Crossover design software advice needed (refitting a driver)

Jump to Latest
E

EL506

Member
Joined 2021
#1
Dear members

I've got some speakers and one of their tweeters has gone silent.... No longer obtainable.

Some candidates have been located that will a) fit, b) have similar performance (and were suggested as replacements).

I'll need to adjust the crosssover to get it "behaving close to the old tweeter".

I've got the plots for the old tweeter, the ones for the new candidates and the existing crosssover details.

Is there software that will allow me to input the old data, extract the response curve of the old setup and then use that in conjunction with the new driver data to correct the crossover?

Thanks in advance.
 
#2
If you want to copy the old crossover you would want to measure the old tweeter so it's good one is working.

Yes, you could do that with something like Xsim.

There are several easier ways and more involved ways to do it depending on your choices and skills/equipment.
 
  • Like
Reactions: EL506
#3
AllenB said:
If you want to copy the old crossover you would want to measure the old tweeter so it's good one is working.

Yes, you could do that with something like Xsim.

There are several easier ways and more involved ways to do it depending on your choices and skills/equipment.
Click to expand...
Thanks, was hoping you would chime in, I'm reading your excellent thread "Introduction to desiogning crossovers without measurement".

I do have LtSpice modelling skills and have a Umik-1 measurement microphone as well as a FocusRite solo. But am suffering from early stages of Parkinsons and moments of being capable of fine movements are far and few between. Prefer to do the "hard work" with modelling to get it close enough.

I'm going to take a look at X-sim.

​

 
#4
That will come in handy.

Some would put the new tweeter in and adjust the resistors (eg with clip leads or a variable L-pad) until it fits, if time is an issue.

Copying the current crossover is easy if you can get factory data for both tweeters or you can do comparative measurements. This assumes the current crossover is worth copying. Sometimes they are, and sometimes it is better than the alternative but not always.
 
  • Like
Reactions: GM
#5
Crossover is quite decent with good inductors and I prefer to only add some resistors to adjust sensitivity (and should it be needed then adjust some capacitors). I may go for something like a ScanSpeak D2905/9500 where I do not have correct too much (or anything).
 
  • Like
Reactions: GM
#6
AllenB said:
That will come in handy.

Some would put the new tweeter in and adjust the resistors (eg with clip leads or a variable L-pad) until it fits, if time is an issue.

Copying the current crossover is easy if you can get factory data for both tweeters or you can do comparative measurements. This assumes the current crossover is worth copying. Sometimes they are, and sometimes it is better than the alternative but not always.
Click to expand...
Opening the speaker cabinet and adjusting the crossover several times is something that I want to avoid as much as possible. Modelling will get you in the ball park.

I managed to get the tweeters and the crossover modelled and found which tweeter candidates matches the best.

Did not know how simple this could be, the marvels of the software available.

Once I got the hang of the software it went quickly. Xsim required the tweeters' *frd and *zma files which were extracted from the graphs in the documentation using FrGraphTracer. Simple, straightforward, the graphs just needed removing some lines in paint in order for FrGraphTracer to pick the correct line. The Scan-Speak D2905-9500 turns out to get the closest to the original (actually it looks better in modelling). All that seems to be required is to have an additional 3 Ohm resistor in series with the tweeter.

The Scan-Speak is probably overkill but looks like having the highest chance of succes.

Many thanks for the help, much appreciated.
 
#8
AllenB said:
Excellent result. The tweeter you mention is relatively well balanced and trouble free, as far as domes go. I had a pair once and used them for some time.
Click to expand...

@AllenB

I have to ask for your input again, I've been trying to find the inromfation but sofar have been unsuccesful.

I am wondering where I need to compensate for the difference of the Z location of the dome. Both are 19mm domes but the difference of the distance between the domes in respect to the baffle is close to 8 ~ 9 mm.

Where and how do I need to enter this in Xsim: I notice two locations, one is on the setting of the FRD/ZMA under "mod delay" and then there is a setting under "derived" to enter the distance. I suspect I have to update the *.ZMA too. This is working with the factory supplied graphs that have been converted into *.frd and *zma with FRGraphTracer.exe.

Thanks!
 
You must log in or register to reply here.
Top Bottom