Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Multi-Way Conventional loudspeakers with crossovers

Help me choose components for my nude dipole build
Help me choose components for my nude dipole build
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 27th June 2019, 10:55 AM   #11
StabMe is offline StabMe
diyAudio Member
 
Join Date: Apr 2017
Agree on a 4-way, Ju.

Now i just have to choose what to use as an HF dipole.

Mundorf amt17d22 is out of question as of now due to its prohibitive cost.



Quote:
Originally Posted by Juhazi View Post
One really important thing about dipoles above 2kHz is, if the backside radiation symmetrical to frontside. Manufacturers dont give that data, but you can guess it by looking. Cones/dynamic drivers are hopeless in this regard and I have a doubt about AMTs.

Dipole cancellation below the dipole peak overrules most wiggles in response, and gives 6dB/oct slope

Every dipole radiator has a limit up where in frequency the dipole action goes. Above 4kHz very small details of the frame become significant. With my AINOgradients I struggled with 3" cones too long before I understood this and purchased the B&G Neo8PDFR.

And yes, it really must be a 4-way! Each driver has only 2-3 octaves of smooth and consistent dipole radiation pattern without excessive distortion.

Attached is Peerless NE95 responses. More details in my thread.
  Reply With Quote
Old 27th June 2019, 12:47 PM   #12
CharlieLaub is offline CharlieLaub  United States
diyAudio Member
 
Join Date: Mar 2007
Location: Michigan
Quote:
Originally Posted by StabMe View Post
Here is a photorealistic representation of how it will look like. Please, do not confuse with the real thing - it is just a mock up. I know it looks almost like a real thing!

Click the image to open in full size.

Do you guys thing i am risking some interference from the sides of the frame which is not vertical?
I am working on a system much like what you are building, with different variations, with an emphasis on symmetric dipolar radiation to both the front AND the rear. The rear radiation is mostly ignored and most people just assume it is like the front, but except for low frequencies this is often not the case unless you use planar drivers like the B&G Neo model, Mundorf, etc.

In any case, what you are doing with the frame looks great. But there is one flaw that I see in your cartoon of the setup: I worry you have not taken into account the center of mass of the drivers. It's NOT on the mounting flange! The magnet is a significant portion of the mass, and it is several inches behind. If you hang the driver via the mounting holes the weight distribution tips it "upwards". The more massive the magnet, the further back the center of mass. Only if you hang the driver by the center of mass will it hang straight.

I found a solution to this problem. You find some sturdy long bolts. Tightly install two of these in the top two mounting holes on either side of the center one (in your pic, at 10 and 2 on the clock face like for the midrange). Then you find a point near the rear of the bolt that lies on the center of mass. When you hang the driver from this point, it hangs straight. These bolts must support the entire weight of the driver, so if your midrange has a large ferrite magnet this will put a lot of strain on them. Unless you can find another point to hang the driver that is on the center of mass, this is the best way to do it. I can hang a 15" Neo magnet driver like this no problem for example.

Another thing you should consider is how to run the lead wires to the drivers. I have found that when the driver is smaller/lighter, the wire itself will tend to pull on the freely swinging assembly. You may run into this with the Neo3PDR. One solution is to bring a stiff wire (like solid core) up to 2' away from the driver and then use some bare tinsel lead or base stranded wire to bridge the gap. You need to work out something one way or another that allows the driver to hang undisturbed.

Yet another issue is alignment of the acoustic centers of the drivers, especially for the tweeter, midrange, and the midwoofer (the H-frame alignment can just be approximate). Just as with a boxed speaker, misaligned acoustic radiation will lead to large phase lead/lag issues. You cannot use delay to fix this problem with a dipole, however, because by "moving" the apparent acoustic center (e.g. back with delay) you are simultaneously moving the acoustic center frontwards when looking from the rear. You have to physically align the acoustic centers of all drivers in a dipole system or what will happen is that to the front you will have good integration between drivers around the crossover points but this will be screwed up to the rear. This would be less of a problem with e.g. the Neo10+Neo3 because they are slim and alignment is almost automatic, but between a planar driver and cone driver you need to find the acoustic center and check alignment from both front and rear vantage points. The AC is usually located back towards the rear of the cone e.g near the dust cap. A quarter of an inch is a significant delay (e.g. phase lag) at frequencies above 1kHz. Do not overlook this important aspect of the dipole system and its crossover design/driver integration.

