How Passive Radiators work

[Steve Gonzales]

I found this short article an interesting read.

http://www.humanspeakers.com/whatis/PR.htm

[pmakres1]

Good reading Steve...thanks for posting it.

Peter :)

[4313B]

"They are mathematically identical"

False.

"An acoustic suspension system also tapers off in its response at only 6 dB per octave. The vented, or passive radiator, system rolls off at 18 dB per octave"

False.

Zilch posted in another thread that he is currently reading the AES Anthologies. Hopefully he can explain why the above two statements are false.

Zilch? Pop Quiz! :p

[Steve Gonzales]

I was hoping to get some minds working on the subject. There seems to be little dicussion of PR's, at least in depth. I know very little about the fine details of how they function and want to learn more about them.

[4313B]

Hopefully Zilch will accept his pop quiz. :)

[Zilch]

An "open book" quiz?

Oh boy, oh boy! :bouncy:

[Three papers to absorb here, first....]

[4313B]

Absolutely! Open book. Open notes.

Sorry if I put you on the spot Zilch.
You've got some really cool reading material there wouldn't you agree?

[Zilch]

You've got some really cool reading material there wouldn't you agree?Yup, it's fun to read it chronologically and watch the concepts develop.

[Did we REALLY usta look like that?] :p

[4313B]

[Did we REALLY usta look like that?] :p'fraid so... :p

[boputnam]

:wtf:

"...vented systems in general are a poor choice for small speakers that roll off much above 40 Hz - there is too much signal present, in the range they respond to poorly, for them to work well."

Gadzooks. Guess I should sell those 4313B's afterall...

"A larger speaker will work better with this sort of set up because at the frequencies that might be troublesome, there is rarely any musical signal present."

Complete :bs:. The inference is, I suppose, that a "vented" system is OK for large cabinets simply because at the LF's "there is rarely any musical signal present".

:barf:

[4313B]

Take it for what it is Bo. Just an opinion based upon the author's personal perception of the universe and everything in it. I'm sure there's some grain in there, I just don't have the desire to take the time to separate it from the chaff.

[4313B]

*****

Here's something else to read -

http://www.adireaudio.com/Files/TechPapers/GroupDelay.pdf

I found it Googling for Passive Radiators (I was actually looking for schematics so I wouldn't have to bother drawing them up myself! :banghead: ).
Pay special attention to the very last paragraph on page 4.


*****

Here's another one and it's The Real Story - I don't feel like reading it right now but here's the link anyway.

http://www.greenhydrant.com/~drees/ht/pr-vs-port.php


*****

GordonW should be able to add to this thread based on his experience with the newer, non-JBL passive radiators and their implemenation.


*****

[Chas]

"An acoustic suspension system also tapers off in its response at only 6 dB per octave. The vented, or passive radiator, system rolls off at 18 dB per octave"


I think the first "fact" is correct, but I have other ideas about the second, but then I don't want to spoil the fun for Zilch:bouncy: . I have spent a bit of time with the AES reprints too, they're great fun, there's stuff there you'll never forget.

I remember building my first woofers some 25 years ago from the AES T/S papers with four Altec 414z's, there were no T/S parameters published ever by Altec that I have ever found. The whole thing had to be learned and measured from scratch.

Going through all this coupled with the JBL learning curve I am having from you guys here has given this analog circuit level guy a real appreciation as to what is involved in high quality transducer (JBL pro stuff shouldn't be called "speakers" IMHO) system design.

[pelly3s]

i've talked to the guy from human speakers and he is a little odd... i tried to sell him some parts before

[4313B]

I think the first "fact" is correctI suppose the opperative word is "mathematical". Since he doesn't elaborate we can only speculate how he arrived at that.


I have spent a bit of time with the AES reprints too
Ok, comparing the acoustical analogous circuits of a vented-box loudspeaker system and a passive radiator loudspeaker system we see that the passive radiator loudspeaker system has the additional element denoted by CAP - the acoustic compliance of the passive radiator's suspension (usually drawn between MAP and RAP in the schematic below).

If I read correctly, the vented-box system and the passive radiator system are only mathematically identical if the compliance of the passive radiator is infinite.

acoustical analogous circuit of a vented-box loudspeaker system for reference
MAP = acoustic mass of passive-radiator diaphragm including air load = acoustic mass of port or vent including air load
CAP = acoustic compliance of passive-radiator suspension = no equivalent for vented box system
RAP = acoustic resistance of passive-radiator suspension losses = acoustic resistance of port or vent losses

[Steve Gonzales]

You've got my attention. I have no way to challenge this information, I'm just a spectator. Keep the idea's and thoughts coming.:bouncy:

[Mr. Widget]

Ok, comparing the acoustical analogous circuits of a vented-box....

