advertisement


The Curse of Ferrofluid

timpy

Speaker Abuser
I notice that both ATC and GR Research, just 2 examples, make the point that their High Frequency units, in at least some designs, specifically don't have ferrofluid in them. This is put across as a positive.

I would like to avoid it, having had some sorry experiences with elderly KEF Uni-Q drivers.

Do any of you chaps worry about / try to avoid ferrofluid in your tweeters? Either as they age or buying secondhand?
 
I can honestly say that, in 30 odd years of speaker fettling, ferrofluid has not crossed my mind even once as a problem.

My interactions with it have been during my time designing loudspeakers when it was added to the coil gap when building tweeter prototypes, and on one single occasion when I replaced voice coils on a pair of tweeters and so replaced the ferrofluid whilst I was in there.

Frankly, I've found tweeters with no ferrofluid to be far more troublesome, a lot less reliable and in the case of some older designs, quite unpleasant sounding. As an example, it was only recently that I discovered that the old 1970s Goodmans DT3 tweeter was designed without ferrofluid. It's a nasty sounding thing unless you put some serious crossover witchcraft in place, and I'd estimate that around 80% of the examples I've encountered have been burnt out.

In summary - ferrofluid is fine by me!
 
I only worried about it when I owned some JBL s2 9800s which retailed at £38k (I bought secondhand, much more cheaply.....) As these grew older I was concerned that if ever had to sort out a horn I was either in for some ferrofluid replacement or a very expensive list price horn!
 
I can honestly say that, in 30 odd years of speaker fettling, ferrofluid has not crossed my mind even once as a problem.

My interactions with it have been during my time designing loudspeakers when it was added to the coil gap when building tweeter prototypes, and on one single occasion when I replaced voice coils on a pair of tweeters and so replaced the ferrofluid whilst I was in there.

Frankly, I've found tweeters with no ferrofluid to be far more troublesome, a lot less reliable and in the case of some older designs, quite unpleasant sounding. As an example, it was only recently that I discovered that the old 1970s Goodmans DT3 tweeter was designed without ferrofluid. It's a nasty sounding thing unless you put some serious crossover witchcraft in place, and I'd estimate that around 80% of the examples I've encountered have been burnt out.

In summary - ferrofluid is fine by me!
Speaking of the DT3 Adam, is there much if any difference between the DT2, DT3, DT4 and the 571071A? I've frequency-swept examples of all four but I can't draw any meaningful conclusions as the frequency responses and distortion profiles of 40+ year old used tweeters are unsurprisingly all over the shop!
 
.... the frequency responses and distortion profiles of 40+ year old used tweeters are unsurprisingly all over the shop!
I did hear of someone who changed any dome type HF units on speakers nearly as often as people change windscreen wipers on a car. He regarded them as consumable items, apparently.

Where the original ones were unavailable, he would get ones fairly close for the design on a "be better than what's there basis".
 
I changed the ferrofluid in my Kef Reference 3.2s recently. It was an easier job than I anticipated, and while there wasn't a night-and-day improvement in sound, I would say there does seem to be a small improvement in the 'smoothness' of the treble. I did measure the before and after frequency response and distortion my a UMIK-1 measurement mic and REW, but the UMIK isn't resolute enough to measure the distortion levels above about 10KHz so the differences on paper were small.

My speakers are about 25 years old so it's certainly not something that is required regularly. It would have no bearing on my choosing of a particular speaker.
 
One of the reasons for using ferrofluid is to help cool the voice coil. Tweeters can overheat, if overdriven, or if an amp oscillates or distorts a lot. So that's why tweeters without ferrofluid might be less reliable.

Ferrofluid also damps a driver's fundamental resonance. Without it, you may well have problems, although it is possible to address this in the crossover.

One drawback is that it will dry up, eventually.

There is a school of thought, also, which believes that ferrofluid can mask fine detail.
 
I have found that some of the Kef T33 ferrofluid variants dull with age, ie 20 years. I have had to replace the ferrofluid in a pair of meridian speakers in the past when they have gone a bit soft which were seas metal domes.

I would not buy a speaker on the basis of whether or not it used ferrofluid in the tweeter just like I tend not to consider whether or not the woofers use foam or rubber surrounds, same thing in my book.

Rgds
Stuart
 
One drawback is that it will dry up, eventually.

I bought some replacement ferrofluid I didn't use at the time. Unfortunately in the 6-8 years I had it, it had turned to treacle in the syringe and sort of "separated".

It may be that like foam surrounds, some ferrofluid lasts better than others?
 
I only worried about it when I owned some JBL s2 9800s which retailed at £38k (I bought secondhand, much more cheaply.....) As these grew older I was concerned that if ever had to sort out a horn I was either in for some ferrofluid replacement or a very expensive list price horn!
Very simple to do on a JBL 435Be compression driver.
 
The "problem" with oils is that unless they are totally synthetic, which is actually no guarantee of anything, is that they are not one thing, but a range of things (most commonly simple hydrocarbons), mixed together.

Low viscocity means low molecular weight, means higher vapour pressure, means faster evaporation (as a very general rule, and even gold has a vapour pressure, albeit verging on mind-numbingly low).

As the low viscocity components in any oil (mixture) are lost (evaporate), viscocity increases.
 
Speaking of the DT3 Adam, is there much if any difference between the DT2, DT3, DT4 and the 571071A? I've frequency-swept examples of all four but I can't draw any meaningful conclusions as the frequency responses and distortion profiles of 40+ year old used tweeters are unsurprisingly all over the shop!
No idea, I'm afraid - I've never encountered any of the others! Frankly, enduring the DT3 is more than enough...
 
No idea, I'm afraid - I've never encountered any of the others! Frankly, enduring the DT3 is more than enough...
So, by extension, you won't have liked most of the 70s Goodmans models, Minster, Havant, Magnum, Magister, Dimension 8, etc, as these all used the DT3 AFAIK? I moved all of mine on (including my Achromat Kappas and Sigmas) before I got a measurement mic so have no idea how these models actually measured. However, when deciding which of the three pairs of Achromat Betas I was going to keep I was intrigued to find that the tweeters in two of the three pairs had a huge peak in their response around 13kHz. I was aware that there was something about the top end in these two pairs that sounded very odd but I would never have known what the specific issue was without measuring!
 
How do one replace the ferrofluid?
Ah now then. Some disassembly required. It lives in the voice coil gap, so the diaphragm needs to be removed from the magnet assembly to gain access.

It’s usually easier than it sounds. Getting the old stuff out can be a bit trying. Only done it once, others will be more experienced.
 
It’s a pain to take out. I’ve done Celestion tweeters that had lost sensitivity due to the stuff hardening over the years. I reassembled with no goo (!) and got treble back.
 
In 40 years of hifi I’ve had one blown tweeter, because I drove it too hard at a party. Never any other problem related to ferrofluid, though I’m careful what I buy if that makes any difference.
 


advertisement


Back
Top