This table should shed some light on the matter. These figures, which I borrowed from magnetapplications.com show that Neodymium suffers far more from temperature related temporary losses in strength than any other type commonly used in speakers when presented with an opposing magnetic field (HC), thus explaining its much lower maximum service temperature than all other types listed and the extensive efforts to dissipate heat from them in loudspeaker applications. Br is each magnet's maximum residual flux density in the absence of an external magnetic field. Alnico, btw, leads the group in both of these categories, although it is more susceptible to permanent flux losses (until remagnetized) than the other types when presented with an extremely high opposing HC.
Reversible Temperature Coefficients of Br and HC
Material....Tc of Br..Tc of HC..Curie Point....Maximum Service Temp.
NdFeB........-0.12......-0.6......310 (590)...........150 (302)
SmCo.........-0.04......-0.3......750 (1382).........300 (572)
Alnico.........-0.02.......0.01....860 (1580).........540 (1004)
Ceramic......-0.2.........0.3......460 (860)...........300 (572)
Perhaps SmCo has not been so widely used for speakers because once the cobalt is paid for alnico gives better performance as long as the speaker is not being greatly abused physically or by overpowering. Mud magnets may have earned their poor reputation partly because of their significant sensitivities in opposite directions in Br *and* HC which probably would tend to create a less coherent sonic presentation than most other magnet types when significant amounts of power are applied to the vc.
Reversible Temperature Coefficients of Br and HC
Material....Tc of Br..Tc of HC..Curie Point....Maximum Service Temp.
NdFeB........-0.12......-0.6......310 (590)...........150 (302)
SmCo.........-0.04......-0.3......750 (1382).........300 (572)
Alnico.........-0.02.......0.01....860 (1580).........540 (1004)
Ceramic......-0.2.........0.3......460 (860)...........300 (572)
Perhaps SmCo has not been so widely used for speakers because once the cobalt is paid for alnico gives better performance as long as the speaker is not being greatly abused physically or by overpowering. Mud magnets may have earned their poor reputation partly because of their significant sensitivities in opposite directions in Br *and* HC which probably would tend to create a less coherent sonic presentation than most other magnet types when significant amounts of power are applied to the vc.
SH grade neo is typically rated to 150 deg C, but EH grade is good to 200 deg C. Both of which are really overkill; to get the magnet that hot your voice coil is at 200+ deg C, which is well beyond the point at which the glues holding it together will fail. In other words, the voice coil will fall apart well before you thermally demagnetize modern SH+ grade neo.
The old L and M grades weren't too good, as they started losing magnetic force at just 100 deg C, which is within the realm of what high power voice coils can withstand.
Neo is like any other magnet material - it has its strengths and weaknesses. It does offer some significant advantages in motor size (weight and physical dimensions), and typical pot-core neo motors are inherently shielded as well. If those advantages are desireable, then use it. Otherwise it may be lower cost to go with ferrite.
Dan Wiggins
Adire Audio
The old L and M grades weren't too good, as they started losing magnetic force at just 100 deg C, which is within the realm of what high power voice coils can withstand.
Neo is like any other magnet material - it has its strengths and weaknesses. It does offer some significant advantages in motor size (weight and physical dimensions), and typical pot-core neo motors are inherently shielded as well. If those advantages are desireable, then use it. Otherwise it may be lower cost to go with ferrite.
Dan Wiggins
Adire Audio
My concerns with neodymium temperature sensitivity would begin long before the speaker would fail a smoke test because of sonic instability resulting from driver parameter shift caused by squirrely magnet system performance. I wouldn't want a heavy bass line knocking a driver's HF sensitivity down even a quarter decibel when the magnet warms just because of any design compromise related to the magnetic material. Even a 30-40C rise will result in some pretty undesirable stuff happening to Neo magnetic performance.
Personally, I wouldn't sweat it. The thermal capacity of the magnet structure is usually so far beyond what the voice coil can take that you will smoke the VC well before the magnet heats up enough to start losing strength.
