Dipole midbasses?

[Feandil]

Well I've been reading about the advantages to a dipole setup (though I've never heard one to compare)

I plan to purchase a cheap cheap pair of pro audio 12" to make a "test" subwoofer to see if I like Dipole sound

I would, however, like for someone to give me a link to a dipole midbass solution

if it works out and it sounds like everyone boasts, I might make a 8-10 speaker bass array with 10" drivers....

but I've been reading and there are certian baffle design problems with a midbass dipole solution. any calculators or programs to help me figure it out would be appreciated

also is there any set distance between the dipoled woofers and each other?

What's the difference between W-shaped and H-shaped?

I believe H-shaped will not allow for use up to 300 hz like the midbasses would play... so I dunno what I need to accomplish this...

 

[Paul W]

Jessica,
Dipole mid-bass is quite achievable. Suggest you look up S Linkwitz site...then read through it about three times He has excellent reference material there, including a baffle calculator.
Paul

My dipoles

 

[ScottG]

http://www.t-linespeakers.org/downloads.html

most solutions require active eq. - some drivers however (if they are sufficently eff. in comparison to the pairing midrange) can work with a "mechanical" eq. via a high Qts.

I haven't had time to play with mine yet, but these are the drivers I selected for just this solution (2*16ohm per side):

http://store.steelsound.com/Detail.bok?no=478

specs here:

http://madisonspeakers.netfirms.com/rawdrivers.htm

as to IF they meet specs and are suitable for this application - well maybe you'll get there b4 me.

 

[Feandil]

His explanation seems rather poor...

and I couldn't find any calculators...

he shows you how to build a dipole but no info on a dipole past 100 hz...

when he takes his dipole outside the response after 100 hz is terrible... I would need that flat reponse to 300 hz or more...

I understand how to make the dipole H-shape etc... but I don't understand in a W-shape or some shape how to create the baffle to work up to 300hz

 

[Feandil]

http://www.t-linespeakers.org/downloads.html

most solutions require active eq. - some drivers however (if they are sufficently eff. in comparison to the pairing midrange) can work with a "mechanical" eq. via a high Qts.

I haven't had time to play with mine yet, but these are the drivers I selected for just this solution (2*16ohm per side):

http://store.steelsound.com/Detail.bok?no=478

specs here:

http://madisonspeakers.netfirms.com/rawdrivers.htm

as to IF they meet specs and are suitable for this application - well maybe you'll get there b4 me.

those drivers are very impressive... tell me what you think of them

I would like to do 16 10" midbasses per tower or 8 15" midbasses in a dipole... and those knights look good especially the price

I've seen dipoles in a push-pull pattern... any info on this??? also the classic w-shaped...

I don't really understand how you calculate the distance between drivers in the w-shaped... and the distance to the ends

 

[ScottG]

SL's is quite "technical" and you realy have to read it numerous times b4 the info really starts to make sense.. that said, most of his information is with regard to active eq. - NOT an acoustic solution.

In particular he DESIRES dipole operation (its cancelation effects) up to 300 Hz + (which is covered by the main panel in the phoenix - NOT by the subwoofer).

In this instance you want a solution that is quite different - a dedicated midbass driver to handle freq.s from about 55 Hz to 300 Hz (correct?). If so then you don't want a "baffle" per se. or rather you don't want anything more than to mount the midbass drivers too. (i.e. don't concern yourself with "W", "H", or "U" channel loading).

(Note that you will still need a subwoofer solution for freq.s below the bandpass mentioned.)

 

[454Casull]

His explanation seems rather poor...

and I couldn't find any calculators...

he shows you how to build a dipole but no info on a dipole past 100 hz...

when he takes his dipole outside the response after 100 hz is terrible... I would need that flat reponse to 300 hz or more...

