Planar magnetic with central tweeter? - Page 4 - diyAudio
 Planar magnetic with central tweeter?
 User Name Stay logged in? Password
 Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read Search

 Planars & Exotics ESL's, planars, and alternative technologies

 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
diyAudio Member

Join Date: Apr 2004
Location: Maine, USA
In order to make it easier to compare the magnetic field's strength, orientation, and homogeneity for different magnet spacings I drew up a FEMM model with three different spacings. I'm still using NdFeB magnets that have 1/4" x 1/4" cross sections. The attached figure shows magnet spacings of 1/2", 3/8", and 1/4" (reading from the left to the right side of the diagrams).

I also calculated the absolute value of the component of the field strength that actually moves the diaphragm---the component of the field that lies parallel to the diaphragm. I assumed the diaphragm and conductors would be 1/8" above the faces of the magnets. The upper graph shows the result.

It's interesting to see the saddles in the graph for the 1/2" spacing case (left end of the diagram). That means that near the edges of the magnets the field lines aren't oriented in the right direction, but they're so closely spaced that the desirable component of the field is still largest there.

For the sake of argument, let's assume 1/4" wide conductors. My take on the upper diagram is that you're better off fitting two runs of those conductors above magnets spaced by 1/2" rather than a single run over the 1/4" gap. The field strength in the 1/2" case is more than half of what it is in the 1/4" case, and by doubling the conductor length you get double the force. There's also a larger radiating area. All this adds up to higher sensitivity and more acoustically open area behind the diaphragm. Can anyone confirm or shoot down this reasoning? I realize some method for ending up with the same impedance in both cases will be necessary in order for this simple analysis to apply. I'm also working with a fixed number of magnets, in order to compare approaches with similar cost.

I hope the graphs make some sense...

Few
Attached Images
 Composite graphic of field lines and tangential mag.jpg (247.2 KB, 215 views)