In my experience, 10-30 seconds of +200 deg C on the voice coil is a death sentence (even with the ultra-high temp epoxy units used in car audio). It takes a few minutes of such delivered power to heat the magnet structure anywhere near that level.
For a 0.25 dB drop, worry more about the heat of the voice coil. That corresponds to ~6% increase in DCR from temperature. With the Tc of Cu being around 0.4%/deg C, that means an increase of the voice coil temperature by ~15 deg C will drop you 0.25 dB. Start at a warm ambient temperature of 30 deg C (warm club environment). You've lost 0.25 dB at 45 deg C on the voice coil from simple heating, well before what the magnet would have any issue.
I'd be more worried about the ability of the magnet to hold up under flux; demagnetization from flux modulation is an actual issue, and one of the reasons why AlNiCo magnets should be avoided for higher powers. The coercive force of Neo is orders of magnitude higher than AlNiCo or ferrite. Even Ferrite is considerably better than AlNiCo (by 2-5X). You can see high instantaneous flux levels from large peaks that can cause significant permanent loss in magnet strength.
Dan Wiggins
Adire Audio
In my experience, 10-30 seconds of +200 deg C on the voice coil is a death sentence (even with the ultra-high temp epoxy units used in car audio). It takes a few minutes of such delivered power to heat the magnet structure anywhere near that level.
For a 0.25 dB drop, worry more about the heat of the voice coil. That corresponds to ~6% increase in DCR from temperature. With the Tc of Cu being around 0.4%/deg C, that means an increase of the voice coil temperature by ~15 deg C will drop you 0.25 dB. Start at a warm ambient temperature of 30 deg C (warm club environment). You've lost 0.25 dB at 45 deg C on the voice coil from simple heating, well before what the magnet would have any issue.
I'd be more worried about the ability of the magnet to hold up under flux; demagnetization from flux modulation is an actual issue, and one of the reasons why AlNiCo magnets should be avoided for higher powers. The coercive force of Neo is orders of magnitude higher than AlNiCo or ferrite. Even Ferrite is considerably better than AlNiCo (by 2-5X). You can see high instantaneous flux levels from large peaks that can cause significant permanent loss in magnet strength.
Dan Wiggins
Adire Audio
Actually, Alnico is the least susceptible of the magnetic materials I have listed to flux modulation. The table makes clear that while Alnico is probably not what you want to place your bets on for high SPL stadium installations, it is still well ahead of neodymium and totally outclasses ceramic for the highest performance sound quality applications where knuckle dragging abuse is not part of the sonic equation.
Speaking for myself, I'm willing to hold my knuckles off the floor to get the best SQ. And intelligent magnet system design, such as, say with an underhung VC, rather quickly obviates that 'problem' with alnico IAC.
VC resistive heating is not as pernicious in its effects IAC, as magnetic system temperature sensitivity because it has no component that corresponds to flux modulation.
Speaking for myself, I'm willing to hold my knuckles off the floor to get the best SQ. And intelligent magnet system design, such as, say with an underhung VC, rather quickly obviates that 'problem' with alnico IAC.
VC resistive heating is not as pernicious in its effects IAC, as magnetic system temperature sensitivity because it has no component that corresponds to flux modulation.
The Aura drivers seems to be nice, however compared to the Excel drivers I fail to see the superior performance from these drivers at least by looking at the manufacturers graphs.
Also if these discrete steps in magnet flux exists shouldn´t they show up in a distortion plot then? Meaning in the end the most linear driver will be the best no matter if ceramic or Neo are being used.
The W15CH do not measure better than the old W15CY. Aura woofers don´t seem to measure better than W26/W22 (within physical limits). And the new Accuton Neo tweeters do not seem to measure better than the Diamonds which I believe uses ceramic magnets.
Maybe everything we need to know can be found in bandwith, Q, decay/waterfall and HM/IM plots..?
/Peter
Also if these discrete steps in magnet flux exists shouldn´t they show up in a distortion plot then? Meaning in the end the most linear driver will be the best no matter if ceramic or Neo are being used.
