I’ve read many threads about ferrofluid used in tweeters. Still, I cannot conclude the fact about it. Moreover, the more I read, the more I am confused. So, I decided to frankly ask here again for clarification and being useful for people who are reading this in 2023.
What’s the effect of tweeters having dried out ferrofluid? Will they create “dull” or “brighter” sound?
Some threads suggested it should be “dull” as the ferrofluid also has some magnetic property, loosing it will decrease total magnetic strength of the driver—summation of magnetic power between permanent magnet and ferrofluid. So the sensitivity of the driver is decreased. HOWEVER, some people stated that the ferrofluid is the “damping” material, loosing it will increase the sensitivity of the driver.
Please clarify the fact
What’s the effect of tweeters having dried out ferrofluid? Will they create “dull” or “brighter” sound?
Some threads suggested it should be “dull” as the ferrofluid also has some magnetic property, loosing it will decrease total magnetic strength of the driver—summation of magnetic power between permanent magnet and ferrofluid. So the sensitivity of the driver is decreased. HOWEVER, some people stated that the ferrofluid is the “damping” material, loosing it will increase the sensitivity of the driver.
Please clarify the fact
Yes , I've heard some contradicting opinions on the subject.
I am of the opinion that the > quality, quantity of fluid + voice coil gap would determine very much.
I would say that for cheaper/low cost tweeters I would go for FF, you'll get better power handling and
a nice natural damping of resonant frequency. I've also heard that for first-order XO, their nearly a must.
Very special expensive tweeters have attributes > meaning no FF. These likely have better detail of clarity.
I have used FF tweeters that were definitely not dull.
I am of the opinion that the > quality, quantity of fluid + voice coil gap would determine very much.
I would say that for cheaper/low cost tweeters I would go for FF, you'll get better power handling and
a nice natural damping of resonant frequency. I've also heard that for first-order XO, their nearly a must.
Very special expensive tweeters have attributes > meaning no FF. These likely have better detail of clarity.
I have used FF tweeters that were definitely not dull.
When my ears were young and I could hear 20k, the D25AG was the best sounding tweeter I met. And I met quite a few, including various ribbons. Up till today I can’t explain why though, or it must have been the excellent linearity and decay behavior. It wasn’t a star in distortion figures IIRC.
Assuming the ferrofluid is in good shape, neither. And both, unfortunately. Ferrofluid taken generally does not determine the response / sonic characteristics of tweeters, so trying to over-generalise in this was is basically a hiding to nothing. 
Something similar applies to if (if) it dries out, since it depends exactly how it's dried out & what effect that has immediately on the behaviour. If it's gummed up solid, you may have no sound at all, as the coil can't move. If it's become more viscous than originally intended (but not a near-solid 😉 ), you might think that would be 'dull' but it might end up being the opposite for x amount of time -see below:
Neither 'dull' nor 'bright' have a technical meaning as such, but since they're descriptive opposites trending should more or less come into play. You can have terribly harsh HF (presumably the polar opposite of what's being meant / assumed by 'dull') in some designs using ferrofluid if an overly heavy type has been applied for the coil / dome mass etc., pushing distortion through the roof. That's just poor design, rather than the fault of ferrofluid itself. A similar instance may (may) occur in some cases if the fluid has become excessively viscous over time relative to originally specified & applied. But in the same way, you can have tweeters sans ferrofluid with relatively poor HF extension, or mediocre off-axis response (neither of which have anything to do with ferrofluid) to give only two examples that may result in what some might describe as a 'dull' presentation. With good designs, it tends to be as much about how you use them as anything else, as usual.
Something similar applies to if (if) it dries out, since it depends exactly how it's dried out & what effect that has immediately on the behaviour. If it's gummed up solid, you may have no sound at all, as the coil can't move. If it's become more viscous than originally intended (but not a near-solid 😉 ), you might think that would be 'dull' but it might end up being the opposite for x amount of time -see below:Neither 'dull' nor 'bright' have a technical meaning as such, but since they're descriptive opposites trending should more or less come into play. You can have terribly harsh HF (presumably the polar opposite of what's being meant / assumed by 'dull') in some designs using ferrofluid if an overly heavy type has been applied for the coil / dome mass etc., pushing distortion through the roof. That's just poor design, rather than the fault of ferrofluid itself. A similar instance may (may) occur in some cases if the fluid has become excessively viscous over time relative to originally specified & applied. But in the same way, you can have tweeters sans ferrofluid with relatively poor HF extension, or mediocre off-axis response (neither of which have anything to do with ferrofluid) to give only two examples that may result in what some might describe as a 'dull' presentation. With good designs, it tends to be as much about how you use them as anything else, as usual.
Last edited:
I had two experiences:
One old seas tweeter pair. The increased viscosity of dried ferrofluid increased resonance frequency. No high frequency rolloff. See this thread.
One pair of aged dynaudio 3way speakers, owned by a friend. He told me there was a lack of high frequency extension which I could also measure. The tweeters slightly rolled off at above 8 kHz. We did not investigate much further but I am sure this was due to dried out/viscous/sticky ferrofluid.
