[Danny Richie]

So, how do I know what frequencies will couple and what frequencies will cancel each other out?

It is all about wavelength and distance.

Take two drivers and a distance to ear (or mic), then vary the distance of one driver in relation to the other from the distance to ear (or mic). Whatever the distance is the wavelength out of phase.

so, do those stacked planars actually increase high frequency sensitivity?

In the top octave or in about half of the top octave, no they do not couple (no increase). However, as frequency decreases the output spreads out more in all directions (off axis response increases) and they begin to couple and output levels increase.

Instead of various ribbons tweeters I would prefer the Carver Amazing 60 inches ribbons and a half dozen of subs like the Gilmore Audio Model1, now with a SPL of 91,5dB. Anyone know if this Carver 60'' is sell as raw driver??

I can get the big BG planars all day long. But honestly, the long line of the smaller Neo's sound better and have much higher sensitivity.

HT oOz, I think you are starting to get this all figured out.

[FullRangeMan]

Dear Mr. Danny,
Just see this BG RD75 at Partsexpress, nice to know Bohlender is making a 75 inches driver. They are magneplanar system with poliester membrane, this is not a problem.
But Iam afraid all this BG RD line is monopole, firing sound front only and need a box as used in VMPS.
The beauty of the Carver Amazing is in the (Dipole) Soundstage, it is really adictive (you do not listen the music, you feel the music etc).
If this BG is a Dipole driver it is a alternative to the widows of the Carver Ribbon like me.
Regards.
P.S.: If this BG are monopole only I have to reject it, it have no effect in my brain or heart, Iam a Dipole junkie.
Impressive statement below from a BG RD75 user at PartsExpress:
--------------
Difficult to use.
Friday, June 12, 2009
Paul from Madison,WI 
I have long been a fan of electrostatic loudspeakers and when I heard that the RD 75 B&Gs are even better than electrostatics I was eager to give them a try. I have found it very difficult to get these to perform at the level I have come to expect. First, there is the 5K anomaly that requires a notch filter to tame. Instead I cross over to a RAAL 70-10B at 4K via a Pass Labs XVR1 electronic crossover to avoid this and to get extension up past 20K. Next, there is the dipole issue. I like the dipole nature of electrostatics, but on the Bohlender...mmm not so much. Soon I will try them in an enclosure to see if this helps. Finally, there is the usable range. As others have noted, you really should cross these over at a much higher frequency that the 150Hz stated. I have tried many and settled with 220HZ with a 4th order 24/db octave slope mated to an array of 6 Usher 8955A in sealed enclosures with 1cuft of space per driver. All in all many dollars spent for sound that doesn't necessarily sound better than my electrostatics. Don't get me wrong, these are phenomenal drivers; however, they are just plain difficult to implement sucessfully.
-----------------

[JohnR]

So, how do I know what frequencies will couple and what frequencies will cancel each other out?

Have a go at using this simulator -

http://www.pvconsultants.com/audio/radiation/vpr.htm

[BPT]

FULLRANGEMAN:
The RD75 can be used as an OB. The rear of the driver is internally damped with acoustic foam, so less high frequency sound is emitted from the rear than from the front. The RD50 can be used as a direct replacement for the Carver driver. I have built both a RD75 OB and an OB line array using multiple 4" full range drivers. The 4" plays louder, needs far less watts to boogie and can be crossed over at 100Hz. The RD75 is more coherent, focused, transparent, faster, less coloration.... You can hear what Danny is describing above 2K. The RD75 allows you to hear (and see in your mind's eye) the individual strings resonating on an acoutic guitar or closely miked violin----with the 4"ers, it is not so clear. Cymbals are cleaner, better defined in size/shape with clearer, longer sustain. On good recordings (like the Blue Coast) the RD75 captures the space between performers far better---you can hear a performer, then a defined dead space, then another performer. With the 4"ers, the individual players spaces are not as defined. With the RD75 depth and layering is better as well. On the down side, the RD75 is expensive, requires a notch filter, dramatically rolls off above 10KHz. and can only play at very high levels when crossed over around 300Hz. or higher. You can get around some of these problems using DSP, but that introduces more cost and other problems (phase). No free lunch, here.
Chris H.