The mathematical models do a fine job of predicting the lower bass performance. One thing they don't do is show you the added upper bass output that passives tend to add. In the contemporary small cube subs this isn't an issue, but in traditional JBL designs like the L220 and many others there is an added upper bass warmth that you either like or you don't. If you don't you should definitely use a port, if you find it desirable and the added group delay isn't an issue for you (it usually isn't for me), then you should opt for the PR.

Widget

[4313B]

I guess I don't get your point with respect to quoting me Mr. Widget. My comparison of the models was to show that they are, indeed, not identical.


If you don't you should definitely use a port, if you find it desirable and the added group delay isn't an issue for you (it usually isn't for me), then you should opt for the PR.I personally see one reason to use a PR and that would be to tune a small enclosure to a low frequency where a ducted port simply would not fit.

[pmakres1]

You've got my attention. I have no way to challenge this information, I'm just a spectator. Keep the idea's and thoughts coming.:bouncy:
Me too! I am watching this thread with interest. :bouncy:



The mathematical models do a fine job of predicting the lower bass performance. One thing they don't do is show you the added upper bass output that passives tend to add. In the contemporary small cube subs this isn't an issue, but in traditional JBL designs like the L220 and many others there is an added upper bass warmth that you either like or you don't. If you don't you should definitely use a port, if you find it desirable and the added group delay isn't an issue for you (it usually isn't for me), then you should opt for the PR.

I LOVE the upper bass/lower midrange of the L220. :D

Peter

[Mr. Widget]

I guess I don't get your point with respect to quoting me Mr. Widget. My comparison of the models was to show that they are, indeed, not identical.

I realize that. I was showing that in addition to their not being identical in low bass performance there is also the difference in upper bass that I mentioned.

Widget

[Mr. Widget]

I LOVE the upper bass/lower midrange of the L220. :D



That's why it's there! For those who do.:thmbsup:

Widget

[4313B]

So you prefer the even steeper rolloff of the passive radiator system as compared to the vented system?

This brings us to "An acoustic suspension system also tapers off in its response at only 6 dB per octave. The vented, or passive radiator, system rolls off at 18 dB per octave".

The acoustic suspension system has a rolloff of 12 dB/octave, the vented system can have a rolloff of 18 dB/octave to 24 dB/octave and the PR system has a rolloff of 30 dB/octave. 6th order assisted alignments have a rolloff of 36 dB/octave. Incidentally, JBL vented systems are often quasi-third order alignments and roll off at ~ 18 dB/octave. Looking at the B380 or B460 real quick - QB3 unassisted (no BX63), and B5 assisted (with BX63) - same ~ rolloff as a classic PR system. None of this is black and white today and the modeling programs allow us to move between the various alignments of old with much greater ease. I would not be surprised to find that the L220 in particular exhibited more of a 4th order response than a 5th order response.

[Mr. Widget]

So you prefer the even steeper rolloff of the passive radiator system as compared to the vented system?

I am not sure if you were asking me...

I do not. I also do not like the added warmth. I prefer a dry sound with lots of extension. Typically I prefer a sealed system, but in the realm of JBLs I typically prefer a vented alignment... even with the Sub1500.

Widget

[boputnam]

If I read correctly, the vented-box system and the passive radiator system are only mathematically identical if the compliance of the passive radiator is infinite. Oddly, that was my one thought reflecting back to my days schooling in Strength of Materials, P-Chem, Diffusion, etc.

[4313B]

Typically I prefer a sealed system, but in the realm of JBLs I typically prefer a vented alignment... even with the Sub1500.WidgetAh! Well... I prefer the room gain one can appreciate with the SUB1500 in a sealed enclosure. So much so that I am willing to suffer the substantial efficiency penalty. However, for high efficiency mains I believe the SUB1500 ported is the ticket. The sealed SUB1500 is not a slam dunk solution, it probably does take more care to properly implement than the ported solution and some EQ might be desired. Case in point - I use a cut filter at the high end of the SUB1500's bandwidth (as opposed to a boost filter at the bottom like the Revel Sub). I will be working on a DX-1 equivalent card in the future which can either be set to cut high or boost low.