I understand how to make the dipole H-shape etc... but I don't understand in a W-shape or some shape how to create the baffle to work up to 300hz

http://www.linkwitzlab.com/images/graphics/p9.gif

Dipole Scan-Speak drivers with a Scan-Speak 9800 tweeter. The crossover points are 100Hz and 1400Hz.

 

[Feandil]

SL's is quite "technical" and you realy have to read it numerous times b4 the info really starts to make sense.. that said, most of his information is with regard to active eq. - NOT an acoustic solution.

In particular he DESIRES dipole operation (its cancelation effects) up to 300 Hz + (which is covered by the main panel in the phoenix - NOT by the subwoofer).

In this instance you want a solution that is quite different - a dedicated midbass driver to handle freq.s from about 55 Hz to 300 Hz (correct?). If so then you don't want a "baffle" per se. or rather you don't want anything more than to mount the midbass drivers too. (i.e. don't concern yourself with "W", "H", or "U" channel loading).

(Note that you will still need a subwoofer solution for freq.s below the bandpass mentioned.)

right I understand the subwoofer thing

I don't understand how you can make a main panel act like a dipole... his explaniation isn't very good and his drawings aren't good either....

I see from the first posters dipole that he just mounted them to the front baffle and that was it... but all the dipoles I've seen are push pull or some thing like that so I don't understand how that would work....

either way... I'm still trying to understand this lintwitz page

 

[ScottG]

those drivers are very impressive... tell me what you think of them

I would like to do 16 10" midbasses per tower or 8 15" midbasses in a dipole... and those knights look good especially the price

I've seen dipoles in a push-pull pattern... any info on this??? also the classic w-shaped...

I don't really understand how you calculate the distance between drivers in the w-shaped... and the distance to the ends

they appear well made, but require EXTENSIVE break-in to "loosen-up" the surrounds - any measurement taken under at least 300 hours of dynamic playback would be erronius.

I thought I heard some buzzing comming from the surround on one of the drivers during a very heavy bass track - but I could have been mistaken (in any case the remedy is fairly easy - rubber cement in the appropriate area). They are quite dynamic for your average dipole solution (in large measure due to the B (of Bl) and the mms for the given operating bandwidth), when compared to most other dipole solutions. The trade-off of course is that they don't go as low, but that IS a benefit in a dipole design because of the excursion required.

Again, the baffle configurations are utilized to essentially REMOVE dipole effects for a given freq. - unless I'm mistaken, this is NOT what you are after so you don't need to concern yourself with it (i.e just make a baffle wide enough to support the drivers - no wider).

Push-pull is used to cancel distortion as freq. decreases. You could incorperate this opperation in several different manners. You could even do this so that the driver becomes MORE directional and approaches a 2nd order gradient.

Moreover IF you wanted to do this and extened the response of the driver you could:

Operate the drivers in phase but with their magnets back to back. The front facing driver will be facing the listener. The rear driver will "face" into a T-line.

 

[Feandil]

they appear well made, but require EXTENSIVE break-in to "loosen-up" the surrounds - any measurement taken under at least 300 hours of dynamic playback would be erronius.

I thought I heard some buzzing comming from the surround on one of the drivers during a very heavy bass track - but I could have been mistaken (in any case the remedy is fairly easy - rubber cement in the appropriate area). They are quite dynamic for your average dipole solution (in large measure due to the B (of Bl) and the mms for the given operating bandwidth), when compared to most other dipole solutions. The trade-off of course is that they don't go as low, but that IS a benefit in a dipole design because of the excursion required.

Again, the baffle configurations are utilized to essentially REMOVE dipole effects for a given freq. - unless I'm mistaken, this is NOT what you are after so you don't need to concern yourself with it (i.e just make a baffle wide enough to support the drivers - no wider).

Push-pull is used to cancel distortion as freq. decreases. You could incorperate this opperation in several different manners. You could even do this so that the driver becomes MORE directional and approaches a 2nd order gradient.

