Something I just though of today that I want to try out soon is to put a copper shorted turn around the outside perimeter of a loudspeaker magnet. The purpose is the same as if there was a shorted turn around the centre polepiece like some better quality speakers have, that serves to make the magnetic field appear dynamically "harder" to the voice coil instead of soft and squashy.
If in normal operation the voice coil field pushes against the permanent magnet field it will in fact try and push the field back to zero. The magnet resists this of course but it cannot do so perfectly. Therefore the bigger the magnet the less likely this is to happen very much. As a result, (I expect) Qes and therefore Qts is lower which means the cone really does what the signal is telling it to do.
If we put a short around the outside of the magnet, any moment to moment variations in the magnetic field caused by the voice coil using the magnet as a punching bag ("fwabadda-dabadda" as Don Martin would say) now induce a current into the shorting loop and *this* current tries to stop the field changing in the first place. The dynamic impedance of the magnet is hopefully lowered.
This is probably not the ideal place to put a shorting loop, but the thing is, you can actually try it yourself without taking your speaker apart.
Hmmm.... how about this...? Drive the speaker as normal, and have a *multi-turn* coil of wire around the magnet as before, and put a low value current-sensing resistor in series with the voice-coil and run this current sense signal to a second power amp that drives the add-on coil! By altering the gain of the second power amp you could presumably reduce the magnet's "impedance" to virtually zero. Could a Qes of zero be attained by doing this??
/Circlotron - waits for mrfeedback to blame Melbourne's water again 😉
If in normal operation the voice coil field pushes against the permanent magnet field it will in fact try and push the field back to zero. The magnet resists this of course but it cannot do so perfectly. Therefore the bigger the magnet the less likely this is to happen very much. As a result, (I expect) Qes and therefore Qts is lower which means the cone really does what the signal is telling it to do.
If we put a short around the outside of the magnet, any moment to moment variations in the magnetic field caused by the voice coil using the magnet as a punching bag ("fwabadda-dabadda" as Don Martin would say) now induce a current into the shorting loop and *this* current tries to stop the field changing in the first place. The dynamic impedance of the magnet is hopefully lowered.
This is probably not the ideal place to put a shorting loop, but the thing is, you can actually try it yourself without taking your speaker apart.
Hmmm.... how about this...? Drive the speaker as normal, and have a *multi-turn* coil of wire around the magnet as before, and put a low value current-sensing resistor in series with the voice-coil and run this current sense signal to a second power amp that drives the add-on coil! By altering the gain of the second power amp you could presumably reduce the magnet's "impedance" to virtually zero. Could a Qes of zero be attained by doing this??
/Circlotron - waits for mrfeedback to blame Melbourne's water again 😉
Melbourne Aquavit ?..........
Hi Graham,
The other day I dropped into a car audio shop to have a listen to what the youngsters are listening to nowadays - unimpressive to my ears.
Anyway I saw a Sony 2 way speaker with the woofer crossover inductor wound around the perimeter of the woofer magnet - I did not get to listen to these however.
I have thought about fitting a copper band around the perimeter of the magnet myself - oneday I will try it to see if any improvement/change.
Years ago I saw the magnet assembly out of a very old small speaker and this had a copper disc glued to the top of the polepiece (under the dustcap).
Perhaps a silver disc would be useful and economical enough instead.
I have tried glueing other (exotic) materials to the back of magnet assemblies and this has effected very nice improvements.
Eric / - Never drinks Melbourne water, not even Fosters.
Hi Graham,
The other day I dropped into a car audio shop to have a listen to what the youngsters are listening to nowadays - unimpressive to my ears.
Anyway I saw a Sony 2 way speaker with the woofer crossover inductor wound around the perimeter of the woofer magnet - I did not get to listen to these however.
I have thought about fitting a copper band around the perimeter of the magnet myself - oneday I will try it to see if any improvement/change.
Years ago I saw the magnet assembly out of a very old small speaker and this had a copper disc glued to the top of the polepiece (under the dustcap).
Perhaps a silver disc would be useful and economical enough instead.
I have tried glueing other (exotic) materials to the back of magnet assemblies and this has effected very nice improvements.
Eric / - Never drinks Melbourne water, not even Fosters.
Further thinking, better idea...