The W15CH do not measure better than the old W15CY. Aura woofers don´t seem to measure better than W26/W22 (within physical limits). And the new Accuton Neo tweeters do not seem to measure better than the Diamonds which I believe uses ceramic magnets.
Maybe everything we need to know can be found in bandwith, Q, decay/waterfall and HM/IM plots..?
/Peter
Re: Re: If neodymium is all that great . . .
I haven't and couldn't because I couldn't find reference to it via Google. Could you give me a link?
That may be so but they have apparently traded excursion for sensitivity. When you cross at 90 to 110Hz (a rather good bi-amping crossover frequency IMO) as I am doing you don't need all that much excursion even with demanding recorded material - 1/2" pk/pk is way plenty for a 90db, 5 incher. I'd rather have the sensitivity.
Anyway, I've read a lot of opinions about neo here. I was sorta looking for the facts but, as a friend used to say, oh well.
usekgb said:
I haven't and couldn't because I couldn't find reference to it via Google. Could you give me a link?
Timn8ter said:
That may be so but they have apparently traded excursion for sensitivity. When you cross at 90 to 110Hz (a rather good bi-amping crossover frequency IMO) as I am doing you don't need all that much excursion even with demanding recorded material - 1/2" pk/pk is way plenty for a 90db, 5 incher. I'd rather have the sensitivity.
Anyway, I've read a lot of opinions about neo here. I was sorta looking for the facts but, as a friend used to say, oh well.
Pan said:
I'm not sure what you're getting at, if only because Aura and Seas Excel don't have at all comparable product lines. At least, I don't know of any 1" - 4" wideband Excel drivers that cost $15-$35 or hugely expensive "statement" Excel subwoofers. Nor do I know of any Aura NRT tweeters or 5"-10" "statement" Aura midwoofers.
The closest comparison is the Aura NS8 to the W22, but even there I think the targets for each driver are so different that comparison makes about as much sense as comparing the W22 to that sweet-looking new neo-magnet 8" pro driver from B&C.
Re: Re: Re: If neodymium is all that great . . .
Originally, you said this:
A lot of discussion ensued about the relative advantages of neo magnets compared to others, also cost constraints, etc - maybe not an 'answer' but informative. Some examples of speakers that use neo magnets has been given (Adire, Morel, many tweets), but you sort of dismissed them for some reason, though it is an answer to your original question. Next you said this:
What 'facts' are you looking for, exactly? Dan Wiggins of Adire added some good info about neo magnets (in particular, look at the text I added bold to):
This seems to be your answer. Any design entails choices and compromises, nothing is perfect. The fact that neo magnets AREN'T used is instructive, as is looking at where they are used (car applications, tweets, A/V, etc). In spite of your unreferenced claims, it may be that neo magnets aren't used that much because their higher cost isn't justified except in specific applications (ie, they don't sound better or they would be used a lot, regardless the cost).
Two other links to peruse:
LDSG Motors
LDSG Magnets
It is pretty clear that the motor is mostly responsible for the sound, magnets clearly of secondary importance.
Originally, you said this:
Bill Fitzpatrick said:
A lot of discussion ensued about the relative advantages of neo magnets compared to others, also cost constraints, etc - maybe not an 'answer' but informative. Some examples of speakers that use neo magnets has been given (Adire, Morel, many tweets), but you sort of dismissed them for some reason, though it is an answer to your original question. Next you said this:
Bill Fitzpatrick said:
What 'facts' are you looking for, exactly? Dan Wiggins of Adire added some good info about neo magnets (in particular, look at the text I added bold to):
DanWiggins said:
This seems to be your answer. Any design entails choices and compromises, nothing is perfect. The fact that neo magnets AREN'T used is instructive, as is looking at where they are used (car applications, tweets, A/V, etc). In spite of your unreferenced claims, it may be that neo magnets aren't used that much because their higher cost isn't justified except in specific applications (ie, they don't sound better or they would be used a lot, regardless the cost).
Two other links to peruse:
LDSG Motors
LDSG Magnets
It is pretty clear that the motor is mostly responsible for the sound, magnets clearly of secondary importance.