 31st January 2012, 03:20 AM #32 diyAudio Member   Join Date: Nov 2006 Location: Indiana Blog Entries: 1 Two conductors, each with half the current? I would try going down to 2 mm, or whatever it takes to get the field to be parallel to the membrane, everyplace there will be any part of a conductor. Otherwise, the force on the current will not be perpendicular to the membrane. Is there some way to set the resolution, in FEMM? It seems like the results should be more uniform. So maybe we're not getting a true picture. What are the units of the field strength, in your plot? I have a DC Gaussmeter with a very small directional probe, and will go try to measure what the field strength is at a few points on my Magnepan MG-12 speakers, just as a frame of reference. Will report back.
 31st January 2012, 04:04 AM #33 diyAudio Member   Join Date: Nov 2006 Location: Indiana Blog Entries: 1 OK. I put the probe's sensor tip against the mylar membrane, so it was measuring the maximum reading, which looked like it occurred when it was seeing the field that was parallel to the membrane. I measured both the tweeter section and the mid/bass section. I measured both when just barely at the membrane and also when pushing the membrane until it was bottomed-out against the magnets. For the tweeter section, I measured below the foil, where there was no foil but there were still magnets. Then I measured in the center of one of the strips of aluminum foil and the measurement was the same. For the mid/bass section, I had to place the probe tip against the side of one of the aluminum wires. With a thin piece (about 0.5 mm) of very-slightly magnetized steel, I tested the probe of the HangZhou BST-200 Teslameter (Digital Gaussmeter). I used the probe with its protective cover removed, so I could use the tip and orient it as needed. The tip is a piece of what looks like PCB material, about 1 mm thick and 3 mm wide, and protrudes about 40 mm from the probe body. At the tip of the protrusion, there is a Hall-effect sensor. I used the edge of the slightly-magnetized thin steel sheet to see where on the probe tip the measurement was occuring, and the peak is in the first millimeter of it, although the first two millimeters respond much more than farther down the protrusion. The first 2 mm are black and look like the actual measauring component. It may be that the measurements below are the average for 1mm back from the actual tip of the probe. I have no way to know. The manufacturer's website is http://www.hzmagnet.com, according to the label on the back of the unit, which I purchased new on ebay. When measuring at the surface of the mylar, if I carefully turned the probe tip 90 degrees, the reading went to zero. So it looks like it is nicely directional. Going another 90 degrees got to the same reading as before, except with the opposite sign. I used what is labeled as the "200 mT" scale, on the meter, which should have a resolution of 0.1 mT, according to the manual. So I just assumed that the units were mT. I could be wrong. 1 mT = 10 Gs, according to the manual. Magnepan MG-12/QR speaker, magnetic field measurements: TWEETER: Membrane at rest: 35 mT Membrane pushed in, touching magnets: 70 mT Membrane distance from magnets: about 1 mm MID/BASS: Membrane at rest: 30 mT Membrane pushed in, touching magnets: 80 mT Membrane distance from magnets: about 3 mm Last edited by gootee; 31st January 2012 at 04:34 AM.
 31st January 2012, 03:20 PM #34 diyAudio Member   Join Date: Apr 2004 Location: Maine, USA Thanks for the real world magnetic field measurements. Very helpful. The units in the graph I posted are Tesla so it looks like the field in the system I modeled would be quite a bit higher than that in the Magneplanars--maybe by a factor of 6 to 8. If I were to reduce the diaphragm-to-magnet distance a bit that factor might ease a bit higher but I don't yet know how much diaphragm displacement I'll need. The point I was trying to make regarding the tangential field strength is that even if the field lines run in some direction other than parallel to the diaphragm they can still contribute usefully to the desired diaphragm motion. If they have a significant component that is parallel to the diaphragm, that component counts. I understand that your concern is that forces in the wrong direction could cause diaphragm deformation rather than semi-pistonic motion. I see conflicting evidence in that regard. I don't doubt the validity of the underlying argument, the question is how things play out in practice. The ongoing discussion in the Analysis Epsilon thread seems to suggest conductors located in less-than-textbook positions are advantageous. I wouldn't want to put a conductor directly over the face of a magnet where the field lines run entirely in the wrong direction, but I'm beginning to think they need not be confined to regions where the field lines run precisely parallel to the diaphragm. I also want to point out that even if the conductors are confined to the ideally oriented part of the field, the parts of the diaphragm that have no conductor are not going to be driven. The result will be alternating regions of driven and undriven diaphragm which could lead to the same rippling or puckering type of diaphragm deformation described above. If anyone can point to some measurements that help shed light on this question I'd be very interested.
 31st January 2012, 03:27 PM #35 diyAudio Member   Join Date: Apr 2004 Location: Maine, USA Hi gootee, I'm having trouble with the hzmagnet link and my searches aren't bringing anything up for the company. I didn't find anything after a quick ebay search either. Do you have any other info that might help put me on the right trail? Thanks. Few
diyAudio Member

Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
 Originally Posted by Few Thanks for providing the link to your ESL mode experience. I was going to try to dig that out as evidence that the drum head modes really do matter but I'm glad you beat me to it. It's too bad resonant systems are so easy to make resonate!
Hello Few,

Not sure if you had seen the attached patent, but thought you might find it useful or at least interesting.
Basically, the patent is for the technique of damping membrane modes in a planar ribbon by strategic variation in the % open area of the plates holding the magnets.
Looks like you need to use pretty low % open area to provide adequate damping though, so maybe not particularly useful for mid/tweeters.
Attached Files
 Thigpen_damping_6760462.pdf (175.4 KB, 34 views)

 1st February 2012, 10:24 PM #37 diyAudio Member   Join Date: Apr 2004 Location: Maine, USA Thanks for the patent link. Reminds me of your exploration of ESL diaphragm damping. The table showing higher SPLs from the same maximum displacement after damping the modes is particularly interesting. I agree that the price paid in the high frequencies in order to damp the low frequency modes may be too high. One of the reasons I'm hoping to get satisfactory performance from a single-ended design rather than a symmetric one using two parallel arrays of magnets is I like the idea of minimizing the obstructions between the diaphragm and the listener. I'm not sure if you come out ahead with this approach but I guess I'll find out. At least the cost of the magnets is considerably less! Some old Infinity EMIT planar magnetic tweeter literature shows much improved cumulative spectral decay (CSD) behavior after adding some fibrous damping between the diaphragm and the magnets, or between the diaphragm and the plate supporting the magnets--can't remember which. ---- Actually, while trying to track down the diagram I was remembering, I ran across this, which describes Infinity's use of laminated diaphragms to control their break-up. I don't think I remembered that feature of their design; maybe others are already well aware of it. They ended up with quite thick diaphragms but report clean CSD results. Few
diyAudio Member