[4313B]

I do not. I also do not like the added warmth.I messed around alot with the PR8 and PR10 and found them to be quite cool at the time, providing the enclosure volumes were kept small. I liked the added warmth. The 124, LE14, and 136 with PR's weren't bad either as subs. Having the stock L220 (and a 136 version of the L220) in the same room as the 240Ti and 250Ti was the final nail in the coffin as far as JBL PR's for me personally. One of my buddies prefers the PR's to this day. GordonW has mentioned the newer PR's are pretty cool. And so it goes...

This is all theory guys. I point everyone back to the last paragraph of the pdf I posted earlier in the thread. We can argue till the cows come home about which is the definitive solution and we just won't solve a thing. We like what we like and that's that. My whole point was to challenge the two statements made as outlined in my first post.

BTW - rumor is the 240Ti and 250Ti were ported on the back to keep the front veneered baffles nice and clean, and not to keep unwanted internal reflections from radiating into the forward listening position. Now of course JBL (and everyone else) states the ports on the back are to prevent just that, which they do. So... I guess we could say the 240Ti and 250Ti benefited quite nicely from what was allegedly intended as a cosmetic decision. :p

[GordonW]

That's right- a PR will behave just like a port, with infinite compliance (ie, acting as a PURE MASS with no SPRING component.

Otherwise, the PR is really a FIFTH-order system, rather than a (quasi-)THIRD or FOURTH order system, as all ported systems behave as. There is a consequence here- the rolloff shape of a PR system is SHARPER than a ported system, below the final resonance of the transfer function as you go down in frequency. But, OTOH, you can oftentimes use the additional term in the transfer function (ie, PR tuning) to eke out just a little more bottom end energy, just at resonance, compared to a ported system, at the expense of just a little upper bass. For many room/speaker combos, this makes for a "further-extended-quasi-EBS" alignment... even more bass extension, at the expense of just a little degraded transient response at EXTREME low frequencies.

My tactic with these, is to use PRs tuned at EXTREME low frequencies... as in, well below 20 Hz box resonance. This usually requires PRs with self-resonances in the 5 Hz range... which can tax the suspension of cheaply-made PRs, due to the substantial mass required on the PR diaprhagm to do this. Also, given that increased compliance also helps... this also requires a very carefully designed PR, to not "flop around" under this type of loading. But, we've seen some real advances in spider and surround geometry on PRs; there's some SERIOUS units out there with SERIOUS volume displacement (ie, they can move a LONG WAY in and out), and still manage to keep things in control...

Regards,
Gordon.

[4313B]

Thanks Gordon! :)

What might be groovy is if we could get the specs on the old JBL PR's. I'm quite certain JBL never ran the numbers on any of them. We'd have to measure them all ourselves... I'm also pretty certain their suspensions aren't up to excessive weight.

[Chas]

Interesting, my recollection was that vented roll off was only 24 db/octave. But I had not considered the various alignment options. Just shows my age, I guess.

Geez, I remember reading somewhere that acoustic suspension designs were first order roll-off, I'll bow to your experience....until I can look it up, of course. ;)

My SUB-1500's work just great in 3 cu.ft. sealed boxes BUT I only operate below 30 Hz for the really deep stuff, not so much for musical purposes, but for the sensation of "being there" ambience, etc., the benefit is no EQ is really necessary in my room. Besides, four 2235's just don't need any help between 30 and 100 Hz!:D

[DavidF]

Geez, I remember reading somewhere that acoustic suspension designs were first order roll-off,


The driver itself has the second-order rolloff (or roll-on I suppose is more appropriate), so you have to start there.

DavidF

[pmakres1]

Keep it going guys...I'm really liking this thread, though I confess I get a bit lost in the enclosure alignment discussions. I'll be re-reading the thread a few times I'm sure. The discussions about first, second, third and so on alignments brings me to a question I've wondered about for some time. The brochure I have on the B460 states the following: "The B460 uses the JBL model 2245H 18-inch (460mm) driver in a vented enclosure of 8 cubic feet (226 liters). The combination is a third-order quasi-Butterworth alignment. The enclosure is tuned to 26Hz." Could some of you fellows who are more up on this than I explain what this means? I was never an expert in this area, and I need at the very least a basic refresher on 1st, 2nd, 3rd and so on order alignments and what it all means. Keep it simple, or, if you like just point me to some relevant reading. Pardon my ignorance, I'm trying to learn a bit here.....:o:

I can tell you first hand that the B460 combined with the L220 is a great combo! :)

Peter

[4313B]

I must be on your ignore list...