After 24 hours mulling over this idea of mine, I now think that a better way would be to wind a coil with the same number of turns as the voice coil around the centre pole of the magnet assembly (yeah, kinda difficult 🙄 ) with as thick a size wire as will fit, and push it right to the very back so it is against the inside of the back plate. Glue it so it cant move around. Also make sure that the voice coil can't touch it at it's fully inward position.
Drill a pair of small holes in the back plate to allow the ends of the coil to exit. Now simply put this magic 😉 coil in series with the voice coil and you are ready to go! When the polarity is correct I think the resonant peak will be not as high as before, and away from resonance the cone p/p movement should be slightly greater for a given drive voltage because the permanent magnet appears to be stronger. Also there will be no eddy-current losses from the coil around the solid centre core because this would happen only if the flux was changing, and we are trying to do is to reduce that (undesireable) change.
*********
Imagine yourself sitting in a small boat on the water. You reach out and take hold of a nearby pier and alternately push away/ pull toward it. Trouble is, the legs of the pier bend to and fro causing the pier to move slightly in the direction you are pushing it instead of causing you and your boat to move the exact amount you want. The pier is speaker magnet of course, and magnetism is all soft and squashy, how much so depends on the strength of the magnet. This is what we have to fix, to make the squashy less squashy.
Well, where all the speaker experts when you want them?? Hello?
After 24 hours mulling over this idea of mine, I now think that a better way would be to wind a coil with the same number of turns as the voice coil around the centre pole of the magnet assembly (yeah, kinda difficult 🙄 ) with as thick a size wire as will fit, and push it right to the very back so it is against the inside of the back plate. Glue it so it cant move around. Also make sure that the voice coil can't touch it at it's fully inward position.
Drill a pair of small holes in the back plate to allow the ends of the coil to exit. Now simply put this magic 😉 coil in series with the voice coil and you are ready to go! When the polarity is correct I think the resonant peak will be not as high as before, and away from resonance the cone p/p movement should be slightly greater for a given drive voltage because the permanent magnet appears to be stronger. Also there will be no eddy-current losses from the coil around the solid centre core because this would happen only if the flux was changing, and we are trying to do is to reduce that (undesireable) change.
*********
Imagine yourself sitting in a small boat on the water. You reach out and take hold of a nearby pier and alternately push away/ pull toward it. Trouble is, the legs of the pier bend to and fro causing the pier to move slightly in the direction you are pushing it instead of causing you and your boat to move the exact amount you want. The pier is speaker magnet of course, and magnetism is all soft and squashy, how much so depends on the strength of the magnet. This is what we have to fix, to make the squashy less squashy.
Well, where all the speaker experts when you want them?? Hello?
Graham
The top plate and center pole piece have both magnetic resistance, adding more squashiness to the squashy magnet. So the best place for shorting rings is indeed close to the voice coil.
I could imagine that it would be worth to try what Eric suggested, i.e. a copper band around the magnet (which has to be soldered properly in order to act as a real shorting ring).
My imagination tells me that this wouldn't reduce Qes but it could decrease 2nd order distortion somewhat.
Regards
Charles
BTW: In the meantime I have found the AES article I mentioned in the "afterburner" thread. Are you still interested in this ?
The top plate and center pole piece have both magnetic resistance, adding more squashiness to the squashy magnet. So the best place for shorting rings is indeed close to the voice coil.
I could imagine that it would be worth to try what Eric suggested, i.e. a copper band around the magnet (which has to be soldered properly in order to act as a real shorting ring).
My imagination tells me that this wouldn't reduce Qes but it could decrease 2nd order distortion somewhat.
Regards
Charles
BTW: In the meantime I have found the AES article I mentioned in the "afterburner" thread. Are you still interested in this ?
Thanks Charles, but I can only manage six fads at any one time 😉
Seriously though, I aren't looking in that direction nowadays. But thanks anyway!
Seriously though, I aren't looking in that direction nowadays. But thanks anyway!
Current-dumping afterburner for a gainclone. 😱
http://www.diyaudio.com/forums/showthread.php?s=&threadid=15742&highlight=
http://www.diyaudio.com/forums/showthread.php?s=&threadid=15742&highlight=
- Status
- Not open for further replies.