Home Theater Forum and Systems banner

1 - 9 of 9 Posts

·
Registered
Joined
·
53 Posts
Discussion Starter · #1 ·
i am new to speaker designing and the passive x-over seems very intimidating to me. i understand basic electrical stuff but i have been reading about passive x-over design and i must say its hard for me to wrap my head around since alot conflict each other with personal preference. the aproach i have come up with is to get a minisp 2x4 and set x-over according to winISD to get a starting point and play with it in increments that i can use to build a simple x-over. i will repaet i am new so this seems like a good idea with the limited kowledge i have.:ponder: i have a dayton DATS on the way so i can get info off the speakers after break in to get a better model in winISD. also purchased some wood tools to make good encloures and other things. i have a 5ch amp to use with the minidsp for testing. i dont have anything to record actaul response other than my ears. REW seems like something i will get into later as i learn more.

am i missing a step or will this work to get started?
 

·
Registered
Joined
·
1,747 Posts
An active crossover probably won't behave the same way as a passive crossover with the same filters, because a passive crossover interacts with a system's impedance curve while the active does not.

But at the end of the day it's not the filters we're worried about simulating. It's the final acoustic transfer functions and driver phase interaction.

A digital active crossover like the miniDSP makes design a lot easier, because you can play around with the delays and acoustic response easily and measurably. For a passive crossover, you normally want assymettric acoustic slopes in order to account for driver offsets, wheras this isn't necessary for an active crossover.

Can the miniDSP be used to give you the desired acoustic response from the passive crossover for a simulated listen? Yes, if done right. But the passive crossover can't necessarily match up to the active crossover you end up with.

If you don't have a measurement microphone, you really don't have much hope of getting a good crossover designed.
 

·
Registered
Joined
·
53 Posts
Discussion Starter · #3 ·
i was planning on a 6.5 and tweeter for my first build to keep it simple. from what i understand the crossover point will determin how close i need the drivers. i can not for the life of me find the calculator i found earlier that will tell you how close they should be. but my understanding is as long as i keep them at least that distance they should act as a cohesive unit.

i do plan on getting a mic but i have to build them before i can measure them. i am really interested in REW as it seems like it has a good following and lots of info for dumb people like me :dumbcrazy:. is this a good choice or do i need to look at another option. im mainly doing this as a hobby in the little free time i have and learning as i go. once i get them built and measured i will post it up so i can see where i went wrong and where i need to do more research. i dont know anyone that is into actual speaker building around me so its just me and the internet.
 

·
Registered
Joined
·
1,747 Posts
i was planning on a 6.5 and tweeter for my first build to keep it simple.
It might seem simple, but it's actually a pretty flawed design, especially with inexpensive tweeters that aren't up to the tast.

from what i understand the crossover point will determin how close i need the drivers. i can not for the life of me find the calculator i found earlier that will tell you how close they should be. but my understanding is as long as i keep them at least that distance they should act as a cohesive unit.
Drivers should be within 1/4 wavelength of each other from center to center. for a 2khz crossover, that's about 1.7 inches max (not only vertically, but also in terms of depth! the acoustic center is at the bottom of the cone where the voice coil is!). for a 4khz crossover that's about .8 inches max. for a 1khz crossover that's about 3.4 inches max. for a 500hz crossover that's about 6.78 inches max. Essentially, these are very unrealistic targets to achieve without going coaxial or unity horn. Getting the drivers as close as possible is a good starting point, but ultimately you have to make compromises, and it starts by optimizing the lobing of the drivers. Since the above targets aren't likely, the next best thing is to avoid these lobes being on the forward axis - and doing so is part of the crossover design.

i do plan on getting a mic but i have to build them before i can measure them.
Nothing wrong with that, but the mic is vital to the crossover design. it is not optional.
 

·
Registered
Joined
·
1,747 Posts
so in short i need to read more?
If you haven't already, I would suggest you start by assembling a well-done reputable DIY design and simply listening to and measuring it. Get yourself familiar with "why" it was designed the way it was. A true reference speaker that you can go back to and pick apart as you try to design your own. Understand what the difference is between a good speaker, and a speaker that is a collection of parts. There's a sense of pride in designing something yourself, but you need to do things one step at a time. Don't just jump in head-first. Reverse engineer something that you have in front of you and learn that way. Of course the theory is important, but you have to know what that theory is supposed to lead you to.

Perhaps the Stance:

http://meniscusaudio.com/stance-pair-p-1353.html

Might be a fun place to start.
 

·
Registered
Joined
·
53 Posts
Discussion Starter · #7 ·
i have already built the parts express tritrix kit. i spent a while playing with polyfill, eq and placement before i was happy with them. i dont have a mic so im not sure how i affected it. i just know i like the way they sound now. i also played with my sat speakers from a cheap martin logan 5.1 system. fiberglassed the insides with a dowel across it. i played with the filling they came with to. i have a emotiva umc-1 and xpa-5 and the umc-1 eq is limited. when i was into car audio i would spend hours listening to certain songs tweeking them till i got the sound i wanted from them.

so you suggest that i get a diy kit and build it to spec and get a baseline then modify it till i like it measuring everytime i try something?
 

·
Registered
Joined
·
302 Posts
Let me try to broaden the conversation a little... Here's a link with a lot of good FAQs at the right.
https://sites.google.com/site/undefinition/diy

I will also repeat the advice to "build a proven design." The term "kit" doesn't apply in the sense of buying a bag of parts, as you source the kit yourself from a bill of materials (BOM), circuit diagram and construction drawings that someone else provides. The trick is that you're not the first to build it, and others have been successful, so you should be, too.

Put another way, folks were building Tritrix TLs long before PE formalized a kit. You've already started making it your own - playing with polyfill, eq and placement - as we all do.

Based on what you already have, all you need is a measurement mic and some simulators. A crossover simulator allows you to upload frequency response and impedance data, input driver locations and start developing filter circuits based on predicted response. Box simulators let you predict bass response and cabinet diffraction.

There's a lot here, and in some ways you're ahead of me (haven't read a speaker design book yet, but I'll still suggest reading Paul's FAQs. They're a great place to start.

HAve fun,
Frank
 

·
Registered
Joined
·
53 Posts
Discussion Starter · #9 ·
thanks for the link. i read it and i feel a bit more comfortable. i really like the tarkus design he has and i think i will shoot for something along those lines or i might get frustrated and just go with his design lol.
 
1 - 9 of 9 Posts
Top