You brought up the question about whether the frame side members will cause any interference problems. I have been making measurements on drivers within a frame like this. I found that as long as there are at least 4" of free space on either side (more is better) between the driver and the hanging frame members it doesn't really show up in the measurement, even above 60 degrees off axis horizontally. With enough free room on the sides, the dipole can form and the frame falls in the acoustic shadow of the dipole null at 90 degrees. As long as the thickness (front to back) of the frame is not more than a couple of inches and the frame is on the order of 20" wide you can ignore it. With larger drivers used at lower frequencies this is less of a concern, it's more for frequencies above 800Hz or so.

I see you listed the Beyma TPL-150 as a possibility, crossed at 2kHz. Is that a dipole? For some reason I thought it was not... If not, I would definitely not cross from a dipole system to a monopole tweeter at 2k. This is best done at 5kHz or higher, or just use a dipole tweeter.

In my system, I have tried to use cone drivers for all but the tweeter. If you cannot source a planar midrange (e.g. the Neo10) I can give you some suggestions via PM on drivers that will work well in this application. I have made a bunch of measurements on various drivers, front and rear, including the distortion profile, and know of a couple of good candidates for this role that work well as a dipole. Many times the magnet structure and basket will cause problems in the rear response when you get to around 1kHz, and it is difficult to tell a priori without buying and measuring. I would not suggest using a 12" up to 2kHz as a dipole without measuring the rear response first. But the same problems happens with drivers of all sizes, and smaller is not necessarily better in this regard. What you want is a driver that can be crossed at 2kHz, since this is about the low end of a dipole tweeter's range.

Anyway, I am encouraged to see you take this path with your system. I originally suggested this kind of design on this site here:
Multi-way OB

Finally, what materials are you planning to use for the "frame" for the hanging drivers? This is one problem I have been wrestling with. I'm not talented at woodworking. Hardwood tends to warp/twist. I have a mockup done with 2x4 lumber and that is already a problem. Plywood could work, but it has to support all the hanging weight and resist sway and twist. A metal frame seems like a good idea, but I fear that it will resonate (that "ping" metal sound you get when you tap on it). My frame is straight up and down, so I will have fewer structural concerns to worry about, and IMO you should think about it carefully.
__________________
Visit my Audio Web Page <<--CLICK TO LEARN MORE-->> Get my LADSPA plugins

Last edited by CharlieLaub; 27th June 2019 at 12:51 PM.
  Reply With Quote
Old 27th June 2019, 02:23 PM   #13
StabMe is offline StabMe
diyAudio Member
 
Join Date: Apr 2017
Hello, Charlie! Thanks for chiming in.

It is your ideas that you shared across different posts on this forum and even a PDF file with a whole approach outlined that got me interested in making a symmetric dipole system with nude hanging drivers.

Quote:
Originally Posted by CharlieLaub View Post
I am working on a system much like what you are building, with different variations, with an emphasis on symmetric dipolar radiation to both the front AND the rear. The rear radiation is mostly ignored and most people just assume it is like the front, but except for low frequencies this is often not the case unless you use planar drivers like the B&G Neo model, Mundorf, etc.

In any case, what you are doing with the frame looks great. But there is one flaw that I see in your cartoon of the setup: I worry you have not taken into account the center of mass of the drivers. It's NOT on the mounting flange! The magnet is a significant portion of the mass, and it is several inches behind. If you hang the driver via the mounting holes the weight distribution tips it "upwards". The more massive the magnet, the further back the center of mass. Only if you hang the driver by the center of mass will it hang straight.

