Differences between OBs and planars

[River757]

I've listened to, but never owned, Magneplanars, Apogees, Bertagni Electroacoustics and Martin-Logans, but the closest I've been to listening to an OB speaker system is a pair of Infinity RS IIs --> three open-backed 5" midbass drivers plus a rear-facing EMIT tweeter, in addition to dual ten inch woofers and a front-facing EMIT.

Since I've been considering building a small OB system, I was wondering if anyone had any opinions concerning the sonic differences and similarities between the two systems. Also, why aren't there more commercially-available OB speaker systems (besides for example Jamo's "R" series), since it seems like:

1) the lack of a full enclosure would be an advantage as far as manufacturing and especially shipping costs (shipping is one of those "hidden" costs but is still a very significant portion of a speaker's street price).

2) building an OB would be much less expensive vs. a planar with its rather delicate and specialized radiating surfaces.

Thanks!

 

[head_unit]

1) True (compared to a box speaker I think is what you meant), but but but...it might actually be harder to package and ship due to the odd shape. And while somewhat less material could be used for the enclosure, it would probably be somewhat MORE expensive. Why? Because the raw material (if veneered/vinyled MDF) is not really that expensive, and there are huge machines set up to churn out rectangular cabinets all day long. The price of things is most affected by volume; any thing unusual thus tends to cost more.

2) An OB is simply not a planar. It is just a speaker with no back, or at least no back over the midrange drivers. The "better sound" of the planar is not primarily due to the open back. That's just a necessary evil to avoid the huge expense of huge cabinets. (To my recollection only the Beveridge system had enclosed planars). The sound of the planar is primarily due to the diaphragm being driven over a greater portion of the radiating area, with magnetic systems having characteristics very different from cone transducer magnetics. Plus some planars avoid the evils of crossovers, or at least within most of the audio range.

 

[River757]

The "better sound" of the planar is not primarily due to the open back. That's just a necessary evil to avoid the huge expense of huge cabinets.

I thought the primary attraction of a planar was its dipolar nature and ability to create an "airy" soundfield. And the cone vs. diaphragm driving system aspect, while important too, was secondary.

Btw I'm not trying to start a war between OB fans and planar fans, I'm just trying to accumulate more info on this subject for my own benefit.

 

[DougL]

To me, OB and plainer speakers have similar sounds.
I attribute that to the dipole radiation and lack of a box.
I have friends that had Quads, Maggie's and I have an OB.

To me, it was no accident that the Dalquest DQ10 had an OB midrange.

Doug

 

[whgeiger]

Acoustic Image

I've listened to, but never owned, Magneplanars, Apogees, Bertagni Electroacoustics and Martin-Logans, but the closest I've been to listening to an OB speaker system is a pair of Infinity RS IIs --> three open-backed 5" midbass drivers plus a rear-facing EMIT tweeter, in addition to dual ten inch woofers and a front-facing EMIT.

Since I've been considering building a small OB system, I was wondering if anyone had any opinions concerning the sonic differences and similarities between the two systems. Also, why aren't there more commercially-available OB speaker systems (besides for example Jamo's "R" series), since it seems like:

1) the lack of a full enclosure would be an advantage as far as manufacturing and especially shipping costs (shipping is one of those "hidden" costs but is still a very significant portion of a speaker's street price).

2) building an OB would be much less expensive vs. a planar with its rather delicate and specialized radiating surfaces.

Thanks!

As the radiating surface of the driver(s) increases, the further behind the loudspeaker the acoustic image of the source recedes.

At lower frequencies, the demands on the driver(s) is increased beyond that of those mounted in a CB enclosure. So, it is not clear that OB loudspeaker would be cheaper to build even if produced and shipped in large quantities.

Regards,
WHG

 

[a.wayne]

I thought the primary attraction of a planar was its dipolar nature and ability to create an "airy" soundfield. And the cone vs. diaphragm driving system aspect, while important too, was secondary.

Btw I'm not trying to start a war between OB fans and planar fans, I'm just trying to accumulate more info on this subject for my own benefit.

Correct , along with a massive drive area .....

 

[River757]

To me, it was no accident that the Dalquest DQ10 had an OB midrange

You know, I've seen pics of that speaker many times, but always figured the felt "blanket" lying across the back of the mid driver was used in a cosmetic manner to cover up a closed-back driver (the 10's use of a plastic piezo[!!] supertweeter prompted that theory). You learn something new every day!