Moreover IF you wanted to do this and extened the response of the driver you could:

Operate the drivers in phase but with their magnets back to back. The front facing driver will be facing the listener. The rear driver will "face" into a T-line.

What kind of transmission line???

I would like to operate this dipole midbass towers to 55 hz or so... maybe to 40 if possible

I'm unsure what you mean about the t-line though....

What I see a lot with dipole towers is:

Front of driver>>>>>>>> opening out <<<<<<<<front of other driver

would this be what I would do???

and about those knights... that seems like a massive amount of break in time... that makes me question the quality

 

[ScottG]

right I understand the subwoofer thing

I don't understand how you can make a main panel act like a dipole... his explaniation isn't very good and his drawings aren't good either....

I see from the first posters dipole that he just mounted them to the front baffle and that was it... but all the dipoles I've seen are push pull or some thing like that so I don't understand how that would work....

either way... I'm still trying to understand this lintwitz page

Virtually ALL driver operate with 2 phases of output. the forward radiation is one phase, the rearward radiation is the opposite - when these two opposite phase signals collide (when there is no box to contain the reward radiation) the point at which the two outputs "mix" (at the sides of the speaker) they cancel each other out. As seen here:

http://www.linkwitzlab.com/rooms.htm

On the dipole diagram the red circle is the forward radiation, the blue is the out-of-phase rear radiation. You'll notice that at the sides there is NO output because this is where the "mixing" is occuring and causing a null.

 

[ScottG]

What kind of transmission line???

I would like to operate this dipole midbass towers to 55 hz or so... maybe to 40 if possible

I'm unsure what you mean about the t-line though....

What I see a lot with dipole towers is:

Front of driver>>>>>>>> opening out <<<<<<<<front of other driver

would this be what I would do???

and about those knights... that seems like a massive amount of break in time... that makes me question the quality

About T-lines generally:
http://www.t-linespeakers.org/

They are essentially a closed box system where the inside of the box has a maze-like labyrinth (not unlike a long tube bent-up) that is calculated based on several of the drivers paramaters.

I see it like this (for the solution I posited):

Listener..Front of FRONT Driver attached to baffle (with width of driver)..rear of FRONT driver and magnet..rear of REAR driver and magnet (abutting rear of Front driver's magnet)..front of REAR driver attached to box containing a T-Line.

(not exactly a beg. project though..)

As to the drivers needing a long break-in period. This is perfectly normal for most drivers - especially drivers with an accordian suspension.

You posted:
"I see from the first posters dipole that he just mounted them to the front baffle and that was it... but all the dipoles I've seen are push pull or some thing like that so I don't understand how that would work...."

I'd suggest starting like this b4 something as complicated as the above - don't worry about push-pull operation. I'm not in-experienced in this area and I've chosen this method with just 2 drivers per side (connected in parallel to give 8 ohms with a resulting increase in output). Its fairly cheap and if it doesn't work for you then you will not have wasted a lot on it.

 

[Feandil]

Well I'm up to the transmission line project...

what I dont understand is to acheive dipole wouldn't one driver have to be out of phase?? either through wiring or placement???

in that guy's dipole is one of those seas out of phase with the other??

also I planned messing with dipole before stepping up to the finished project...

I know what a t-line is.... but I kind of don't understand in that situation would it would solve... where are you using that t-line to??? where does it come out (the sound)?

 

[Feandil]

reading up a t-lines I'm unsure about how I could practially impliment this... I understand what you are saying with the creation of an extra chamber "in front" of the rear speaker to act as an enclosure

What made this work in my mind is to think about it like it's ONE HUGE enclosure...

where for each back motion of the front facing speaker it moves the rear speaker out into the "box" that the TL is attached to.... just like a normal sub in a TL box would do...

now I understand what you would do... you would probably run the TL's "Port" to the front of the tower as well... creating a very directional dipole??????

I'm unsure... seems like thats makes sense though

 

[5th element]

You have not understood what scott said.