I found a solution to this problem. You find some sturdy long bolts. Tightly install two of these in the top two mounting holes on either side of the center one (in your pic, at 10 and 2 on the clock face like for the midrange). Then you find a point near the rear of the bolt that lies on the center of mass. When you hang the driver from this point, it hangs straight. These bolts must support the entire weight of the driver, so if your midrange has a large ferrite magnet this will put a lot of strain on them. Unless you can find another point to hang the driver that is on the center of mass, this is the best way to do it. I can hang a 15" Neo magnet driver like this no problem for example.
Yeah, i understood that hung by the mounting wholes, a heavy driver would tilt upwards. I was thinking about hanging it both by the mounting holes AND by the magnet so that it kinda balances the weight and make the driver hang by the straight line. Your solution sounds good, i am just not sure if it is applicable for a 12-incher.

Quote:
Another thing you should consider is how to run the lead wires to the drivers. I have found that when the driver is smaller/lighter, the wire itself will tend to pull on the freely swinging assembly. You may run into this with the Neo3PDR. One solution is to bring a stiff wire (like solid core) up to 2' away from the driver and then use some bare tinsel lead or base stranded wire to bridge the gap. You need to work out something one way or another that allows the driver to hang undisturbed.
Charlie, i can order a thin and narrow CNCed frame made of wood in which Neo8 and Neo3 will be put and hanged as a single unit. This way such unit becomes heavier and won't be disturbed by wires. Or do you think vibrations from Neo8 will affect the sound of Neo3?


Quote:
Yet another issue is alignment of the acoustic centers of the drivers, especially for the tweeter, midrange, and the midwoofer (the H-frame alignment can just be approximate). Just as with a boxed speaker, misaligned acoustic radiation will lead to large phase lead/lag issues. You cannot use delay to fix this problem with a dipole, however, because by "moving" the apparent acoustic center (e.g. back with delay) you are simultaneously moving the acoustic center frontwards when looking from the rear. You have to physically align the acoustic centers of all drivers in a dipole system or what will happen is that to the front you will have good integration between drivers around the crossover points but this will be screwed up to the rear. This would be less of a problem with e.g. the Neo10+Neo3 because they are slim and alignment is almost automatic, but between a planar driver and cone driver you need to find the acoustic center and check alignment from both front and rear vantage points. The AC is usually located back towards the rear of the cone e.g near the dust cap. A quarter of an inch is a significant delay (e.g. phase lag) at frequencies above 1kHz. Do not overlook this important aspect of the dipole system and its crossover design/driver integration.

My idea was to use metal rods parallel to the floor on which to hang midrange and tweeter. I will then put the mic in the listening position and move the drivers until impulse responses are aligned.

Quote:
You brought up the question about whether the frame side members will cause any interference problems. I have been making measurements on drivers within a frame like this. I found that as long as there are at least 4" of free space on either side (more is better) between the driver and the hanging frame members it doesn't really show up in the measurement, even above 60 degrees off axis horizontally. With enough free room on the sides, the dipole can form and the frame falls in the acoustic shadow of the dipole null at 90 degrees. As long as the thickness (front to back) of the frame is not more than a couple of inches and the frame is on the order of 20" wide you can ignore it. With larger drivers used at lower frequencies this is less of a concern, it's more for frequencies above 800Hz or so.
That's good to hear. I am still contemplating on what material to use for the frame. I am not good at woodwork (basically zero skills), so i was thinking just hiring somebody locally to make a frame out of metal profile. Steel, Alu? I will add bracing where it is appropriate and where it won't interfere with the sound.

Quote:
I see you listed the Beyma TPL-150 as a possibility, crossed at 2kHz. Is that a dipole? For some reason I thought it was not... If not, I would definitely not cross from a dipole system to a monopole tweeter at 2k. This is best done at 5kHz or higher, or just use a dipole tweeter.