Join Date: Nov 2006
Location: Indiana
Blog Entries: 1
Quote:
 Originally Posted by Few Hi gootee, I'm having trouble with the hzmagnet link and my searches aren't bringing anything up for the company. I didn't find anything after a quick ebay search either. Do you have any other info that might help put me on the right trail? Thanks. Few
This looks like the correct website:

Wholesale BST200 Gaussmeter - Hangzhou BST Magnet Co.,Ltd.

And here are a lot more:

This ebay one (below) looks fairly close. Not the same but looks like the same probe. The black part with the rounded end can be unscrewed and removed, to expose the small flat probe.

DC/STATIC MAGNETIC FIELD/TESLA/GAUSS METER/TESTER,AC | eBay

Also, do a search for gaussmeter on ebay. Just watch out for all of the ones that don't go down to DC!

Last edited by gootee; 2nd February 2012 at 05:29 AM.

 2nd February 2012, 05:55 AM #39 diyAudio Member   Join Date: Nov 2006 Location: Indiana Blog Entries: 1 How would we determine the optimum gap between the mylar and the magnets, for given magnets and given the range of amplifiers that might ever be used? There are equations for the force on the conductors: Magnetic Force on a Current-Carrying Wire Magnetic field - Wikipedia, the free encyclopedia And there are equations for the static and dynamic position and motion of the diaphragm (won't go into that here). And there are also constraints (or at least one constraint): If we had a testbed in which the diaphragm-magnet gap was adjustable, we could apply the expected "max upper limit" DC current in the direction that pulled the diaphragm toward the magnets and adjust the gap so the two weren't touching. We could use an adjustable DC current source to determine the max upper current limit for a particular gap size. We could use an adjustable DC current to easily measure the basic static linearity of the membrane displacement versus the current. AC tones could be could be used to get measurements of dynamic linearity of displacement versus amplitude at various frequencies. Software like Arta and a measurement microphone could help us understand the sensitivity of distortion versus the non-linearity. Maybe ideally the gap should even be adjustable, depending on average signal amplitude, to get the most-accurate sound reproduction at lower volumes but also allow for much larger displacements for high volume settings. Just thinking out loud. Last edited by gootee; 2nd February 2012 at 06:06 AM.
 3rd February 2012, 01:30 AM #40 diyAudio Member   Join Date: Apr 2004 Location: Maine, USA Thanks very much for the links. I just found a source of Hall effect sensors (Allegro) willing to send up to 25 samples so I've emailed the local distributor. It looks like you apply 5 VDC and ground to two pins and measure the voltage on the third. Even I can handle that. The problem will be zeroing the DC offset (they provide a rough calibration curve in mV/gauss). I may start with that approach if the free samples come through. I'll definitely check out the links, though. I fear that DC current-based measurements may lead us astray, depending on what we think we'll learn from them. Resonances within the driver's bandwidth (or even outside that bandwidth if driven acoustically, as Bolserst pointed out) are likely to lead to larger displacements than the rest of the frequency range and DC measurements won't reflect that effect. I think AC signals will be necessary to measure displacement limitations. I do agree an adjustable diaphragm-to-magnet distance would be handy. I hope to build a system with interchangeable spacers for that reason. Thanks again for the links. Few

 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 OffTrackbacks are Off Pingbacks are Off Refbacks are Off Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post baffleman Planars & Exotics 14 18th April 2014 05:01 PM WrineX Planars & Exotics 71 9th August 2012 02:48 PM rubicon Swap Meet 8 1st December 2010 08:22 PM Borat Planars & Exotics 40 8th September 2009 03:47 PM luvdunhill Swap Meet 9 7th December 2008 04:51 PM

 New To Site? Need Help?

All times are GMT. The time now is 02:58 PM.