but identical graphs.
I regret I don't have the smarts to model this design on that site - one intended for band pass anyway.
I will try Claudio's site. I've skimmed through it before church this morning. Seems DESIGNED to model this design w/out getting creative.
Thanks dude. Tony
Just what kelticwizard, I, Weems, and the guy authoring this site...
The Subwoofer DIY Page - Dual-chamber ported systems
has alluded to or clearly said.
thanks for your input. I'm not interested in a BP design so I'm struggling w/ how this "helps."
Over & over again I get the feeling from some replies: "this is a complicated alignment." I see it as barely, slightly more than that of an ordinary BR closure.
1. Model a driver as BR.
2. PLAN for extra space for the "extras." (divider and ports)
3. build it
4. test & tune
5. install extras per post #11...
Enjoy bass reflex frequency response w/, per kelticwizard and the link above, less cone motion, hence reducing the likelyhood of destroying your driver @ high levels.
What am I missing?!!! Please, please, PLEASE(!!) someone explain it to me in a way I'll understand.
I'd like to think MAYBE I could, as I posted before, "tune" to the effect of acoustic EQing between, say, 50-100hz. Unfortunately if I'm too dumb to model this on Claudio's site, I'll follow my own advice, not get fancy, and be happy w/the results.
Thanks Mike. Tony
A classic Weems DCR doesn't have a huge effect on cone movement at the upper resonance. To me, the main advantage is that you can build a long/tall enclosure without the enclosure length resonance intruding into the passband.
FWIW, all speaker cab alignments technically are BP since they only operate between the speaker's resonant frequency [Fb, Fc, etc.] and petering out to the driver's upper mass corner [~2*Fs/Qts].
Anyway, MC was just showing you that the second chamber is a 4th order BP that 'fills in' the gap between the driver's output and its vent output where it controls the cone's motion a bit more than a reflex, so if you don't want this 'BP', then you don't want to use a double bass reflex alignment.
GM
Thnx Don & GM for chiming in - enough to make my bulb glow a little brighter. I will build it the DCR. Modeling & narrowing down to size versus WAF.
The Micka.de simulator is pretty accurate for these. I based a DCR for a 4" TB fullrange on the results.
(no correction for port dimension)
An externally hosted image should be here but it was not working when we last tested it.
(no correction for port dimension)
An externally hosted image should be here but it was not working when we last tested it.
- Status
- Not open for further replies.