As the radiating surface of the driver(s) increases, the further behind the loudspeaker the acoustic image of the source recedes.

As Ed McMahon used to say, I did not know that. Thanks.

At lower frequencies, the demands on the driver(s) is increased beyond that of those mounted in a CB enclosure. So, it is not clear that OB loudspeaker would be cheaper to build even if produced and shipped in large quantities.

A good use for a small subwoofer then and just allow the OB to reproduce the frequencies that would allow the price to stay competitive?

 

[head_unit]

I thought the primary attraction of a planar was its dipolar nature and ability to create an "airy" soundfield. And the cone vs. diaphragm driving system aspect, while important too, was secondary...

Well, as a speaker engineer I guess I have a certain bias in how I look at things. Really planars have several things going on at once, which are not really separable.

I do believe the dipole effect is a kind of accident of design. I could be wrong, but I'm pretty sure the idea of driving a large surface came first, to:
- Have a very large surface enabling low excursion at each frequency (bass excluded).
- The diaphragm driven over a greater % of the surface, which seems to give better sound.
The magnetics also have a very different field leading to a very different excursion versus distortion relation, and the large rectangular diaphragms have quite different breakup modes compared to cones, both factors leading to different sound versus cone speakers.
In the case of the Bertagnis, in-ceiling or in-wall mounts are not dipoles, and dipole was not a big deal to those guys. I worked with them, that's how I know .

Then with such a big diaphragm, it would have been wildly expensive to enclose it-so they made a dipole instead. The dipole then happened to create this "open" sound due to a very different cancellation in the room vs. monopole.

Then some folks said "hey, let's just not put a baffle behind our midrange drivers and get a similar "airy" effect. Others decided to try reflectors and extra tweeters and 360 degree radiating surfaces like the Ohm Walsh as different approaches for "airiness"...
Vive la différence!

 

[John_E_Janowitz]

The biggest advantage to a ribbon, planar, electrostatic, etc over a standard VC driven driver is the lack of the VC itself. In a VC driven driver where you have a coil of wire, you have inductance to deal with. This inductance leads to issues with flux modulation and the non-linearities in the impedance curve which has a significant effect on the upper end of the response curve due to excursion. Lowering the effects of inductance in a VC driven driver has some significant benefits. This can be accomplished by a properly designed shorting ring. Per our white paper:

AE Speakers --- Superb Quality, Unforgettable Performance, Definitely.


"The copper effectively short circuits the inductance of the VC to an incredibly low level. This further lowers the influence of the inductance variance on the high frequency response of the driver. Also it forces a flatter phase curve for the driver. The main difference between electrostatics, ribbons, and VC driven drivers has always been the inductance. Here we can finally get a woofer to mate with these other drivers as closely as possible."

John

 

[mige0]

The biggest advantage to a ribbon, planar, electrostatic, etc over a standard VC driven driver is the lack of the VC itself.

Allow me question that - and that white paper too - the tenor is more of your sales strategy than anything else


Michael

 

[sreten]

Hi,

An immediate difference is completely different dispersion characteristics,
amongst a lot of other things when drivers are planar and not the normal
type drivers used in boxes.

rgds, sreten.

see : Quarter Wavelength Loudspeaker Design

 

[a.wayne]

The biggest advantage to a ribbon, planar, electrostatic, etc over a standard VC driven driver is the lack of the VC itself.

John

Very true John in the case of a ribbon driver it is as close to a pure resistor as possible ...


regards,

 

[River757]

Lots of great info here gentlemen (and ladies? ) - thanks! I don't fully understand everything presented so will continue to sit back and "absorb" it.

head_unit: this may not be the exact Bertagni I listened to back around 1984, but it looks identical:

"BES SM 280 Score!"

IIRC the irregular shapes of the VC-driven styrofoam diaphragms helped reduce or cancel out certain unwanted sonic artifacts.

 

[head_unit]

The biggest advantage to a ribbon, planar, electrostatic, etc over a standard VC driven driver is the lack of the VC itself...The main difference between electrostatics, ribbons, and VC driven drivers has always been the inductance...

Actually the MAIN difference is the diaphragms are often larger in area than cone speakers as well as different in shape (therefore dispersion as someone noted). The next main difference is they are typically much lighter than a cone, sometimes approaching down near the weight of the air mass load. And, perhaps more importantly, the diaphragm is driven over a greater portion of it's area. Instead of just driven from a small ring compared to a big cone, the driving force is distributed over most of the diaphragm. A further big difference is the electromagnetics are quite different versus cone drivers.