http://audioheritage.org/vbulletin/showpost.php?p=53594&postcount=22

http://audioheritage.org/vbulletin/showthread.php?t=4557

The B380 and B460 enclosure volumes and tuning frequencies are such that the 2235H and 2245H exhibit responses equivalent to a quasi-third order filter. Tune the stock B380 volume up a bit to ~ 29 Hz and you get Keele's maximally flat response for the 2235H which is closer to a classic fourth order Butterworth response (still QB3 due to the low Q's of the transducers). Think of enclosure/transducer systems as passive filters. Add a BX-63 to a B380 or B460, with it's second order bump filter, and you end up with a 5th order system -similar rolloff to a classic PR system. One JBL transducer that fits almost perfectly into a classic 4th order Butterworth vented box alignment without forcing any "compliance shifting" would be the LE10 with it's Qts close to 0.38

[pmakres1]

I must be on your ignore list...

http://audioheritage.org/vbulletin/showpost.php?p=53594&postcount=22

Sorry Giskard,

I did read that post, and I was going from memory of the B460 brochure, which I thought had said quasi fifth-order. When I pulled the brochure again it said quasi third-order. I should have re-read your post before I asked the question. (no you are not on my ignore list). What about the term "Butterworth" alignment?
I apologize again for my ignorance..I'd still like to see some basic reading on the driver alignments.

Peter

[4313B]

Butterworth - they're people's names. Invent a new filter and we'll name it after you. Linkwitz-Riley, Zobel (which should probably be called Boucherot?), etc.

I'd still like to see some basic reading on the driver alignments.

There has to some good stuff out there on the net. Some of the AES stuff is too deep without a Mathematics Degree from M.I.T.

This might be a place to look - http://www.diysubwoofers.org/

This might have some stuff - http://www.mhsoft.nl/default1.asp

[Steve Gonzales]

Wonderful discussion!. I will say that in my limited (so far), non LHS reaserch, as in the case of my original link, it is hard to tell who's the real deal (right), as far as the bottom line. I would love to be able to see a read-only article in the reference section on this site. It could be used as a learning tool for budding builders. I want to thank every participant for their input. I'm definately much more aware of what is going on and what to expect, when using a PR. I would like to make one observation. I have owned 3 pairs of (sorry)-ESS AMT1A,B,D's in the past and these speakers use a VERY crude and simple PR. Their bass output is nothing short of AMAZING!. As I see all the pitfalls one will have to overcome to eventually come out with a good sounding product when using a PR, it is even more amazing. Thanks again, Steve G

[pmakres1]

Butterworth - they're people's names. Invent a new filter and we'll name it after you. Linkwitz-Riley, Zobel (which should probably be called Boucherot?), etc.

I'd still like to see some basic reading on the driver alignments.

There has to some good stuff out there on the net. Some of the AES stuff is too deep without a Mathematics Degree from M.I.T.

This might be a place to look - http://www.diysubwoofers.org/

This might have some stuff - http://www.mhsoft.nl/default1.asp

Thanks Giskard!

Very helpful.

Best regards,

Peter

[4313B]

Hopefully others can add in some links. I've noticed that some of my old links are now broken. Books are still worth owning. :p

[Chas]

Okay here's what Villchur, AES Sept.1956, had to say regarding Baffled direct-radiators which I interpret to be what I used call to infinite baffles and assuming these would have the same charecteristics as acoustic suspension boxes:

"Below resonance, however, the speaker mechanical system is no longer mass-controlled. If it were resistance-controlled, velocity would remain constant, and output would be attenuated at a rate of 6db/octave due to the continuing drop in air-load resistance. In fact, it is compliance-controlled-the predominant impedance is that of the elastic restoring force of the system-and bass output is therefore attenuated at a rate of 12 db/octave."

This must have been what I was thinking about earlier.

I also found an old article in the AES book where Olsen describes a "drone cone" i.e. a passive radiator. He thinks that effeciency goes up with a PR due to the fact that losses are less with a PR versus a vent. He wrote that in 1947.

[4313B]

Ah!

Thanks Chas :)

This might be interesting - http://www.audioasylum.com/forums/pi/messages/2062.html

[Chas]

Holy crap! I think you are introducing too many variables just to be argumentative!:D

Ever check out a range of delta T of ambient air or maybe even altitude versus change in FB in a vented system?:blah:

Very interesting graphs based on even more variables, though.....:cheers:
Just proves to me once again, build, listen, change, listen, etc....science. art and ears all combined. That's what makes this so much fun.

