modding Yamaha NS-150. Measuring small-thiele parameters & new crossover

[zafira1981]

Hi!!

I want to improve my 3-way speakers since I damaged the tweeter of one of them.

I opened the speakers enclosure and I was so disapointed with the crossover

Electrolitic capacitors and small inductors.



So I decided to do some improvements to the speaker system:

- [DONE] change both tweeters (dome) with new ones (Audax TW010F1) . Since the audax tweeter is not a silk dome like the originals, i have to upper a little the second crossover frequency.
- build an entire new crossover network, with decent inductors and capacitors

I started looking for the small/thielle parameters for each driver but I was unable to find them. So I decided to measure them using this guide:
http://sound.westhost.com/tsp.htm

I got the Fs, but I cannot continue measuring because I dont know what is Is, and the page didnt says.



So, if anyone can help me, I have a few questions.

1- what is "Is"? (look at the excel screenshot)
2- what do you think about the driver?
3- both "subwoofer" and "mid" are the same driver. Is that OK ??

[zafira1981]

some pics

[zafira1981]

why nobody answer my post? I need some help, and there are lot of you that can help me!!

[mjf]

hello.
holiday...............sitting in the garden (and drinking water or so)........
i am no speaker expert,but
have a look at figure 6 in the text.................
rs = 10 ohm
vref = voltage drop in the linear region above resonant frequency fs (in your diagram around 200.......300Hz; and between 10 and 20Hz)
and Is should be vref/rs (by ohm's law).
the driver has a high fs for a woofer...............no real "deep bass" if your measurement is well done.
looks like this is an "two and a half way sytem" (woofer and woofer-midrange).
have you drawn a schematic of the original network?
perhaps you do not need all this parameters because you have a box with it's speakers - so you only need to upgrade the parts and adapt the network for the tweeter................
greetings...............

[Inductor]

Originally posted by zafira1981
why nobody answer my post? I need some help, and there are lot of you that can help me!!

Right...
http://www.diyaudio.com/forums/showt...65#post1843565

[zafira1981]

Originally posted by mjf
hello.
holiday...............sitting in the garden (and drinking water or so)........
i am no speaker expert,but
have a look at figure 6 in the text.................
rs = 10 ohm
vref = voltage drop in the linear region above resonant frequency fs (in your diagram around 200.......300Hz; and between 10 and 20Hz)
and Is should be vref/rs (by ohm's law).
the driver has a high fs for a woofer...............no real "deep bass" if your measurement is well done.
looks like this is an "two and a half way sytem" (woofer and woofer-midrange).
have you drawn a schematic of the original network?
perhaps you do not need all this parameters because you have a box with it's speakers - so you only need to upgrade the parts and adapt the network for the tweeter................
greetings...............

thank you very much, I've figured out how to do it.


I have a new question.

I want to connect a zobel network in the speaker, but according to this page http://sound.westhost.com/lr-passive.htm I have to know the inductance of the voice coil, wich I dont know. Using the excel spreadsheet I know the RLC parallel values of R, L and C, but I dont know the R and L series (I supose that L series is what I need)



my question is how to find the L series, and if it is the one that should I use for the calc of the zobel network.

thanks!!!

[mjf]

hello.
there is annother way to determine a rc network to compensate the increasing impedance in the upper frequency region.........
by experiment and measuring:
first measure the coil res (rvc) with a dmm(ohmmeter).
r in the zobel network should be around 50% bigger than rvc:
r = 1,5 x rvc
(powerrating a few watt).
and second try to find a cap that makes the impedance flat........
i would begin with 10uf (non polar elco) in your case.
build the network onto the speaker and measure the impedance........
change the value of the cap a little bit if it does not fit to find the flattest impedance curve............
greetings............

[zafira1981]

thank you very much, I will try to find the C by testing like you say.

By the way, I just finished building the box for measuring the VAS.

Greetings from Argentina!