Shorting rings and other magnet structure improvements can definitely improve the performance of cone drivers-but they cannot somehow magically make a cone into a planar, they are just physically different entities.

 

[head_unit]

head_unit: this may not be the exact Bertagni I listened to back around 1984, but it looks identical:
IIRC the irregular shapes of the VC-driven styrofoam diaphragms helped reduce or cancel out certain unwanted sonic artifacts.

Ah, what a blast from the past. The company became Sound Advance and eventually Sonance absorbed them. Their best idea was making speakers to look exactly like ceiling tiles, for a super fast drop-in mounting. The acoustic design of the speakers was not through any sophisticated simulation-this was long before AkAbak©-but through a lot of experimentation.

I always suspected the shape was to accomodate a space to fit a tweeter and also to look like an ear. Probably the irregular shape also distributes resonances across various frequencies. The dispersion of their speakers was very good at high frequencies.

 

[John_E_Janowitz]

Actually the MAIN difference is the diaphragms are often larger in area than cone speakers as well as different in shape (therefore dispersion as someone noted). The next main difference is they are typically much lighter than a cone, sometimes approaching down near the weight of the air mass load. And, perhaps more importantly, the diaphragm is driven over a greater portion of it's area. Instead of just driven from a small ring compared to a big cone, the driving force is distributed over most of the diaphragm. A further big difference is the electromagnetics are quite different versus cone drivers.

Shorting rings and other magnet structure improvements can definitely improve the performance of cone drivers-but they cannot somehow magically make a cone into a planar, they are just physically different entities.

While some things you mention are common, they are not always that way. Diaphragm area can be different but doesn't a have to be. Small planar tweeters like the Infinity Emit's are no larger than a comparable dome tweeter. The driving area is also nearly the same in the case of those vs a dome tweeter. A large electrostatic panel could have much more mass than a small midrange driver. The difference though is that a VC driven driver always has a VC and these other drivers do now.

The main point I was making though is that a planar, ribbon, electrostatic, etc has virtually no inductance because it does not have a moving coil of wire in it. It is almost a purely resistive load across the board. This impedance doesn't vary with excursion as it does in a typical VC driven driver. As there is no energized moving coil of wire, there is also no flux modulation. The effects that these issues have on an impulse response and the phase of the driver are substantial. These are the reasons that these other types of drivers are able to be more accurate than VC driven drivers.

To get a driver with a VC driven driver to come close to what one of these other drivers can do you need to address those issues. Flatten the impedance curve and resulting phase. Lower overall Le and make it more linear with respect to excursion. Eliminate flux modulation. Obviously this doesn't all of a sudden make a cone driver into a planar, but it gets things closer and allows for more proper mating between the two.

John

 

[mige0]

To get a driver with a VC driven driver to come close to what one of these other drivers can do you need to address those issues. Flatten the impedance curve and resulting phase. Lower overall Le and make it more linear with respect to excursion. Eliminate flux modulation. Obviously this doesn't all of a sudden make a cone driver into a planar, but it gets things closer and allows for more proper mating between the two.

John

Again, I beg to disagree.

Flattening impedance has nothing to do with sonic difference in the scope of this thread - besides this, this feature can be found at many, many dynamic drivers nowadays and they definitely do not sound any like planars from this.

As said, you should not confuse people with your sales strategy versus the matter at hand.

As for flux modulation - this certainly is a different animal as its not easily EQ-able - but also - its by no way at the the core of "cone OB versus planar" IMO

Michael

 

[John_E_Janowitz]

Again, I beg to disagree.

Flattening impedance has nothing to do with sonic difference in the scope of this thread - besides this, this feature can be found at many, many dynamic drivers nowadays and they definitely do not sound any like planars from this.

As said, you should not confuse people with your sales strategy versus the matter at hand.

As for flux modulation - this certainly is a different animal as its not easily EQ-able - but also - its by no way at the the core of "cone OB versus planar" IMO

Michael

I would first ask you to re-read what I said. The question was what makes a planar different and I answered that question. In EVERY case, a planar does NOT have a VC, nor does it have the issues that come with a VC driven driver. A planar or ribbon has flat impedance. It doesn't have a huge rising response as frequency goes up, nor does it have a huge peak at the FS like a standard cone driver does. Because of having almost zero inductance this impedance curve does not change with excursion like a VC driven driver.

