I didn't know you could get zma files from an equivalent circuit. Would you mind describing the process?
Draw the equivalent circuit only in XSim using datasheet and for the dcr values of coils and capacitors use the smallest value you can.
Then in Impedance window under file, by exporting system zma, you will have the drive unit impedance ready to be used
for your driver.
Then in Impedance window under file, by exporting system zma, you will have the drive unit impedance ready to be used
for your driver.
I finally found the time to do some more on this, and did the mids' ZMA from its equivalent circuit, as before this its FR was looking just like the tweeter with 2 peaks and a big valley. Turned out ok. Did the bass driver also even though the FR was looking ok and not much change there.
My next question is: if I reverse the polarity of the B110 only as per the factory DN12 crossover, there is a big valley in the system FR (black line) between the T27 and the B110. If I also change the polarity of the T27 this valley disappears, but the DN12 doesn't have it reversed. Any ideas about whats going on here?
My next question is: if I reverse the polarity of the B110 only as per the factory DN12 crossover, there is a big valley in the system FR (black line) between the T27 and the B110. If I also change the polarity of the T27 this valley disappears, but the DN12 doesn't have it reversed. Any ideas about whats going on here?
Attachments
Those are not easy drivers to work with. I did do a try (from start) myself with the data provided and came to this:
This is only working off course if the data provided is correct. But it gives also an id how to solve the issues you have with your corssover.
This is only working off course if the data provided is correct. But it gives also an id how to solve the issues you have with your corssover.
Wow thank you Waxx, that looks pretty smooth! I will keep it for future reference as I modify the crossover in the future. What I am trying to do first is create a fairly smooth and coherent baseline with the factory DN12 crossover std values, and try to learn why my sim is not showing a reasonable result as the factory crossover must surely provide. Then from this baseline I will make small changes in the future, such as a notch for the unruly B110 mid, perhaps an aB section for the T27 etc. Any thoughts on why I need to reverse polarity on the tweeter when the DN12 doesn't do this?
You'll learn more if you select phase to be shown.. https://www.diyaudio.com/community/threads/xsim-configuring-the-plot-window.406106/
The factory crossover is not smooth on those, it never was to start. The Kef Concerto's (the original ones) were soon modified by diy'ers because the crossover was made wrong. You can find a lot of info about that on the web if you google. Kef in general makes good speakers but some were not, and the Concerto's were among them, mainly due to a bad designed crossover.
I'll turn on phase and try to work out what its telling me. Thank you. Phase has always been one of my (many) blind spots.
Waxx, Ive never used the DN12 myself as I built some TLs with SOTA/Frisby crossovers a long time ago but wasn't fully happy with them. I have some ideas about how to improve them by 'borrowing' sections from other versions of the crossover, but really want to be able to simulate it all in Xsim, otherwise I could be chasing my tail forever. I did plug in some spare DN12s I aquired a few weeks back and have to agree with you - I didn't like what I heard, except for the bass which was pretty good. They are pretty old though and could use a recap.
Ah that makes sense. So delay could be throwing everything out of phase? I missed that. I'll read up on how to set up delay. Thanks again.
In short, let us assume midrange unit is behind the tweeter about 1", and woofer perhaps 2".
edit: have you simulated baffle step?
Would you attach midrange and woofer datasheets?
edit: have you simulated baffle step?
Would you attach midrange and woofer datasheets?
Last edited:
While it could be a factor, you'd be getting ahead of yourself in working out why one wants reversing and the other doesn't. It can be normal for it to be that way depending on certain choices. (It's only a guess without seeing phase.)
Btw, a friend (not on this forum) supplied me those, he did redesign and rebuild the filters of his 1975's Kef Concerto more than a decade ago on measurements of the drivers in the box (he is a retired sound engineer with knowledge about speakers) and after some conversation he don't mind that i share his crossover for this. These are his main speakers and this version is in daily use for about 12 years now. He did change the crossover a few times during the decades that he has this speaker, but redid it with simulation software (i had to help him with it then) as before it was more the old "trial and error" way (before simulation software). I heared this setup quiet some times myself and it sounds very good.
Here's the driver phase of the factory DN12, and with the T27 inverted. Still trying to understand it.
Attachments
Data sheets above. I haven't simulated baffle step as still trying to come to grips with factory crossover and learning to use Xsim.