Re: Re: Re: If neodymium is all that great . . .
Sorry Bill, I don't have a link for you. This is the driver that JBL uses in many of their new high power PA speakers. This is in their 12" line array, the SRX712M, and their new mini-line. I have a fair amount of experience with JBL's latest Neo drivers, and I am very pleased with them. Sorry I couldn't give you more info.
Cheers,
Zach
Bill Fitzpatrick said:
Sorry Bill, I don't have a link for you. This is the driver that JBL uses in many of their new high power PA speakers. This is in their 12" line array, the SRX712M, and their new mini-line. I have a fair amount of experience with JBL's latest Neo drivers, and I am very pleased with them. Sorry I couldn't give you more info.
Cheers,
Zach
BF wrote,
"Anyway, I've read a lot of opinions about neo here. I was sorta looking for the facts but, as a friend used to say, oh well."
I´m so sorry I contaminated your pure neo-thread. Just tell me to leave with my nonsense and I´m out starting my own "no facts only speculations and opinions neo thread".
Or maybe you can lighten up a bit..
/Peter
"Anyway, I've read a lot of opinions about neo here. I was sorta looking for the facts but, as a friend used to say, oh well."
I´m so sorry I contaminated your pure neo-thread. Just tell me to leave with my nonsense and I´m out starting my own "no facts only speculations and opinions neo thread".
Or maybe you can lighten up a bit..
/Peter
Pallas said:
As far as I know there are a 10" Aura woofer and also a 10" Excel woofer. If you look at the graphs published at the companys homepage you´ll see that the performance are not superior on the Aura woofer. Even the W22 has better upper bass than the Aura woofers. I also said "withing physicall limits" meaning that obviously the 18" Aura has less distortion than the W26 at high spl when W26 run out of excursion.
However for many listeners the performance at low to medium spl is what counts and (again) judging by the manufactureres measurements there are no superior perofrmance from the UH neo approach in this case.
Please also take my other points about other drivers into consideration.
Now, was it so hard to understand really?
By a quick glance on the web there is nothing that indicates a neo based motor as being inherently better than a ferrite dito.
/Peter
Maybe I miss something but... the original issue discussed resolution performance. Can anyone deduct resolution performance (and dynamic compression etc) from frequency response graphs and manufacturer's spec sheets alone? I don't think so - different Q's, some peaks here and there etc., will be addressed by the eventual speaker designer but say nothing much about resolution and transparency. They only tell you how to use the driver in a speaker. The power of the engine and the size of the brakes tell you nothing about a car's handling.
Manufacturer's specs usually include neither linear distortion / energy storage data (important for transparency), nor dynamic compression data, nor resolution data. In fact you're lucky if you get THD data. So I really can't imagine how to judge on these parameters from (mostly in all likelihood) a single on-axis FR graph.
Manufacturer's specs usually include neither linear distortion / energy storage data (important for transparency), nor dynamic compression data, nor resolution data. In fact you're lucky if you get THD data. So I really can't imagine how to judge on these parameters from (mostly in all likelihood) a single on-axis FR graph.
thoriated said:
Dan was refering to permanent demagnitization, not AC flux modulation. Alnico is resistant to flux modulation so long as the current through the voice coil is within a certain range, but go beyond that range and you get permanent demagnitization...well permanent is not the right word exactly, you lose magnetization until you put the driver in a magnetizer and give it a good zap. If you've got a vintage alnico driver, chances are it could do with a zap.
John
MBK,
Yes I think you missed some things.
"the original issue discussed resolution performance."
Resolution and distortion is related don´t you think?
"Can anyone deduct resolution performance (and dynamic compression etc) from frequency response graphs and manufacturer's spec sheets alone?"
I don´t thihnk anyone ever claimed that but to some degree yes, if the measuremetns are good and detailed enough. Resolution needs low HD/ID distortion, flat FR and fast decay.
"So I really can't imagine how to judge on these parameters from (mostly in all likelihood) a single on-axis FR graph."