StigErik used TPL-150 with the back cover removed. I should have mentioned that too. It is just a contemplation, though. But i think i will skip on TPL since i have already ordered Neo8 and i need something for HF part. Will be considering options mentioned in this thread. Neo3's quality is a concern


Quote:
In my system, I have tried to use cone drivers for all but the tweeter. If you cannot source a planar midrange (e.g. the Neo10) I can give you some suggestions via PM on drivers that will work well in this application. I have made a bunch of measurements on various drivers, front and rear, including the distortion profile, and know of a couple of good candidates for this role that work well as a dipole. Many times the magnet structure and basket will cause problems in the rear response when you get to around 1kHz, and it is difficult to tell a priori without buying and measuring. I would not suggest using a 12" up to 2kHz as a dipole without measuring the rear response first. But the same problems happens with drivers of all sizes, and smaller is not necessarily better in this regard. What you want is a driver that can be crossed at 2kHz, since this is about the low end of a dipole tweeter's range.
As i mentioned earlier, i have already bought planar midrange - BG Neo8. It will be crossed to 12G40 at about 700-800Hz and planar HF driver. I hope, i can get dipole radiation can be guaranteed this way. Do you think 12G40 will maintain dipole pattern up to 700-800Hz?


Quote:
Anyway, I am encouraged to see you take this path with your system. I originally suggested this kind of design on this site here:
Multi-way OB
This post is basically where is started



Quote:
Finally, what materials are you planning to use for the "frame" for the hanging drivers? This is one problem I have been wrestling with. I'm not talented at woodworking. Hardwood tends to warp/twist. I have a mockup done with 2x4 lumber and that is already a problem. Plywood could work, but it has to support all the hanging weight and resist sway and twist. A metal frame seems like a good idea, but I fear that it will resonate (that "ping" metal sound you get when you tap on it). My frame is straight up and down, so I will have fewer structural concerns to worry about, and IMO you should think about it carefully.
I am leaning towards a metal frame. A flipped over T frame with horizontal bracing at the top to hang the drivers, in the middle between midrange and midwoofer and at the bottom will make it more stable. I have an option to weld frame members to each other, then i will have them painted balck, i think. A wood frame with mesh to make it look a bit better is also an option. As for the resonating sounds - i think an aluminium frame is a better choice?


Thank you for such a thorough reply, Charlie.
  Reply With Quote
Old 27th June 2019, 09:10 PM   #14
StabMe is offline StabMe
diyAudio Member
 
Join Date: Apr 2017
Any thoughts on HiVi RT2C-A? Possible to remove the back cover and make it a true dipole? Can have a pair locally.

Same for Fountek NeoX 1 or 3 - those can be had for a reasonable price here. Not sure if one can remove a back cover for dipole application.

Last edited by StabMe; 27th June 2019 at 09:18 PM.
  Reply With Quote
Old 27th June 2019, 10:54 PM   #15
evanc is online now evanc  United States
diyAudio Member
 
Join Date: Sep 2009
Location: New Jersey. About 1 hour from NYC and 1 min. from the beach
Help me choose components for my nude dipole build
Quote:
Originally Posted by StabMe View Post
Hey Evan!

Well, i wanted a dipole system which will exhibit similar rear radiation with what it has in the front. ...... from what i can see in the datasheet, it would more like an xo at about 1.5Khz?
I bet you are correct on both points. I worked with what I had and am very happy with the results.

If the tweeters are reasonable price how about one on facing forward and one facing backwards to achieve dipole radiation? (this is what I have done)
__________________
http://www.evancotler.com
  Reply With Quote
Old 28th June 2019, 04:04 AM   #16
ScottG is offline ScottG  United States
diyAudio Member
 
Join Date: Feb 2003
Location: US
Quote:
Originally Posted by StabMe View Post
Any thoughts on HiVi RT2C-A? Possible to remove the back cover and make it a true dipole? Can have a pair locally.

Same for Fountek NeoX 1 or 3 - those can be had for a reasonable price here. Not sure if one can remove a back cover for dipole application.

No.

