Build One Planar Speaker - sourcing parts

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Long time reader, but not poster, to the DIYaudio. For the last week I've been reading up on Planar speakers, and I'd like to start by building one speaker. I build guitar tube amps and effects, so I'm experienced with electronics and diy, but just not with Planar. My goal is to build one speaker "just to see" how it performs for electric guitar. It would be nice if the planar were around 12" x 12" with a usable frequency response of 100Hz to 4kHz (or better).

I'm fine with building the thing, I've read all the posts here related to building planars, and, while there isn't a 1, 2, 3 step tutorial anywhere, I believe I understand the principles. What I don't know is the dimensions of the magnets to buy. If someone could tell me, "buy 200 of ____ model", and tell me how to orientate them, then I'd be ready to go.

Thank you for your help.

CHAD
 
25 x 6 x 1.5 mm N45 Ni-Cu-Ni magnets from e-Magnets UK - Supplier of Neodymium Rare Earth Magnets Sold 100 a time in plastic container. I used A4 size 12" X 8" with a 8" x 41/2" cutout for the magnets. Placed horizontally. NSNSNS ETC. 4 columns wide and about 140 magnets altogether. Or can make the cutout smaller to fit fewer if required. Your choice! Run the tape between the magnets backwards and forwards until the end. Check for forward movement of the diaphragm with battery, so if you make another one you can get them in phase. Have fun, ask if not sure.
 
When buying magnets try to get the most material you can for your money as Cost per Cubic Inch (ci) or Cubic Centimeter (cc) for whatever grade you decide to use.

If you want use Neo's and find a really good price, remember that the smaller sized magnets are harder to work with, especially when you are trying to fasten them down end to end with the same poles pushed together. ;)

When you stack magnets for more strength it is the same as wiring Batteries (voltage sources) in series.
And every time you double the stack (thickness) you double the magnetic flux strength ( or lines of forces).

This is why I say to get the most volume of material for your dollar, providing it is of a workable size for your particular project.

I only know of 2 or 3 really good suppliers here in the states when it comes to prices, and a Google search will bring them up for you.
I have posted such links in another thread but I don't have them handy at this time.

FWIW

jer :)
 
Last edited:
25 x 6 x 1.5 mm N45 Ni-Cu-Ni magnets from e-Magnets UK - Supplier of Neodymium Rare Earth Magnets Sold 100 a time in plastic container. I used A4 size 12" X 8" with a 8" x 41/2" cutout for the magnets. Placed horizontally. NSNSNS ETC. 4 columns wide and about 140 magnets altogether. Or can make the cutout smaller to fit fewer if required. Your choice! Run the tape between the magnets backwards and forwards until the end. Check for forward movement of the diaphragm with battery, so if you make another one you can get them in phase. Have fun, ask if not sure.

See, this is where I run into problems. I've read many of your posts with magnet suggestions and have yet to find the exact magnet referenced. Can you post the specific part number or link to the exact page on emagnets? There are so many options.

I feel like my wife sending me to the store to by a box of cereal. It sounds easy until you walk down the isle. Healthy? Tasty? Chocolate? Rice? Wheat? Frosted? Hot or cold?

CHAD
 
You would need about 7 or 8 of those packages to have enough for a 12"x 12" size panel.

This is assuming 15 rows spaced 10mm apart and 30mm long.

This should give you enough room to possibly use up to 4 traces of voice coil in between each row of magnets.

This should give a decent efficiency and a usable resistance range for your amplifier.

jer :)
 
They could be layers if each layer is insulated from the other or they could be side by side, just in the Magnetostatic design.
As it is an Apogee clone it has two traces per section.

Just make sure that all of the separate circuits (traces or Voice Coils) are in series and not in parallel or else it will be no different then having just one trace.

jer :)
 
Last edited:
I only ever have 1 continuous so called trace as all my builds (over 80) are full range, and I never have space between my magnets NS. Most of my neos are 50 x 12.7 x 3 mm, and my ferrites are 50 x 18 to 24 mm x 10 mm. I have used 50 x 12.7 x 6 mm neos for a epsilon layout design. As you have found out all my pictures are on different audio sites on Facebook.
 
