Crossover designing help needed.

[viki_v2]

Hello,
I have just completed building my MTM floorstanders, Now i need to build a crossover for it.Kindly help.

Also i need help for winding inductors.

[pinkmouse]

You need to measure them first. Come back when you have the FR and phase curves for each driver in your enclosure, then we can help.

[viki_v2]

I have data sheet of the drivers , which shows fr and curves. Will that help.
If not tell me how to do that.

[ajazz]

Here's a simple Xover calculator I found it on the Web.

If you have selected the Drivers right it should work for all or else U have to do a lot of tweaking to get the response right.

Good Luck

[wintermute]

Hi viki, I think that you may get a better response if you asked for info on what you need to read about in order to make a start. Your response to pinkmouse shows your ignorance. Published driver curves may bear no resemblance to the performance in YOUR implementation.

As a starting point I'd recommend reading David Weem's Designing Building and testing your own speaker system Designing/Building Speaker Systems 4th Edition Book | Parts-Express.com

For more in depth I would recommend Vance Dickasons Loudspeaker design cookbook Loudspeaker Design Cookbook 7th Edition Book

I would also recommend Joseph D'appolito's Testing loudspeakers Testing Loudspeakers Book

If that isn't enough you may also want to get Vance Dickason's Loudspeaker recipee's book 1. Loudspeaker Recipes Vol 1 Book

Yes I own all of them. and I probably bought them (over a period of 5 years) in the order listed.

Does it make me an expert having read them? Of course not.. but at least I know enough now to ask meaningful (I hope) questions....

If that seems daunting, then making someone elses proven design (and sticking to it exactly) is the safest alternative.

Tony.

[adason]

Originally Posted by wintermute

For more in depth I would recommend Vance Dickasons Loudspeaker design cookbook Loudspeaker Design Cookbook 7th Edition Book

I have this one too and can send it free to anyone, just pay shipping

[viki_v2]

That is wonderful and very generous of you. I will surely do that, but have to wait till 10th of next month. that is my salary time.

[BobEllis]

There are other online references that will help guide you through the process of designing a crossover. Claudio Negro wrote a manual for Speaker Workshop that is also a tutorial on basic design and measurement techniques. http://www.claudionegro.com/

Since you posted here, I assume that you want to build an active crossover. See Linkwitz Lab - Loudspeaker Design, don't let the math intimidate you. It took me a few times through to get the concepts well enough to be able to understand the math. Eventually you will discover relatively simple formulas for designing the crossover building blocks. If you plan a passive design, multi-way is a better forum.

You'll need to do a bit of research to fill in the holes, but the basic process I use to designing a crossover once I have selected the drivers and built the box is:

1. Measure on axis frequency and phase response of the individual drivers. Measure your Ms as one connected the way you will in the system. If you are using a passive crossover, also measure the impedance curves of the drivers. I use Holm Impulse and a Cross Spectrum Labs calibrated microphone. Other options are available.

2. Use these responses to model a system in one of the many CAD programs available. There are free programs as well, try BASTA or ARTA demos.

3. Build a crossover based on the model.

4. Measure filtered individual driver and system on axis response.

5. If not close to predicted go back to step 1. There may be an issue with measurement technique.

6. If close to predicted, some people stop here.

7. Adjust baffle step for intended positioning (against wall or out in the room)

8. Measure system off axis response. The sum of measurements is called power response.

9. Look for sharp discontinuities in off axis response and adjust crossover to smooth them. Usually this results in giving up some on axis flatness, but usually it is worth the compromise.

10. Measure on and off axis response. Repeat steps 7-9 as needed.

11. Adjust values to voice the system to your liking. Some people like a little BBC dip, or a slightly rolled off top end.

12. Measure on and off axis again to ensure that you haven't introduced an anomaly that sounds good at first but is inaccurate.

13. Go to step 2 and choose a different crossover point, and slope, repeat the rest of the steps. I actually do this fairly often since I got my Frequency Allocator DSP crossover system running. Measuring a set of response curves takes longer than completely reshaping the crossover. Of course you can skip to 15 if you are happy with your results so far.

14. Eventually decide that this is the right version of the crossover.

15. Enjoy the music.

As you can see, GOOD measurements are very important to the process. There is a lot of technique involved in measuring that took me quite a while to get consistently useful results. Plenty of reading on the web, and Joe D'Appolito wrote a book on measuring speakers.

This doesn't mean that you can't get good results without measurement, but it is a lot less likely and will be a longer process. Start with published curves in a CAD program, which will be a lot better than using a generic crossover calculator. You'll need a good reference to "calibrate" your ears - Linkwitz suggests some headphones, but they are either no longer available or cost as much as a good measurement rig. Watch for headphones that intentionally have non-flat response to meet marketing goals.

Good luck.

[tinitus]

being a member it is your right to open a thread and ask questions


moderation

off topic personal posts removed

[viki_v2]

Tinitus i am sorry for all this fuss going on, I will try to be more careful about the questions i ask.

My problem is i have no formal electronics education, I have learned some things here and there, some by committing mistakes.

If some say go and have some education and then come, I joined for doing the same.

at some point even the learned ones , might have had some questions or doubts and asked some others for the same. this forum is the best place to do so.

Just with small help and guidance i will learn a lot.
I did not want to trouble any one.