This is HUGE in designing a crossover and integrating with the rest of the system. In a passive crossover it doesn't require a zobel or any other impedance compensation. This does play into the sonic characteristics of the driver. More importantly though you can design a crossover that is accurate at more than just small signal levels. A typical VC driven driver will never have a crossover that gives a perfect transition to the next driver because. The varying inductance changes the impedance magnitude and phase as well as the frequency response magnitude and phase with every stroke of the driver. If you design your crossover with small signal levels in mind, it is valid only if the driver is not moving. As inductance typically goes up on the inward stroke and down on the outward stroke, there is no way to compensate for a change in both directions. You design for the middle ground but as you increase the excursions the crossover gets more and more off.

For this reason planars often sound better when implemented in a system because of the way they interact more properly to the crossover. Keep in mind that varying inductance also has effect on the time domain which a planar will not see.

Because the planar do not have an electromagnet(energized VC) moving through the gap and pushing the permanent field around, the planar has no issues with flux modulation. This again has a huge effect on the time domain and resultant phase match at the crossover point.

I'll take a RAAL ribbon or a TPL150 over any dome tweeter on any day. I'm a big fan of RAAL and I use them in nearly all the high end custom systems we do. There are clear advantages. Yes there is more moving area and that means you can get louder with less distortion. The ribbons are light, but that mostly plays into efficiency more than anything. A ribbon can be heavy and sound just as good as a lighter ribbon, the change is in efficiency. The main thing that makes them different in EVERY case though is the magnetic circuit. That is why you get lower distortion, especially as more power is applied, and where the sonic characteristics come from.

Clearly a VC driven driver does not magically become a planar by putting a shorting ring on it. I simply pointed out that if you want a better match, the issues that make the VC driven driver inferior should be addressed. A high inductance driver that has wildly varying Le with excursion, huge amounts of flux modulation and a response curve that changes with every stroke can never be considered as even a remotely good option to mate to a planar or ribbon. However if you can get rid of the flux modulation and keep the impedance curve flat and consistent at the crossover point you are much closer to a good match.

John

 

[mige0]

Again I beg to disagree.

As for the core topic of this thread:

Since I've been considering building a small OB system, I was wondering if anyone had any opinions concerning the sonic differences and similarities between the two systems.

So - its not exactly about difficulties to bring out the best or to work around the flaws of one principle or the other.



---------

Further, you strongly mix your statements regarding flux modulation and rise in inductance, as your sales strategy is to point to the flattish impedance curve of your speakers.
Those two have not *that* much in common regarding the sonic pattern of a speaker *if* applied correct.

As said - a flattish impedance curve is nothing exceptional nowadays for the first and also does not play *that* big of a role regarding XO design - not even in passive, as you always have to adapt to what you get - FR wise and impedance wise.

Flux modulation is different - I agree - but then, we do not have that much of *precise* data and measurements regarding this topic all over the bunch of speakers available.
More generally seen - its impact is embedded in the THD figures and at least for some speakers those figures are available. Some clever DIY magazines even show THD behaviour over signal level (pretty revealing sometimes ).

But all in all - these "distortion" related effects with its roots in the magnetic circuit are more or less a tear drop in the sea compared to other effects that come into play with planars versus cone.


So I stick to my conclusion that your posts here are not *that much* related to the original topic but rather are "fishing in the pot" - ie of commercially driven interest.


Michael

 

[mige0]

To add something more in the scope of the thread as well:

....
Since I've been considering building a small OB system,
....

Best advice is: go ahead and start out with something proven.
As you seem to be "new" to OB DIY, just don't expect to cut down cost of one of the mentioned speakers to that of a handful mediocre chassis plus a weekend of making sawdust.


If you are interested - besides above - in identifying sonic patterns and sort it regarding physical principles - welcome - this is a wide white field on the map still !

For a starter IMO planars are not planars in the sense of the topic given - meaning - a planar tweeter is quite a different animal than a planar (full range) speaker for your ear brain systems.

Though both may - or may not - relay on the same working principle plain ribbon, AMT, ESL, Magnetostat, etc - the big difference is first :

- are they used in dipole configuration or not ?
- does the used bandwidth cover the comb filter area ?
- how pronounced and how "smeared" is CMP behaviour ?

All above makes a heck of a difference in sonic perception, besides a lot more like circumstances where and how auditioned (room, placement), extending bandwidth beyond reasons, some more subtle underlaying mechanisms etc, etc.


Michael