[Danny Richie]

FULLRANGEMAN:

Chris is correct on all accounts. I completely agree.

Personally though, I prefer a line of the smaller Neo's. The top end is more extended, but they don't play as low. But I have found that running the big ribbons down low is not an advantage. 300Hz is the heart of the mid-range and crossing in that range with good integration is not easy. With like drivers, it can be done, but with dissimilar drivers with different off sets it is pretty tough. I prefer crossing them in the 850 to 1kHz range and letting a row of small woofer handle up to that point.

The smaller Neo's have much higher sensitivity to. And to me sound cleaner and more dynamic. 

[AK]

I'm thinking, high frequency cancellation may be an advantage for a short array of full range drivers.
most of paper full range drivers have rising frequency response.
say we build sealed or ported enclosure and add just enought drivers to balance low frequency gain and high frequencies cancellation.
of course top end response will be rough, but I bet it will sound better that single full range used without any sort of filter.

[Danny Richie]

I'm thinking, high frequency cancellation may be an advantage for a short array of full range drivers.
most of paper full range drivers have rising frequency response.
say we build sealed or ported enclosure and add just enought drivers to balance low frequency gain and high frequencies cancellation.
of course top end response will be rough, but I bet it will sound better that single full range used without any sort of filter.

Nope, it will have a worse response than a single driver and the dips and peaks will vary with distance and height.

Just let the single full range driver cover the full range and bring the other drivers in to cover lower frequency ranges only.

[FullRangeMan]

Thankyou Danny/BPT for sharing your impressions about the BG RD75, if it have a nice Dipole Soundstage it have some sex appeal to me,
cause it is bigger than the 60'' Carver Ribbon, bigger is always better, as all women knows (sorry).
Iam illeterate in xovers, what is this notch filter after all??
Kinds,  Gustavo

[HT cOz]

Thankyou Danny/BPT for sharing your impressions about the BG RD75, if it have a nice Dipole Soundstage it have some sex appeal to me,
cause it is bigger than the 60'' Carver Ribbon, bigger is always better, as all women knows (sorry).
Iam illeterate in xovers, what is this notch filter after all??
Kinds,  Gustavo

Gustavo,

How can you have an icon saying Use a crossover?  AARGh and also If tweeters look like targets then shoot them and then ask a question like what is notch filter?

It just doesn't add up!!!

[FullRangeMan]

Excuse-me if you like xovers.   I do not know xovers why I do not like then, and mainly tweeters.
I have a friend repair man that like shape the sound modifying the xover, and do not have interest in fullranges...   
What I can do for him? He likes two way pocket monitors!! Patience!
Regards,
(sorry for my explicit banners)

[BPT]

Fullrangeman:
A notch filter is usually a combination of a capacitor, inductor and resistor designed/tuned to  reduce/dampen a resonance in a speaker driver.

I was mistaken before. The RD48 is designed for use with the Carver and Genesis speakers.

Chris H.

[Browntrout]

I don't see how anyone could like a crossover except when comparing one to another. Surely if we can do without then all the better? I don't see why all discussions need to conclude with the same similar consensus.
  I have learnt that you can have the most faithfully reproduced sound that lacks all emotional involvement (most hifi to be honest) but you can allow your mind to wonder and let the emotion of music take the concentrations of your consciousness away and leave you with unperfect sound and perfect involvment.
 If we can see past measured responses I think a better understanding of why we listen will arise. The best performances I have experienced have not been so because I was sat in the correct seat of the hall but the ones where I have formed an emotional response to the music and performer.
  How do we put emotion into the music we make?

[JohnR]

Well said.