However,

Beston RT002A-Dipolar Ribbon Tweeter 6 Ohm


Beston RT002A-Dipolar Ribbon Tweeter 6 Ohm | eBay
__________________
perspective is everything

Last edited by ScottG; 28th June 2019 at 04:06 AM.
  Reply With Quote
Old 28th June 2019, 04:15 AM   #17
ScottG is offline ScottG  United States
diyAudio Member
 
Join Date: Feb 2003
Location: US
Quote:
Originally Posted by evanc View Post

If the tweeters are reasonable price how about one on facing forward and one facing backwards to achieve dipole radiation? (this is what I have done)
Normally not a great idea, and it sort of defeats the no-enclosure sound.. BUT at higher freq.s with a cheap AMT (with very little distance separation between the two drivers): it might work well (crossover dependent).

Doubles the cost though.

You can also double-horn load a good inexpensive compression driver like the one from Peerless. Radiation isn't going to be quite the same because the front has a phase-path, but below the top octave for a 1" exit driver it shouldn't be that dissimilar with the equal horns (front and back). Horns can also provide a dispersion pattern within their bandwidth that can be more controlled with more dipole-ish pattern.
__________________
perspective is everything
  Reply With Quote
Old 28th June 2019, 01:11 PM   #18
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
Here is the legendary, classic thread of dipole tweeters, OP by Rudol Finke! Double monopoles tested too. The problem comes always from separation of sources vs. short wavelength.

On the directivity of dipole tweeters

Linkwitz and Kreskowsky too must have accepted this is mission impossible. A rearside tweeter adds late reflections similar to the pure dipole, so phase summation is irrelevant above 6kHz anyway, they say. And I do support that too.

Tweeter integration
__________________
Radikal aktivist AINOgradient speaker project
  Reply With Quote
Old 1st July 2019, 10:08 PM   #19
StabMe is offline StabMe
diyAudio Member
 
Join Date: Apr 2017
As i mentioned before, I have chosen a woofer driver for my dipole build - 18Sound 15LW1401 4Ohm, which shows promising 106dB at 60Hz in Hornresp simulation with its being a limiting factor 9mm x-max in an H-Frame.

Now, Hornresp does provide data with max SPL and the overall FR image and according to some correspondence between David McBean and some other guy from the forum Hornresp's h-frame models (modeled as compounded horns) are a lot like those provided by MJK MathCad models.

The only thing that i don't see in the Hornresp model is the resonance. Just out of my head i used and H frame with 25cm wings which give 50cm total depth. To make it look symmetric, H and D are also 50cm. Will this thing play up to about 180-200Hz? Aside from MJKs spreadsheets, what is the way to find out?
  Reply With Quote
Old 2nd July 2019, 12:26 PM   #20
StabMe is offline StabMe
diyAudio Member
 
Join Date: Apr 2017
Hmmm...

Seems like i am misunderstanding something about modeling h-frame in Hornresp. Two models in 2xPi:

H-Frame with 25cm wings from each side:

Help me choose components for my nude dipole build-1-jpg

Its calculated combined power response with MaxSPL function:

Help me choose components for my nude dipole build-2-jpg

Nude driver (compounded horn with 0,1cm wings):

Help me choose components for my nude dipole build-3-jpg

Its calculated combined power response with MaxSPL function:

Help me choose components for my nude dipole build-4-jpg

The upper part of the response changes, yes, but not the low end extension.

Am i missing something? Any other better way to model H-Frame?
Attached Images
File Type: jpg 1.jpg (18.8 KB, 89 views)
File Type: jpg 2.jpg (75.1 KB, 87 views)
File Type: jpg 3.jpg (37.7 KB, 80 views)
File Type: jpg 4.jpg (75.4 KB, 79 views)
  Reply With Quote

Reply


Help me choose components for my nude dipole buildHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Help choose a guitar amp to build? wicked1 Instruments and Amps 36 1st December 2014 08:34 AM
the three way nude swinging dipole thread twest820 Multi-Way 134 23rd January 2013 01:19 AM
Please help me choose my next amplifier build! ByronInPortland Solid State 1 21st October 2012 11:38 PM
What would you choose for Quad ESL-63, dipole or TL? tf1216 Planars & Exotics 19 14th June 2010 08:55 PM


New To Site? Need Help?

All times are GMT. The time now is 11:02 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki