The copper ring acts as a 'shorted turn' for any modulating magnetic
flux in the pole piece, which causes rise in voice coil inductance.
Aluminium can also be used but it has to be thicker to have the same
effectiveness. A good example is the Acoustic Elegance drivers which
have a thin copper sleeve over the pole piece but also thick Al shorting ring
on the outside magnet - both to the same effect. These drivers have
amongst the lowest VC inductance I have seen - so it seems to work well.
Another example is the large copper sheet straps you see on old power
transformers (often used in valve amps). The copper sheet is wrapped
around the outside of the laminations and also acts as a shorted turn
for external fields and supposedly reduces them.
cheers
Terry
You own graph proves that as an effective measure for magnetic shielding, copper stinks. Orders of magnitude poorer and all but ineffective. This is due to copper's lack of ferromagnetic properties. Eddy current losses as an engineered mechanism to create a magnetic shielding effect is pathetic.
Aren't there distortion reduction considerations as well in as much as the effect will be reduce the magnitude of the air gap flux modulation? That is, reducing variation of BxL during operation.
Electrical fields are related to voltage. Magnetic fields are related to current. In a device where voltage is high and current is low, it would be expected that as a better conductor of electricity, aluminum would be much more effective than steel as an electrical field shield. How effective steel is at shielding a region of space from high magnetic fields under short circuit conditions depends in part on the steel alloy being used. Only magnetically soft materials that are low carbon usually aluminum killed steels with low coercivity and very high saturation such as permalloy, supermalloy, and mu metal are good for shielding. The B-H curve for a particular alloy is a good indicator. Saturation should be in the range of 15,000 gauss and the hysterisis loop tall and thin indicating high saturation and easily mobile reversal of polarity of the magnetic domains. Openings in the enclosure allow flux to divert back into free space. As Amuneal's tutorial says, up to five times the diameter of the opening. The effective design of a magnetic shield requires experience in that art, not a one time trial and error experiment. Look at Maxwell's equations again. Mu zero is the magnetic coercivity of free space. Mu for the material being used is what is important. The test is to measure magnetic flux with and without different sheilds in the area being shielded. This is the only way to determine if that is the variable which affects what you are trying to achieve in the way of magnetic shielding.
Here is an example of a driver with an AL shorting ring:http://www.vifa.de/chassis/peerless/pdf/SLSW10-3908.pdf
The shorting ring minimises the change of inductance of the voicecoil over throw. Ideally the inductance shoud be constant over all positions of the voicecoil that are under Xmax,
P-P. It supresses two effects : One is that the force on the membrane is more when the inductance is high ( i.e. the force is low when the inductance is low ) and the other effects is that the driver will have a changing frequency respose over throw. The less the inductance the more extention in the upper region. In an uncompensated drives that happens when the coil moves away from the center position because it sees less iron.
That modulation of L gives a very unpleasant, shrill sounding distortion so a good driver shoud have a shorting ring. Terry has already mentioned that copper and aluminum can be combined to advantage and it is more economical then heavy copper alone.
BL changes are usually tackled in another way. That has to do with the shape and material of the polepiece and the magnet.
The shorting ring minimises the change of inductance of the voicecoil over throw. Ideally the inductance shoud be constant over all positions of the voicecoil that are under Xmax,
P-P. It supresses two effects : One is that the force on the membrane is more when the inductance is high ( i.e. the force is low when the inductance is low ) and the other effects is that the driver will have a changing frequency respose over throw. The less the inductance the more extention in the upper region. In an uncompensated drives that happens when the coil moves away from the center position because it sees less iron.
That modulation of L gives a very unpleasant, shrill sounding distortion so a good driver shoud have a shorting ring. Terry has already mentioned that copper and aluminum can be combined to advantage and it is more economical then heavy copper alone.
BL changes are usually tackled in another way. That has to do with the shape and material of the polepiece and the magnet.
A bit off topic but I believe the distortion mechanism, or one of them is signal
modulated VC inductance.