Back on topic (this is the open baffle circle...) - Gustavo, one way you go about it might be with nine drivers wired series/parallel for an 8 ohm load (assuming 8 ohm drivers).  And if it doesn't work out, then rewire to run the top driver full range (I am assuming we are talking about smallish drivers here), and rewire the bottom 8 drivers to 4 ohms and drive them with a separate filter and amp. Since this last part is just augmenting the bottom and compensating for dipole rolloff, it doesn't really count as an AAARGH! crossover

Just a thought

[FullRangeMan]

Thankyou for the info BPT!!  >> Who knows do it live! (Brazilian saying).
Someone know what is the Brand or maker of the woofers of the GilmoreAudio (Carver Amazing clones) ?
I ask in amail to Mr. Mark Gilmore, but after 6 months seems he will not reveal...
Thanks,

[Paul Hynes]

Browntrout,

Once again I have to agree with you. This is why I am not particularly keen to mess around with anything that could take away the emotional hit that I get with the B200 line array in my system. I am not concerned how it measures as the presentation has that magic that eludes much audio equipment.

By the way, I like your avatar. It triggers an emotional response to the area that I live in, as the scenery has similarities. There are also several thousand fresh water lochs along the island chain and many of them have a good stock of wild brown trout.

General discussion,

I still think the series/parallel verses parallel wiring issue should be considered further. As I am not commercially active in line array loudspeaker production I have no vested interest either way. I do not have a PHD in Mathematics so do not expect to see any rocket science equations in this post. It is purely an exercise in logical thinking and if I miss any points you think are relevant please feel free to add to the discussion. There will not be any measurements either as I do not have the specialised test equipment required for this.

The reactive impedance of a moving coil drive unit is more complex than just inductance. It consists of resistance together with capacitive and inductive reactance. These parameters are not fixed values, as they are modified by mechanical motion, drive unit loading, driver resonances and temperature. The resulting load impedance and phase angle varies considerably with frequency. The output impedance of the amplifier is in parallel with this varying load impedance and would tend to act as a damping mechanism reducing the effect of the back EMF caused by the reactive loudspeaker load. There is also some self-damping caused by the resistive element of the drive unit impedance.

If two drive units are placed in series the amplifier damping is reduced considerably as each drive unit sees the reactive impedance of the other drive unit between itself and the amplifier. The resistive impedance of the drive units will double, the inductance will double and the capacitance will halve. This will allow the drive unit back EMF to reach a higher amplitude, and exhibit more influence on the resultant acoustic waveform generated by the drive units, compared to the single drive case. As the increased back EMF voltage is time delayed, when it mixes with the following program signal, it is going to smear the waveform more. I would suggest that this should have profound effects and could easily cause increased confusion in time and space in the presented music.

If two drive units are wired in parallel they both see the amplifier damping and the resistive value of the composite drive unit load impedance is halved, the inductive value is also halved and the capacitive value is increased although the reduced parallel resistance would tend to reduce the effect of the capacitance increase a little. The back EMF voltage shows a small reduction in level caused by the reduced inductance and the lower parallel resistance compared to that of a single driver. This is going to have much less effect on the music signal than the series/parallel case.

In line arrays it is logical to conclude that increasing the number of drive units wired in series will exasperate the problems caused by series wiring. Conversely when drive units are wired in parallel the back EMF damping is more optimally maintained, by the amplifier output impedance and the reduced inductance and impedance of the parallel drive units. This should allow better signal integrity.

The difference series/parallel and parallel wiring is easy enough to hear as opnlybafld found out with his basic experiments with his B200 drive units. There seems to be a reluctance to accept this concept and I can understand the hassle factor of re-wiring a line array and building new/more amplifiers, but on a DIY level it is relatively inexpensive to implement. You will never know until you try it.

Incidentally comb filter effects are not confined to line arrays. They are also produced by two or more sources of the same frequency and even one drive unit and the reflections from room boundaries can generate them. Because of this we are all used to comb filter effects in our sound systems, and, as I do not see everyone rushing to dump their loudspeakers in the bin, I would suggest that comb filtering is relatively benign on the grand scale of annoying distortions or perhaps the majority of us have learned to filter the effect out of our minds.

Regards
Paul

[FullRangeMan]

Excelent balanced (pardon the pun) panoramic vison about the issue Mr.Paul.   Maybe a big parallel output  Bryston amp have some utility, after all.
Cheers.

