DIY crossover inductor question

[70's Rockfan]

I am getting ready for a diy speaker project and am
planning on using 2nd order 3 way crossovers. does the
Inductance MH on the speaker and the MH on the
crossover Inductor need to match each other.
Thanks

 

[bowtie427ss]

The 3 main factors that will determine inductor value are the actual impedance of the load @ the intended filter frequency along with the type/slope of the filter being employed. There are other factors, but those are the main 3.

 

[70's Rockfan]

so your saying that they do not have to match.
I have a 8" woofer 8ohm that has a 1.1mh inductance.
in one of the crossover calculators I put in 8ohm and 500hz
that calls for a inductor of 5.09mh and a cap of 19.9uf
so that would be correct ?
at what Frequency do you recommend for a 2nd order 3 way
for the woof , mid and tweet.

 

[Spe@kerBox]

The problem is not matching the two inductances, but rather what impact does the voice coil inductance have on the impedance of the woofer at the crossover frequency? At 500 hertz you are adding an ohm or so to the load the xover sees. So you have a several choices: design the crossover so that the VC inductance is part of the xover(not easy), add a zobel network to the woofer to flatten the impedance, or lower the xover frequency,

 

[70's Rockfan]

I just looked at some zobel network design and calculator. interesting.
I changed the value in the crossover calculator to 8ohm 800hz
that makes the inductor 3.18mh and the cap to 12.44uf
would that be better for the 1.1mh of the speaker.
as you can tell I'm just learning this and still a little confused.

 

[Spe@kerBox]

I just ran the numbers for the 1.1mH VC at 500 hertz and if the DC resistance of the VC is between 6 and 7 ohms you are probably OK. You might want to check that value though.

 

[70's Rockfan]

I checked them and one is 8.2 and the other is 8.3

 

[homebrew]

If you're looking at standard crossover calculations on the net, it'll take you forever tweaking to get close. There are reliable crossover design programs you can download. You're much better off spending your time learning to use them.

http://audio.claub.net/software/jbabgy/jbagby.html

 

[RTally]

If you're looking at standard crossover calculations on the net, it'll take you forever tweaking to get close. There are reliable crossover design programs you can download. You're much better off spending your time learning to use them.

http://audio.claub.net/software/jbabgy/jbagby.html

And then get a RTA and calibrated mic to do the final tweaking. You should also obtain an impedance curve to ensure that the speaker is well behaved across its FR.

 

[70's Rockfan]

Thanks for the link. I was planning
on trying something like that as I learned
a little bit more about crossovers.
Thanks for all the help guys.

 

[bowtie427ss]

You should also obtain an impedance curve to ensure that the speaker is well behaved across its FR.

he needs the impedance curve to determine the BASIC impedance at the intended crossover frequency. It will likely be anything but the nice neat square 8 ohms he keeps plugging into the calculator. It's more relevant than the inductance of the driver. Zilch called it "wanking", it's an appropriate term.

 

[70's Rockfan]

Thanks for your help guys.
I'm going learn how to use the design program that homebrew linked.
once I start to understand it better I'll be back.

 

[RTally]

he needs the impedance curve to determine the BASIC impedance at the intended crossover frequency. It will likely be anything but the nice neat square 8 ohms he keeps plugging into the calculator. It's more relevant than the inductance of the driver. Zilch called it "wanking", it's an appropriate term.

The impedance curve will also tell him if he needs a Zobel network to even out any undesired impedance swings. The more level the impedance seen by the amp over the FR, the better the amp will work. I believe that a lot of the difference in amps that people hear is due to how the various amps respond to impedance variations over the FR.

 

[toxcrusadr]

What drivers are you using? Is there data (such as impedance curves) available for them anywhere? This would save you having to make your own measurements. If you had the impedance curve for the woofer, for example, you can read off the impedance at the desired xover frequency and plug that into a calculator instead of 8 ohms. Similarly for the other drivers.

Of course, a sophisticated piece of software is better, one that can take in the curves for all the drivers, and the crossover parameters, and generate a result for the entire system. But the above is at least a step above using 8 ohms nominal impedance for everything.

 

[70's Rockfan]

I have been looking at the data on a few different woofers.
I have not decided on any yet. is this the data I should be looking at.
https://www.parts-express.com/pedocs/specs/295-356--rs225-8-spec-sheet.pdf
Thanks

 

[Spe@kerBox]

I checked them and one is 8.2 and the other is 8.3

So at 500 hertz the magnitude of the impedance will be approximately the square root of (8.2 squared + (6.28x500x0.0011) squared) = 8.9 ohms. So you could design for that resistance (~9 ohms) or use a zobel to flatten for 8 ohms. You also want to know that you are crossing over at a high enough order so that the woofer does not receiver out of band frequencies. Look at the frequency response graph for that info.

 

[70's Rockfan]

The more I learn about this , it sounds like the zobel would be a good idea.
Thanks for your help.