Vifa XT25 help (Weird Series Capacitor value)

[force of 1/2]

Hello,

I'm building a 2-way using the Vifa XT25tg30-04. When I model the crossover I get some funny results for the HP filter. In both Speaker Workshop and Passive Crossover Designer (I am aiming at an acoustic L/R2 at around 2500hz) I end up with a series capacitor value exceeding 30uf! Does this seem reasonable?

Has anyone else modeled this tweeter and found similar results? Are my FRD and ZMA files way off? The other odd thing is that I have these in a box with an old Dahlquist crossover I had laying around, which appears to be a pretty normal 12db/octave crossover, and they sound great.

Just thought I'd check with someone before I shell out the money for those huge film caps.

Thanks,
John

[ashok]

You can always try inexpensive electrolytics back to back with small value film in parallel to get 30 uF and try it out. If it is OK you can buy the expensive film type. ( say 2x47F + 2x10uF elcos will give 28.5uF . That's within 5 % of your value). However you might have to measure the cap value as elcos have a very wide tolerance range.

This doesn't answer your question if the value is correct but thought I'd suggest that you confirm it works well ( if calculated correctly!) before plonking good money for expensive film caps.

[DualTriode]

Hello,
I am using those tweeters and like them. I have not yet built a XO i am using a digital XO.
Before you go expensive try this capacitor. Dayton DMPC-30 30uF 250V Polypropylene Capacitor | Parts-Express.com
I have used other values and liked the results.
DT
All just for fun!

[chlorofille]

Don't forget to add a notch filter at the tweeter's resonance (450Hz).
[100uF + 1mH + 4.7R] across the tweeter.

[Moondog55]

It is about double the value I would expect, have you double checked the impedance and DCR values entered into the program??

[force of 1/2]

Thanks, everybody, for your replies.

My first thought was to use cheap electrolytics for testing purposes but I am concerned that the actual filter slope will extend too deep into the bass and damage the tweeters-- I read somewhere they have xmax issues and can be easily damaged.

I have checked and rechecked the DCR values, and the ZMA files and they both look accurate. If I model a tweeter with similar characteristics like the RS28a I get a similar result. This really does seem about double to triple the Series capacitor value I would expect.

One more thing... when I try to model a known filter like Zaph's ZD5 or one of Troels Gravesen's designs the SPL graph look really strange. I must be doing something wrong, right?

No else experience this?

John-

[mgm31]

I did a 2 way (Peerless HDS PPB 830860).
Crossover freq. is about the same but the capacitor is only 6.8uF.

Could you post your schematic ?

[sreten]

Originally Posted by force of 1/2

Thanks, everybody, for your replies.


One more thing... when I try to model a known filter like Zaph's ZD5
or one of Troels Gravesen's designs the SPL graph look really strange.
I must be doing something wrong, right?

John-

Hi, very highly likely, cross checking needs to give consistent results, rgds, sreten.

[force of 1/2]

Originally Posted by mgm31

Could you post your schematic ?

So this is the circuit that modeled as being nicely flat with an acoustic L/R2 at 2500hrz.

[PeteMcK]

what is the value of Rlpad1? is that 170 ohms (way too high)?
the impedance of the tweeter is ~ 4 ohms at 2500Hz, so for LR the cap should be ~ 8uF, the inductor ~ .5mH