One old seas tweeter pair. The increased viscosity of dried ferrofluid increased resonance frequency. No high frequency rolloff. See this thread.
One pair of aged dynaudio 3way speakers, owned by a friend. He told me there was a lack of high frequency extension which I could also measure. The tweeters slightly rolled off at above 8 kHz. We did not investigate much further but I am sure this was due to dried out/viscous/sticky ferrofluid.
Ferrofluid damping is very useful for tweeters used in 1st-order crossovers. Choose a tweeter with fs a little lower than the crossover frequency, and the effect will be to limit the effects of excessive (and very unwanted) out-of-band excursion which is usually the biggest problem with these crossovers.
Don't forget, for 1st-order high passed drivers, excursion increases below crossover frequency - doubling every octave below - until fs is reached (below which the 12dB rolloff comes to the rescue). And damping the resonant peak would be pretty important in this context...
Don't forget, for 1st-order high passed drivers, excursion increases below crossover frequency - doubling every octave below - until fs is reached (below which the 12dB rolloff comes to the rescue). And damping the resonant peak would be pretty important in this context...
Maybe I should add:
Also increased harmonics due to nonlinear (sticky) ferrofluid behaviour could lead to a brighter sound impression.
This could lead to an higher output at the increased resonance frequency (probably near tweeter x-over) and as a result to a "bright" or "sharp" sound.
Also increased harmonics due to nonlinear (sticky) ferrofluid behaviour could lead to a brighter sound impression.
Issues arise not only with 1st order crossovers. And various viscosities of ferrofluid have various impact on the damping of the resonance. The impedance curve tells enough, such as if you have to correct it for the crossover to function properly.
I had dried out ferrofluid in my TDLs and I tested the tweeters. The coil was stuck at the bottom, so the dome rocked rather than pistoning. This lead to a big peak at the LF resonance, little output across most of the range and horrible oil can breakup above 10 KHz.
After careful cleaning, I got a reasonable response, a bit peaky and I had to tame it with a series resistor due to the increased efficiency
After careful cleaning, I got a reasonable response, a bit peaky and I had to tame it with a series resistor due to the increased efficiency
A few years ago I reworked a pair of then 20 years old speakers I had built for a friend with Morel MDT30 tweeters. The sound was more "blurry" than "dull", i.e. there wasn't "less treble" but the transients sounded positively "smeared". When I opened them, sure enough the ferrofluid had turned to molasses. I replaced it and they came back to life.
On a side note, recently I updated a pair of 30 years old Vifa D25TG-55 with fabric domes and I must say the ferrofluid looked just fine, I replaced it for peace of mind more than anything else.
Cheers,
Cabirio
On a side note, recently I updated a pair of 30 years old Vifa D25TG-55 with fabric domes and I must say the ferrofluid looked just fine, I replaced it for peace of mind more than anything else.
Cheers,
Cabirio
Yup, thx. I liked the -35 version better, but more than adequate, even by modern standards. Mind you that this measurement has been made from an old driver. Judging by the soldered leads Yevgeni had a used one at his disposal, guess it was more than 20 years old already. Talking about ferrofluid stability, I noticed that too with some Vifa D26 lying around here, 30 years old, those.
Interesting notes about old tweeters >
I would be thinking (although not sure) that with so many years since the 'invention'/ application
the quality of FF today would have to be much better across the board.
All manufactures must surely keep abreast of advances and old problems.
I would be thinking (although not sure) that with so many years since the 'invention'/ application
the quality of FF today would have to be much better across the board.
All manufactures must surely keep abreast of advances and old problems.
Yes, they look very clean and well performing, through the useable range, despite age and clearly used from before. No nastiness in impulses, impedance, csds are clean, and the response is easy to work with.
Distortion performance is not groundbreaking, but that is not a dealbreaker either. Used with passive components, less of a issue even 🙂
It generally is; they've got on top of some of the evapouration issues, and they also tend to use lighter types in general than was widely the case in the '90s etc. The pricier models usually had an edge too. I'm the world's worst driver-fiddler (possibly a good thing 😉 ) but when I pulled a set of 20+ year old 9300s (IIRC it was that one) apart there was nothing wrong with the ferrofluid; ditto a similarly elderly set of T25CF002. I can't say the same about a pair of much-cheaper Peerless units (I can't remember the model number now) but they were both a lot cheaper, and even older. Either way, I think I can see the way the OP is going, and it's tempting to make the leap and say 'ferrofluid always dries out within a probable lifespan, this always causes xyz issue, and therefore (making another jump) tweeters with ferrofluid always have uvw characteristics'. The reality isn't that simple: it doesn't always dry out within respectable usable lifespans (especially now); when it does in some old types, this doesn't always result in the same audible issues (except at the extreme end) and its use in a tweeter in general doesn't always produce a specific set of characteristics.
Not shabby for a fairly old (and not especially pricy at the time) model is it? I'd take that over a few I can think of even now.