Sorry to be picky but the trapezoidal shape only changes the distribution of resonance frequencies. It doesn't reduce resonance---even trapezoidal drumheads resonate. In the case of speakers the resonances are thought to be less intrusive if the they don't overlap in frequency and the trapezoidal shape is intended to reduce the overlap.
Few
 
To further lower the Q of the resonance, shouldn´t any of the membrane´s sides be parallel to another side?

What material are you planing to use anyway?

I am not sure what kind of material, but depending on size, I have plenty of large mylar ESD bags for electrical equipment storage.

Speaking of size (and not knowing if this will perform well enough to buy $120 in magnets), I may want to start with a very smallish speaker, with perhaps 40 or 60 magnets... 8 inches x 6 inches maybe. Is 500Hz a reasonable -6db LF guesstimate?

CHAD
 
ESD bags may be a little thick, A mylar of .5 mil or thinner is more desirable.

However, I have used such materials such as Monokote and other model airplane coverings as well on some of my earliest models and it worked very good.

You need a film that has a good temperature and mechanical stability factor such as some PET films are good for (Mylar or similar), with Kapton being the best, however it is quite costly.

As the voice coil heats up this can/will reduce the tension of the diaphragm depending on how much power you pump into it.
This may or may not be an issue for you.

Most all Model Airplane coverings are PET films.

On the Trapeziodal shape,

No, it does not eliminate the resonance's but it does distribute them over a very slightly wider range depending on the actual dimensions.

Therefore instead of all of the energy being focused on one particular frequency, it is distributed among a few of them instead on just one, thus reducing the amount of resonance (however slight it may be) for that one particular frequency.
But it does not eliminate them as a whole.
The most likely effect will be the lowering the Q factor for that lowest frequency of diaphragm resonance in question.

I may be wrong in my explanation, but that is the best of how I understand it.

I had planned on doing some experiments on using some very irregular shapes such as that of an egg or even an extremely long triangular shapes.

But then I realized that an even and constant horizontal dispersion factor is not possible throughout the vertical span with such shapes.

Even upon my investigation of this subject I have found little evidence that it makes any significant amount of difference in the sound to begin with.

So, It is on the back burner list of my things to worry about.

The Trapezoidal shape diaphragm that the Apogee Duette's have is only a difference of 3", for 11" on top, and, 14" bottom for a 47" length.

This is not enough of a difference to justify any difference in the 6db roll off.

However, The outside casing is twice as wide at about 24 1/2" at the top and 28 1/2" at the bottom and this will shift the roll off down one more octave at best.

There have been many many discussions that can be found in these threads on this subject and one just recently on the distributed resonance methods that Sound Lab uses on their ESL design's.

The physical width of the diaphragm of the design is much more of a concern as this will determine your horizontal dispersion characteristics for your operating range of frequency's.
The Beaming effect!! ;)

If it is to be run as a Dipole than the overall dimensions come into play as well and then you have to deal with the 6db roll off as previously mentioned.

I think (know) that once you get a working version made you will be astounded at the overall clarity of such a design, I have done this with both Magnetic and ESL versions in the past.

FWIW

jer :)

P.S. Even the thin Plastic tops on your common TV dinner is a good cheap source of Mylar (or PET film of some type) in small quantity's!! ;)
 
Last edited:
Thanks, jer!

Here are my views, I think that they are the same as yours. But please correct me if I am wrong, I'm still learning.

fs:
Provided that the compliance is the same, it is actually the mass difference that makes the Q of the fs lower, if that difference can be considered "local" enough. That is, how stiff is the membrane.
As I am doing a bass membrane, it is not only the fs itself but the excitation of the frame construction that I want to be more evenly, and thus lowered, spread.

Dipole roll-off:
If the baffle size isn't reduced in order to make the trapezoid shape, it could be good to lower the Q of that effect as well.
I usually have different sized sides on the back of the baffle, not too long though - don't want to make it into a U-baffle. Those sides can be irregular shaped as well.

Edge/Beaming:
If the trapezoid shape of the membrane manifests on the baffle as well, then the Q of the edge effect also will be lower and it will not be so abrupt. If the membrane is large I think this is the main cause for having a trapezoid shape.

All in all, for me, a trapezoid shape of the membrane and baffle is a must but it is not worth it if it not easily built.
I will try 70 cm at the bottom and 60 cm on the top for a 85 cm high membrane. Picture later in my thread.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.