If you imagine a loudspeaker drive unit, the membrane or cone of it vibrates when a signal is applied to it.

Now the cone moves fowards and backwards, the cone produces (almost) EXACTLY the same signal from the front of the cone as it does from the back.

So from the front you have someone singing, from the back you also have the same person singing exactly the same as the front of the cone. Now the difference is the sound at the front of the cone is 180degrees out of phase with the sound produced from the reverse of the cone.

When these two signals mix they cancel = zero sound. What frequency this cancellation occurs at is controlled by the width of the panel the driver is mounted in.

 

[Saurav]

what I dont understand is to acheive dipole wouldn't one driver have to be out of phase??

Not necessarily. You can "open the back" of the enclosure, thus allowing the back wave of the driver into the room. Then the two out-of-phase sound waves are the back wave and the front wave of the driver (which are obviously out of phase with each other). So you can get dipole operation from a single driver, if you put it in an open baffle. If you use a closed box, then I guess you'd need to mount a second driver on the back and wire it out of phase or something. But that's not the only way to do a dipole. You can do it with a single driver, or with multiple drivers wired in phase, if you allow the back wave of the driver to interact with the front wave of the driver.

Hope that makes sense.

Edit: Or, what 5th Element said

 

[ScottG]

Yup, what 5th and Saurav said (..some times it takes a lot just to understand the basics).

a dipole "driver" is nothing more than a driver without the rear output enclosed in a box. i.e. a single 16 ohm madison 10" driver without a box operates in dipole automatically (without any additional drivers).

when I suggested the use of 2 drivers (on the simple config.) - it was NOT in a compound load (like push-pull, etc.), rather it was just the two drivers (one on "top" of the other) in a vertical orientation (wired in parallel). In other words as the listener you see two drivers - like SL's "main panel" on the phoenix (without the tweeter being between the two drivers).

http://www.linkwitzlab.com/m_panel.htm

This is about as basic as you can get (with regard to acoustic loading and cabinet building complexity).

 

[Feandil]

You have not understood what scott said.

If you imagine a loudspeaker drive unit, the membrane or cone of it vibrates when a signal is applied to it.

Now the cone moves fowards and backwards, the cone produces (almost) EXACTLY the same signal from the front of the cone as it does from the back.

So from the front you have someone singing, from the back you also have the same person singing exactly the same as the front of the cone. Now the difference is the sound at the front of the cone is 180degrees out of phase with the sound produced from the reverse of the cone.

When these two signals mix they cancel = zero sound. What frequency this cancellation occurs at is controlled by the width of the panel the driver is mounted in.

I understand what phasing is from the back of a driver... I've been in audio for awhile just my first attempt at dipole...

 

[Feandil]

Not necessarily. You can "open the back" of the enclosure, thus allowing the back wave of the driver into the room. Then the two out-of-phase sound waves are the back wave and the front wave of the driver (which are obviously out of phase with each other). So you can get dipole operation from a single driver, if you put it in an open baffle. If you use a closed box, then I guess you'd need to mount a second driver on the back and wire it out of phase or something. But that's not the only way to do a dipole. You can do it with a single driver, or with multiple drivers wired in phase, if you allow the back wave of the driver to interact with the front wave of the driver.

Hope that makes sense.

Edit: Or, what 5th Element said

AH... so all along people using OB line arrays were trying to acheive this...

hmmm... does this mean that like what is always assoicated with OB line arrays, that the tower has to be 4 feet or so away from a wall??

if so I don't have that much room

also if I do a closed Dipole would this allow usage closer to a wall as all the backwaves will be allowed forward?

 

[Saurav]

That, I don't know. I've just moved to a smaller room with a dipole speaker, and I'm trying to figure out how to make it work. I *think* I might be able to get it closer to the wall if I can find a way to absorb/diffuse the back wave so it doesn't reflect off the wall and come back into the room. But that's a guess, I'm not sure how it's going to work out.