T
Calculation of low-frequency magnetic shielding of a substation using a two-dimensional finite-element method
The case in question is thicker than the skin depth for that material at the frequency of interest, I assume. SEs charts are for thin materials.
rgds
jms
The case in question is thicker than the skin depth for that material at the frequency of interest, I assume. SEs charts are for thin materials.
rgds
jms
The effective design of a magnetic shield is very difficult. Amateur efforts invariably fail. The $100,000 shielding of the 4000 amp 480 volt bus I referenced in an earlier post even using mu metal was not satisfactory. There were too many alternate flux paths around it. Steel chosen at random probably will not work well even if the geometry of the shield is correct. For example, at the oppostite extreme from mu metal is stainless steel which like aluminum has no ferromagnetic properties to speak of. Despite the iron atoms in the material, the chromium atoms in close proximity negate the external ferromagnetic effect. Try a refrigerator magnet against stainless steel flatware and see.
The superiority of steel as a magnetic shield can be easily demonstrated with a strong refrigerator magnet, a thin sheet of steel such as the cover of a steel food can, and equivalent thicknesses of paper and of folded aluminum foil. The paper and aluminum foil offer no resistance to the magnetic flux of the magnet attracting the steel refrigerator (assuming it is not a stainless steel refrigerator), all it offers is the distance of its thickness. The food can lid will reduce or eliminate the force of the magnet on the refrigerator when placed between them by diverting the magnetic flux into the steel can lid effectively creating the magnetic equivalent of a short circuit.
I never said copper or aluminum were as good as ferromagnetic materials for the same thickness.
I was simply pointing out that they CAN provide shielding from magnetic fields.
You stated that the decoupling that Ed measured between the two coils was entirely due to shielding from the electric field. That's clearly not the case as aluminum does provide some shielding of magnetic fields.
se
Current flowing in wire as the result of a difference in electrical potential produces both electrical and magnetic fields. Measuring the change in electrical performace resulting from shielding and then ascribing the change by inferring it one or the other is just guesswork and is usually wrong. The only way to know is to measure both. Magnetic fields are measured with a gaussmeter which is designed to measure magnetic fields and only magnetic fields.
Electrical fields are related to voltage. Magnetic fields are related to current. In a device where voltage is high and current is low, it would be expected that as a better conductor of electricity, aluminum would be much more effective than steel as an electrical field shield. How effective steel is at shielding a region of space from high magnetic fields under short circuit conditions depends in part on the steel alloy being used. Only magnetically soft materials that are low carbon usually aluminum killed steels with low coercivity and very high saturation such as permalloy, supermalloy, and mu metal are good for shielding. The B-H curve for a particular alloy is a good indicator. Saturation should be in the range of 15,000 gauss and the hysterisis loop tall and thin indicating high saturation and easily mobile reversal of polarity of the magnetic domains. Openings in the enclosure allow flux to divert back into free space. As Amuneal's tutorial says, up to five times the diameter of the opening. The effective design of a magnetic shield requires experience in that art, not a one time trial and error experiment. Look at Maxwell's equations again. Mu zero is the magnetic coercivity of free space. Mu for the material being used is what is important. The test is to measure magnetic flux with and without different sheilds in the area being shielded. This is the only way to determine if that is the variable which affects what you are trying to achieve in the way of magnetic shielding.
Thanks for your valuable advices. I do not need to go for Maxwell equations. Maybe real condition example would help.
We have 63kA 50Hz short circuit current. This makes dI/dt = 28A/us. From slope of current you get magnetic field intensity H = 4.45 million A/m at 1m distance from conductor (standard measuring condition). This, for relative permeability = 1 makes magnetic induction B = 5,6T (Tesla). For a ferromagnetic material with relative permeability of 15000 you get magnetic induction 84000 T. Would it work linear, what do you think.
Unfortunately he didn't supply graphs for both conditions
rgds
jms
I sent that to Ben. It was a passive L/C filter to enhance a SoundTech 1700 distortion analyzer. The coil was air core and a pair of diagonal cutters brought near created increase in thirds at the -140dB level. The same effect was caused by the steel bench top I was on. I had to put my fixture up on wooden blocks.
Thanks all of you guys, for the enlightening explanations for those copper / alu / combined pole-piece rings.
jan didden
Haven't you heard, John? According to Banquer and Khomenko, neither Ott nor Morrison know what they're talking about.
se
- Status
- Not open for further replies.