I suspect DCR on the main woofer coil is the answer here. I used 0.257 ohms (I think speaker workshop changed it on me, should have been 0.21 ohms) which is the dcr of jantzen pcore 2.5mH. I thought PE had one but I can't see it there now...
dcr on the 200uH was 0.177 ohms (not sure where I got that from, as I was sure I put in real values of jantzen coils, the erse 18ga on PE is 0.18.
330uH on tweeter had 0.2ohms DCR
Tony.
In my PCD simulation, I use the Parts Express ERSE 18-gauge laminated I-core for the 2.5 mH with 0.29 ohms DCR. For the notch filter, I used JantZen 18-gauge air core with 0.21 DCR.
The tweeter coil is 0.35 mH JantZen 20-gauge air core with a 0.40ohm DCR.
I use resistors to tailor the total resistance which worked out well for me.
The FR curve is as flat as it can be.
What I try jRevae's partial BSC, I may end up with a smaller than 2.5 woofer coil.
Yes that would be good, as 2.5mH is getting big and expensive! Jreaves network with 2.2mH worked well with my original tweeter network. it was pretty much interchangeable! with the slightly modified current one it would probably need tweaking a little but should work too. I was rather surprised that two fairly different woofer networks went well with the one tweeter network, but I guess this demonstrates what I said earlier, that if you get the curve matching the target then things start to fall into place 
Attached are sim results using the highpass I last posted, and a slightly modified jReave lowpass (to match the move to 2.2Khz)
The phase tracking is better with this combination (though of course we are guessing the acoustic centre).
Jreave's low pass is more elegant than mine and should be cheaper too, it was only the phase that worried me a little but the overall phase is fine.
You will need to go back and look at jReaves original post to see how to implement this in PCD (some trickery required due to the inability to specify DCR for L in an RLC network) Being able to use a 2.2mH coil I think will drop the cost quite a bit.
I should also do a combined network so I can get the impedance curve.. speakerworkshop doesn't allow use of the optimizer if there is more than one driver in the network, hence modeling separately and combining the result.
Tony.
Attached are sim results using the highpass I last posted, and a slightly modified jReave lowpass (to match the move to 2.2Khz)
The phase tracking is better with this combination (though of course we are guessing the acoustic centre).
Jreave's low pass is more elegant than mine and should be cheaper too, it was only the phase that worried me a little but the overall phase is fine.
You will need to go back and look at jReaves original post to see how to implement this in PCD (some trickery required due to the inability to specify DCR for L in an RLC network) Being able to use a 2.2mH coil I think will drop the cost quite a bit.
I should also do a combined network so I can get the impedance curve.. speakerworkshop doesn't allow use of the optimizer if there is more than one driver in the network, hence modeling separately and combining the result.
Tony.
Attachments
I hope that the wonderful work on the crossover by many posters, particularly jReave and Wintermute, will encourage DIYers to look at the drivers used in this project. I picked them to be the good value drivers for under $250 total.
The crossover should be applicable to both vented or closed box designs.
The crossover should be applicable to both vented or closed box designs.
Add a picture of the finished crossover.
I cannot emphasize enough the importance of using good quality component. I used those from Madisound and Parts Express. Not cheap but affordable. The circuit board was from the bay, but I cannot find it anymore. The crossover PCB from Parts Express is junk.
I cannot emphasize enough the importance of using good quality component. I used those from Madisound and Parts Express. Not cheap but affordable. The circuit board was from the bay, but I cannot find it anymore. The crossover PCB from Parts Express is junk.
Last edited:
Nice
Tony.
It sounds exactly as expected and beautiful.
Of course, I am very biased. Some more pictures of the finished speaker.
Last edited:
To be honest. The speakers would not have sounded as good as it does without your help on the crossover. I am greatly indebted to you.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.