Neither can I and no one ever claimed they could.
/Peter
Yes I think you missed some things.
"the original issue discussed resolution performance."
Resolution and distortion is related don´t you think?
"Can anyone deduct resolution performance (and dynamic compression etc) from frequency response graphs and manufacturer's spec sheets alone?"
I don´t thihnk anyone ever claimed that but to some degree yes, if the measuremetns are good and detailed enough. Resolution needs low HD/ID distortion, flat FR and fast decay.
"So I really can't imagine how to judge on these parameters from (mostly in all likelihood) a single on-axis FR graph."
Neither can I and no one ever claimed they could.
/Peter
Hello Bill,
these issues have ensued discussions on this forum before, here is the link:
http://www.diyaudio.com/forums/showthread.php?threadid=35936&perpage=10&pagenumber=6
Cheers
these issues have ensued discussions on this forum before, here is the link:
http://www.diyaudio.com/forums/showthread.php?threadid=35936&perpage=10&pagenumber=6
Cheers
thoriated posted:
There's a pretty good paper on magnets at Arnold Magnetics where they even point out that the coercive force of AlNiCo is low, meaning it is susceptible to demagnetization:
http://www.arnoldmagnetics.com/mtc/pdf/TN_0205.pdf
tomtt posted:
Of course, steel is typically about as good ast AlNiCo itself (~2 MS/m) and copper (~60 MS/m) or aluminum (~35 MS/m) are really the way to go for shorting rings.
If you're relying on your magnet to make your shorting ring, it's really not a good thing - there are MUCH better ways to go, since even the "best" magnet from an electrical conductivity standpoint is about equal to the steel in your motor.
hancock posted:
And of course the amount of magnet volume will also affect the coercive force - how well the magnet resists demagnetization. Volumetrically, AlNiCo is pretty good against ferrite, because you get much more force from a given volume of AlNiCo (or Neo) as compared to ferrite. Of course, if you're talking flux output, then we typically need 10X the volume of ferrite, and that really equalizes the way the magnets resist flux modulation.
Dan Wiggins
Adire Audio
Hmmm... The coercive force is what indicates resistance to demagnetization. Typical AlNiCo is in the 1200-1900 range. Ferrite is 3500-4500. Neo is 12,000-14,000. AlNiCo is a lot easier to demagnetize than neo or ferrite. I know we run a LOT lower charging voltages when demagnetizing AlNiCo magnets.
There's a pretty good paper on magnets at Arnold Magnetics where they even point out that the coercive force of AlNiCo is low, meaning it is susceptible to demagnetization:
http://www.arnoldmagnetics.com/mtc/pdf/TN_0205.pdf
tomtt posted:
I think Ragnar was referring to the electrical conductivity of the magnet itself working as a shorting ring. Ceramics (ferrites) are much worse than neo or AlNiCo in this area! Typical N40 grade neo is around 700 kS/m, while AlNiCo 6 is ~3 times that (around 2.2 MS/m). Ceramics are much worse, down around 60 kS/m.
Of course, steel is typically about as good ast AlNiCo itself (~2 MS/m) and copper (~60 MS/m) or aluminum (~35 MS/m) are really the way to go for shorting rings.
If you're relying on your magnet to make your shorting ring, it's really not a good thing - there are MUCH better ways to go, since even the "best" magnet from an electrical conductivity standpoint is about equal to the steel in your motor.
hancock posted:
Bingo. Ferrite will start to bend a bit earlier than AlNiCo, but you can't get it fully over without toasting the voice coil. AlNiCo, on the other hand, holds up fairly well but push it just a bit to far and it's gone.
And of course the amount of magnet volume will also affect the coercive force - how well the magnet resists demagnetization. Volumetrically, AlNiCo is pretty good against ferrite, because you get much more force from a given volume of AlNiCo (or Neo) as compared to ferrite. Of course, if you're talking flux output, then we typically need 10X the volume of ferrite, and that really equalizes the way the magnets resist flux modulation.
Dan Wiggins
Adire Audio
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.