[4313B]

Holy crap! I think you are introducing too many variables just to be argumentative!:D Nope, just wanted a quick and easy way to show how the tuning/volume can affect the knee of the curve with a transducer/box system just like doing something like adding DCR to a parallel inductor in a filter does, hence the link. My intent should be clear. If it isn't, then I suck at explaining things... whatever...


Ever check out a range of delta T of ambient air or maybe even altitude versus change in FB in a vented system? http://audioheritage.org/vbulletin/images/smilies/blah.gifPerhaps this will help -

Sensitivity of Thiele's Vented Loudspeaker Enclosure Alignments to Parameter Variations
Volume 21 Number 4 pp. 246-255; May 1973
Additional information on the use and application of Thiele's alignments for the vented loudspeaker cabinet is presented. A rewritten alignment table which has all the frequency terms normalized to the speaker resonance frequency is included. Computer-run frequency responses for all the alignments are displayed along with a new fourth-order Chebyshev alignment beyond no. 9. Variations and sensitivity functions for the vented cabinet output with respect to various system parameters (both Thiele system parameters and driver physical parameters) are derived and plotted.
Author: Keele, Jr., D.B.

[Chas]

Quote: "My intent should be clear." It is, I was kidding. I should also have added "measure" to my diatribe too.

[4313B]

Just proves to me once again, build, listen, change, listen, etc....science. art and ears all combined. That's what makes this so much fun.Understood.

I'm going to generalize something I heard recently -

Some guys are indispensable when it comes to theory and mathematics but their implementations kind of suck, other guys aren't so good with theory and mathematics but their designs wouldn't reflect it.

[tomt]

I'm going to generalize something I heard recently -

Some guys are indispensable when it comes to theory and mathematics but their implementations kind of suck, other guys aren't so good with theory and mathematics but their designs wouldn't reflect it.



here is a man who has a degree from mit and has 'invented', 'filters',

as such,named after himself(post 34) and has even as this site says

get this,Repaired Transistors! What a Talent!

http://web.mit.edu/invent/iow/bose.html

Repaired Transitors? How?

must have put them in piles of two.
then did it again

[4313B]

Are you posting an example of the former as opposed to the later? :p

[tomt]

Are you posting an example of the former as opposed to the later? :p



both.

This Man Transends Mortal Status

http://www.bose.com/pdf/pro/tech_data/panaray_402/td_panaray_402.pdf

under "general description'

"low impedance, edge-wound Ferrite V ceramic magnets"!

[4313B]

I guess we all have to have our heros? :dont-know

Here's one of mine :thmbsup: - Rick Rescorla (www.medaloffreedom.com/RickRescorla1.htm)

[tomt]

yep...

amar bose
ann coulter-after the sex change/before the weight loss-
http://www.
http://www.
http://www.
rush limmbauh
ray dolby
james randi
and many more
to me these are anti-heros

real heros-

bart locanthi-
http://www.sabl.com/~bart/bnl2/
john curl
sherman skolnick-
http://www.rense.com/Datapages/skolnickdatapage.html
henk van rensberge-http:www.abandoned-places.com
wayne weslowski
susumu sakuma-
http://www10.big.or.jp/~dh/sakuma/index.html
tom yeago
nelson pass

[DavidF]

I guess we all have to have our heros? :dont-know

Here's one of mine :thmbsup: - Rick Rescorla (http://www.medaloffreedom.com/RickRescorla1.htm)

I'll second that...

DavidF

[tomt]

a few days ago i bought inexpensive speakers (kenwoods or somthing)
for some passive radiator experiments.
woofer as pr,vc wired to different combinations ofcomponents
for adjustable response,
been done before though has'nt it?

i wonder maybe your man rick might be an intelligence agent

apogees for unsavory links, seen that on outside asylum a few days a go

any thing i would ever post about bose is a joke
much like there own literature
really now, 'edge wound magnets'?
from their own literature

[Mr. Widget]

Hi Tomt... not to be an old prude, but I think it would be best if you deleted those three links. It really won't change the tone of your post.


Widget

[tomt]

links deleted



woofer as pr,vc wired to different combinations of components
for adjustable response,
been done before though has'nt it?

has anybody done such a thing?

[Steve Gonzales]

Thank you Mr. Giskard for all your informative work on this thead, and others. What is all that other stuff? Moderators....?. I have read a new report about the Definative Technology Reference Sub and it certainly matches my personal listening experience. They tested this appro. 16" /cubed subwoofer to a very USEABLE (100db) at 12hertz- yeah 12 hz!. It features a 14" active driver and TWO 14" passives with a Class D, 1800 watt amp. Pretty amazing stuff! :)