So have these old speakers that I can hear the potential of but not quite satisfied with the character of them. Tried to fix the problem with simple L-pads, but still heard peaks and valleys. Then I tried modelling the original x-over in LTspice - a roller coaster 😀 decided do redesign them. The criteria: use only the inductors I have and as many capacitors as possible (bought them to replace the old ones). Here's where I'm at:
Response as modeled:
Any comments on foreseeable problems? Tips?
Thanks in advance! 🙂
Response as modeled:
Any comments on foreseeable problems? Tips?
Thanks in advance! 🙂
Search the forum for the procedure of measuring drivers, simulation software and evaluating measurements. Then you will know better how much work you have ahead of yourself. To tune a speaker and design a 4 way XO filter would take me couple of dozens of workhours practically. You have made an inadequate simulation by turning drive units into resistors, not to mention no acoustic measurements.
^ What he said.
To expand a little, drive units have complex impedance curves and frequency responses. Assuming they're each a resistor with a flat frequency response will get you nowhere.
FWIW I've been doing this stuff for a few years now, and I wouldn't attempt a four-way passive speaker without a measurement setup and a big box of crossover parts. Even then, I'd probably go for a 3-way.
Chris
To expand a little, drive units have complex impedance curves and frequency responses. Assuming they're each a resistor with a flat frequency response will get you nowhere.
FWIW I've been doing this stuff for a few years now, and I wouldn't attempt a four-way passive speaker without a measurement setup and a big box of crossover parts. Even then, I'd probably go for a 3-way.
Chris
Thank you for the input! Forgot about speaker impedance curves.
This is my first DIY speaker project so I'm not expecting perfection. Currently I have no measuring tools at all (except a multi meter) but I see it's possible to do primitive measuring using a decent sound card and some simple kit.
As for measuring driver responses - don't think it matters much since I'm using the same crossover points as in the original schematic but will do some primitive measurements, too. The reason for the 4-way system is simply because this is a mod, not a ground-up build. Do not have enough time for that since I have a day job.
This is my first DIY speaker project so I'm not expecting perfection. Currently I have no measuring tools at all (except a multi meter) but I see it's possible to do primitive measuring using a decent sound card and some simple kit.
As for measuring driver responses - don't think it matters much since I'm using the same crossover points as in the original schematic but will do some primitive measurements, too. The reason for the 4-way system is simply because this is a mod, not a ground-up build. Do not have enough time for that since I have a day job.
A day job is simply an excuse, admit it.🙂
ARTA measurement suite can do all the measurements you can think of, and with a mediocre soundcard and a measurement mic, you are as pro as needed. In fact, if you were a hard core DIYer, you could build a mic of your own with a Monacor MCE-4001 and put it in a pencil enclosure. I have done it once and it worked as good as Audiomatica microphone, only it was a Panasonic WM61. Good luck!
ARTA measurement suite can do all the measurements you can think of, and with a mediocre soundcard and a measurement mic, you are as pro as needed. In fact, if you were a hard core DIYer, you could build a mic of your own with a Monacor MCE-4001 and put it in a pencil enclosure. I have done it once and it worked as good as Audiomatica microphone, only it was a Panasonic WM61. Good luck!
Hi Reapper,
You might want to check out my blog on the LM-1. It will refer you back to the DIYaudio site for XSim and files. You can play around and get a feel for simulating speakers and crossovers more fully with it. The design files are free, and include frequency and impedance measurements already.
https://speakermakersjourney.blogspot.com/2016/02/the-lm-1-bookshelf-version.html
Best,
E
You might want to check out my blog on the LM-1. It will refer you back to the DIYaudio site for XSim and files. You can play around and get a feel for simulating speakers and crossovers more fully with it. The design files are free, and include frequency and impedance measurements already.
https://speakermakersjourney.blogspot.com/2016/02/the-lm-1-bookshelf-version.html
Best,
E
Unfortunately, developing a new crossover is pretty much as much work as developing a speaker from scratch, and arguably even more work as you typically have to determine TSPs and measure acoustic response from scratch instead of picking a driver conveniently supplied with your speaker simulation software. Sometimes TSPs and response measurements (even if not necessarily taken under standard conditions) can be found in old data books, and graphs can be converted into a series of data points using appropriate software. Obviously you have to know what the drivers are, and it's only an option for ones made since about the 1980s anyway.
4-way sounds like a 1970s job, probably means you're going to have to do it the hard way.
For starters, make a TSP measurement adapter to be used with the application LIMP that comes with ARTA, there's not an awful lot to it. If you've got these, you'll get a pretty good idea of how especially lower-frequency drivers behave, minus some nonideal nastiness like cone breakup. Something to get you started with meaningful simulation in any case.
4-way sounds like a 1970s job, probably means you're going to have to do it the hard way.
For starters, make a TSP measurement adapter to be used with the application LIMP that comes with ARTA, there's not an awful lot to it. If you've got these, you'll get a pretty good idea of how especially lower-frequency drivers behave, minus some nonideal nastiness like cone breakup. Something to get you started with meaningful simulation in any case.
- Status
- Not open for further replies.