Play with Crossover Design and Simulation - Free

[ErikSquires]

Hi Everyone,

Since a lot of viewers have gone from here to the LM-1 page I wanted to make sure you were aware of an oversight in explaining the project on my part.

When I say the LM-1 speakers are "bookshelf" speakers I mean exactly that. Their frequency response is designed for a desktop or bookshelf and as such need no more BSC than included. Sorry for any confusion caused. If you try these as standmounts you'll need to add some EQ. This is the opposite of most speakers designed quasi-anechoically, which then need EQ in a bookshelf.

This also means you should not compare any of my measurements with designs using quasi-anechoic techniques. Won't match up.

Perhaps a future project I'll re-work the crossover to make a stand-mount version. Or maybe a kind DIY community member will take that on for themselves.

Best,


Erik

[dlr]

Glad you found it!

So am I. It worried me that I may have had a long standing error in the code!

dlr

[ErikSquires]

Glad you found it!

[Brian Steele]

That does not mean something isn't wonky though.

Yeah, something was wonky - with how I laid out the circuit in XSim :-). Turns out that I had a shunt resistor in the tweeter section of the x-over in the wrong location, grr. Under WinPCD, it was directly across the tweeter. In XSim, I had it located across the tweeter and a series resistor. I've corrected that, and now the impedance curves look identical.

[ErikSquires]

Brian, just humor me. Take a look at the driver impedance drawn by XSim and see if the values make sense. It could be some odd read error, or perhaps you accidentally loaded the wrong file. It will take you 10 seconds.

I should also state I have a lot of confidence in XSim's impedance so far. I've designed 3 different 2 way speakers (1 was a center channel MTM), two of which are on my blog post, and compared the impedance to the actual measurements from DATS and they were spot on. Well, within the limits of part variance.

That does not mean something isn't wonky though.

[Chris Roemer]

Right. So I'm thinkin' that if the woof & tweet Zs look the same (before summing), then the discrepancy must (could) be in their summing (which might be due to the algorithms that sum the Z-data in relation to the Z-phase).

[Brian Steele]

Could the imported Z file be incorrect? Double check each driver's impedance by itself. Otherwise, come over to the XSim discussion at DIY. The author hangs out there often.

I'm using the same driver impedance data in both the WinPCD and XSim simulations.

[Chris Roemer]

So . . . it sounds like maybe the Z phase data doesn't sum exactly the same?

[ErikSquires]

Could the imported Z file be incorrect? Double check each driver's impedance by itself. Otherwise, come over to the XSim discussion at DIY. The author hangs out there often.

[Brian Steele]

I'm pretty sure you probably got a decimal point wrong.

Trouble shoot it by removing first one filter (HP or LP) , then the other and comparing the impedance plot. You should see one section glaringly different. Alternatively you may need to substitute a 50-ish ohm resistor to keep things more normal.

Nope - checked the x-over components several times over. It's a pretty simple x-over, with just three elements on the woofer and three on the tweeter. Also, when I look at each section separately, the impedance curves match for both the woofer and tweeter sections. It's just the combined impedance that seems to be off.

[ErikSquires]

I'm pretty sure you probably got a decimal point wrong.

Trouble shoot it by removing first one filter (HP or LP) , then the other and comparing the impedance plot. You should see one section glaringly different. Alternatively you may need to substitute a 50-ish ohm resistor to keep things more normal.

[Brian Steele]

I'm running into a strange issue comparing the results from WinPCD with those from XSim.

The first issue is minor differences in the predicted impedance curves. In this particular example (see pictures below) XSim is predicting a much higher high frequency impedance peak than WinPCD is predicting, for the same x-over components and driver impedance data. It's also predicting a slightly different response through the x-over region, which I guess is to be expected if the predicted impedance response curves are different.

[ErikSquires]

Just FYI, I have added the companion speaker, the MTM LM-1C XSim files to the same discussion thread, for anyone who would like to look at those as well. The LM-1C measurements were probably a little off in terms of absolute level, and there was no near-field measurements at all, however it should prove educational.

[ErikSquires]

I"m surprised if there are not already very good video tutorials here at PE.

I can however offer you a little help with my blog on crossover basics.

http://speakermakersjourney.blogspot.com/2016/02/crossover-basics.html

I suggest you try those examples with XSim first. Play with the parts, and play with the blocks. Get a visceral feel for how the components fit together. Once you get past that, read about speaker distance. After that, try the A26 crossover simulation, or the LM-1.


Best,


Erik

[viyohepav]

hi eric, tq verymuch
if i may suggest, can you do a video tutorial with example for a dumb like me will be perfect.
i tried a view free software including excell spreadsheet out there but i always got stuck.
i hope this software will be something better.
maybe a video example for crossover design for high pass or low pass, and
a video example for crossover design for with 2 way TM.