Comparing tweeters with or without modern low viscosity FF, I honestly don't hear a night and day difference between the two. That wasn't the case with the older FF used on tweeters. They were rather temp sensitive, so you'd notice a warm up period, but the new stuff not so much. I have also started adding FF to some compression drivers with excellent results. Originally the goal with it was to reduce the likelihood of VC failure in speakers I designed for PA application customers, and it did the trick with a specific customer who was really pushing his drivers hard.
When I noticed the benefits of adding FF and realized it didn't nearly affect SQ as much as with the older type, I started trying it on my own HF drivers. CDs like the JBL2407 ring radiator greatly benefit from this in the lower mids around the main the resonance peak, which audibly cleaned up that area. Measuring the difference showed a slight overall reduction in 3rd order HD on the PRV D260MYB, which is a similar 35mm VC ring rad to the JBL. It sounds like a higher end dome with a 3.5k xover using 300uL of FF in the gap. It was mounted on a B&C ME20 WG.
On the topic of FF with conventional soft domes, the addition of FF also didn't show the typical veiled sound as with previous FF types. There is only a subtle difference but IMO the benefits outweigh the slight drawbacks when the correct amount was used. As someone stated before, using the peak impedance method is the most accurate way to add the appropriate amount of FF. I use an insulin syringe to add/remove FF, which allows very accurate, repeatable results and less waste. Adding it by using those little plastic sleeves it comes in just pulls too much out at a time via the magnetic force in the gap.
When I noticed the benefits of adding FF and realized it didn't nearly affect SQ as much as with the older type, I started trying it on my own HF drivers. CDs like the JBL2407 ring radiator greatly benefit from this in the lower mids around the main the resonance peak, which audibly cleaned up that area. Measuring the difference showed a slight overall reduction in 3rd order HD on the PRV D260MYB, which is a similar 35mm VC ring rad to the JBL. It sounds like a higher end dome with a 3.5k xover using 300uL of FF in the gap. It was mounted on a B&C ME20 WG.
On the topic of FF with conventional soft domes, the addition of FF also didn't show the typical veiled sound as with previous FF types. There is only a subtle difference but IMO the benefits outweigh the slight drawbacks when the correct amount was used. As someone stated before, using the peak impedance method is the most accurate way to add the appropriate amount of FF. I use an insulin syringe to add/remove FF, which allows very accurate, repeatable results and less waste. Adding it by using those little plastic sleeves it comes in just pulls too much out at a time via the magnetic force in the gap.
I also have to add that FF migration is one of the main issues using FF, especially in drivers which aren't originally designed to use it. I've noticed that with larger VCs and wider gaps, FF tends to follow gravity and will bleed into any unjoined tight gaps which present capillary force to the FF. Any paper or porous material close by will wick FF up over time. The way around this is using a light coating of wax from white candle paraffin or children's crayon. Any driver which failed from overheating the VC and with FF already in the gap should have the FF replaced. The heat which affected the VC will have heated rhe FF up to the point it starts to deteriorate and increase its viscosity. Morel used some really susceptible FF chemistry back in the 80s, as well as Celestion and.Tannoy. The Japanese stuff was better ie. Foster, Fostex, Pioneer / TAD. The only way to test the integrity of FF is to perform a smoke point test, which is destructive in nature. This was only done to verify existing batches of FF for suitability as service replacement. FF present in drivers would be suspect if more than 10 yrs old and from a failed driver.
Speakers are NOT Linear Time Invariant; LTI. We speaker gurus just often find it convenient to assume that they are. 🙂 The biggest audible difference with Ferrofluid is it makes them somewhat more TI rather than any L improvements. Today you can use Prof. Angelo Farina's swept sine to get a really good handle on this. Just compare Frequency Resp. using 1s, 2s, 4s & 8s sweeps ... but the advantage is easily heard.
Celestion didn't use Ferrofluid until the mid 90s. I know cos I introduced it there.
@kgrlee I came across some tweeters from celestion with FF in them. I don't know if those came that way originally or someone added it, but looking at impedance curves, it looked to be a factory equipped thing. Not sure...
Sorry, don't understand what you mean by the abbreviations L and TL?
Sorry, don't understand what you mean by the abbreviations L and TL?
In the 25mm unit, the larger voice-coil potentially allows for better overall power handling than the 19mm unit with ferrofluid, and the larger diaphragm also allows for larger maximum sound pressure levels. The smaller diaphragm can produce a more extended, smooth high-frequency response, with better off-axis dispersion at higher frequencies. Of course, there exist 34mm tweeters that make use of exotic material (beryllium), which can be quite smooth on-axis even out to 30kHz or so. But their off-axis radiation pattern will be narrower than that of 25mm and 19mm tweeters. The 25mm designs seem to locate themselves in a bit of a "sweet spot" as far as overall performance and technology goes.