[FullRangeMan]

...dramatically rolls off above 10KHz.

Dear BPT,
...dramatically rolls off above 10KHz. >> I think the human hearing love attenuated treble, at least this ears around me. I work some time with pro-sound and my ears has been fried manys nights by the horns of the stage/PA system, strong treble is awful.
Cheers.

[JohnR]

Incidentally comb filter effects are not confined to line arrays. They are also produced by two or more sources of the same frequency and even one drive unit and the reflections from room boundaries can generate them. Because of this we are all used to comb filter effects in our sound systems, and, as I do not see everyone rushing to dump their loudspeakers in the bin, I would suggest that comb filtering is relatively benign on the grand scale of annoying distortions or perhaps the majority of us have learned to filter the effect out of our minds.

Yes indeed. Yet one might have hoped that in the open baffle circle, where interference between wavefronts is a given, some consideration might have been given to the room and other effects.

[Browntrout]

I think we have been talking about open baffle speakers. Even my non technical post was intended to be on topic. 

[Paul Hynes]

JohnR,

Considering room interaction was part of the design exercise when assessing which design to use for my system loudspeakers. Reducing cabinet interaction was also given consideration as I have a fair amount of experience with loudspeaker cabinet construction. I assumed, rightly or wrongly, that open baffle enthusiasts would be aware of the benefits of removing cabinet interaction from the equation. There is plenty of discussion about this topic in this forum circle. Combining these OB benefits with those of a line array is less well documented in DIY circles as few enthusiasts consider building a line source and fewer still make it open baffle.

I chose to use a full range driver open baffle line source array for a number of reasons. Firstly I do not like using crossovers to split the frequency range into sections and I prefer the musical presentation from full range drivers. This is purely my preference. Feel free to differ. Aside from the sonic benefits I perceive with the full range drivers there are financial benefits, when constructing a line array with full ranger drivers, as you do not need a crossover or more amplifiers. This helps compensate for the cost of the drivers, baffle etc.

Another reason is the theoretical advantage of the way the waveform is propagated into the listening room in the near-field area, which in my room includes the listening area. The OB line exhibits a dipole response, which reduces output to the sides of the array. You can easily hear this by walking over to the array and into the null area. The cylindrical wave front that the array propagates into the room reduces floor and ceiling bounce allowing noticeably less colouration from room interaction. There may be some reflection from the side walls where the cylindrical wave front hits the wall, but I suspect these are delayed enough to not interfere with the initial direct wave front that we use to mentally construct the image of the recorded musical events. The reflections sound similar to acoustic reverberation and add air to the performance, as does the reflection from the wave front propagated from the rear of the array that bounces of the rear wall.

The reason that I chose to use eight inch drivers in the line is that I wanted to be able to move enough air to reproduce big drums without resorting to a sub woofer and the attendant interface issues. I have used eight inch full range drive units in the past, and achieved a satisfactory musical performance with them on open baffles, so I was not too concerned about the frequency extremes of the B200.

I have already gone through my reasons for parallel wiring the drive units. On a DIY basis it is easy enough to build a purpose built amplifier to deal with the low impedance load of parallel drivers, certainly 8 B200s in parallel. All you are really doing is re-scaling the voltage and current relationships in the output stage and ensuring the output stage can handle the required current. There is also a benefit I did not highlight. My line array is quite efficient and requires little power to generate high sound pressure levels. I am currently using a stereo amplifier that is configured for 2.5 watts into 0.75 ohms. The voltage swing required for 2.5 watts into 0.75 ohms is around 2 volts peak. This is the sort of voltage swing you would easily expect from a line preamp. In practice this allows me to use a Lightspeed LDR volume control followed by a high power mosfet source follower. Absolutely no voltage gain is required in these two stages, as all my music source equipment can easily provide enough output to go well beyond this voltage swing requirement. This allows very simple low distortion circuitry and the system is particularly transparent to the musical performance because of it.

If anyone wants to discuss the above further or has anything to add that might be of benefit please feel free to join in.

Regards
Paul