Sorry for the sloppy pic, working from my phone.
Raise this section ~3db and see how you like that. Start by increasing the inductor size on your woofer.
@jhollander said:
When I see a low value inductor my first thought is that its probably going to sound a bit thin. Did you splice in the woofer nearfield?
I did not.
I have tried using FRD response blender but wasn't able to export an FRD file. Do you have a program that works for you?
Sorry for the sloppy pic, working from my phone.
Raise this section ~3db and see how you like that. Start by increasing the inductor size on your woofer.
I am going to take some more measurements tomorrow, including nearfield of the woofer, and hopefully I can get something more accurate on the low end and go from there.
I would try and relax the transfer function on the tweeter, it is a common mistake early on in the learning curve to try and fill in a depression in the treble by creating a peaking high pass filter such as the one you have. It unnecessarily stresses the tweeter and consumes extra parts. I think you will find the depression you were filling will fill in off-axis - if it is flat there on-axis it will "bloom" off-axis. This translates to bright and can cause you to pull your hair out trying to find the right padding amount.
@jr@mac said:
I would try and relax the transfer function on the tweeter, it is a common mistake early on in the learning curve to try and fill in a depression in the treble by creating a peaking high pass filter such as the one you have. It unnecessarily stresses the tweeter and consumes extra parts. I think you will find the depression you were filling will fill in off-axis - if it is flat there on-axis it will "bloom" off-axis. This translates to bright and can cause you to pull your hair out trying to find the right padding amount.
That's better, still a bit of a shelf there. Try going third electrical on the tweeter. That tweeter can be a royal pain to get right on the bottom end due to the waveguide corner frequency coinciding with the impedance peak at Fs.
@jr@mac said:
That's better, still a bit of a shelf there. Try going third electrical on the tweeter. That tweeter can be a royal pain to get right on the bottom end due to the waveguide corner frequency coinciding with the impedance peak at Fs.
I tried third order. The transfer function on the tweeter looks much better but I struggled with the crossover. The fact that the woofer is at the end of its rope is proving difficult for me on this. If I could get the woofer to go any higher I think this could work.
Took another swing at 4th, I think this looks a little better than the last one
Remember for the coil you can always unwind some of it's too big. With an air core if it's too small you can increase inductance by putting an iron/laminate core through the middle, even a bolt would work. Worth noting, never ground the core; it should always be electrically floating.
Here are some things that might help:
WAGO 221 Series 25pc Compact Splicing Wire Connectors Assortment Pack | Includes (10x 221-412), (10x 221-413), (5X 221-415) | 221 Series Assortment with Case https://a.co/d/bZGzqcu
Sumnacon 10 Pcs Multimeter Electrical Test Dual Lead Test Hook Clips - 5 Colors Silicone Flexible Test Cables for Electronic, Component, Testing Connecting https://a.co/d/61vhxsJ
I do not mean literally twisting leads, that is just how I refer to the process. I use home made test leads - the factory built ones I used had ridiculously high resistance, high enough to cause a lot of head scratching on tweeter levels post-prototyping.
I used largish alligator clips and 16awg wire to make mine. I also use a lot of bare alligator clips, especially on the tweeter circuit. Gets me closer to point-to-point.
Comments
I made a mistake on the diagram when I drew it in Xsim:
The frequency response from PCD was with the 1.5mH coil on the woofer.
Just trying out PCD, still find it easier to understand when I draw it out in Xsim.
Can you explain what you are seeing? I understand BSC, I am relatively new at this and not certain what needs attention.
It looks good to me. I might suggest making R3 a little smaller to fully minimize that peak in the woofer.
Sorry for the sloppy pic, working from my phone.
Raise this section ~3db and see how you like that. Start by increasing the inductor size on your woofer.
When I see a low value inductor my first thought is that its probably going to sound a bit thin. Did you splice in the woofer nearfield?
I did not.
I have tried using FRD response blender but wasn't able to export an FRD file. Do you have a program that works for you?
I am going to take some more measurements tomorrow, including nearfield of the woofer, and hopefully I can get something more accurate on the low end and go from there.
I would try and relax the transfer function on the tweeter, it is a common mistake early on in the learning curve to try and fill in a depression in the treble by creating a peaking high pass filter such as the one you have. It unnecessarily stresses the tweeter and consumes extra parts. I think you will find the depression you were filling will fill in off-axis - if it is flat there on-axis it will "bloom" off-axis. This translates to bright and can cause you to pull your hair out trying to find the right padding amount.
More like this?
That's better, still a bit of a shelf there. Try going third electrical on the tweeter. That tweeter can be a royal pain to get right on the bottom end due to the waveguide corner frequency coinciding with the impedance peak at Fs.
I tried third order. The transfer function on the tweeter looks much better but I struggled with the crossover. The fact that the woofer is at the end of its rope is proving difficult for me on this. If I could get the woofer to go any higher I think this could work.
Took another swing at 4th, I think this looks a little better than the last one
Much better, now to dial in a few more db of baffle step compensation and concurrent padding on tweeter.
On your attempt with the 3rd order, did you try reversing the polarity on the tweeter?
I did, there was little difference.
This is the reverse null on that last 4th order crossover
Took a quick measurement of the woofer with a really long gate time and it looks like this:
What would you do to bring down that bump at 150hz?
Nothing about 150hz, that is an artifact.
More concerned with the acoustic energy below the XO frequency.
Thanks for your help and your patience JR.
Increased the coil on the woofer from 1.5 to 2.5mh
I would build it and give it a listen.
Yeah, I have a shopping cart loaded up with the crossover parts I need, just need to settle on a coil for the woofer.
Yep, time to start twisting leads.
Remember for the coil you can always unwind some of it's too big. With an air core if it's too small you can increase inductance by putting an iron/laminate core through the middle, even a bolt would work. Worth noting, never ground the core; it should always be electrically floating.
Here are some things that might help:
WAGO 221 Series 25pc Compact Splicing Wire Connectors Assortment Pack | Includes (10x 221-412), (10x 221-413), (5X 221-415) | 221 Series Assortment with Case https://a.co/d/bZGzqcu
Sumnacon 10 Pcs Multimeter Electrical Test Dual Lead Test Hook Clips - 5 Colors Silicone Flexible Test Cables for Electronic, Component, Testing Connecting https://a.co/d/61vhxsJ
I do not mean literally twisting leads, that is just how I refer to the process. I use home made test leads - the factory built ones I used had ridiculously high resistance, high enough to cause a lot of head scratching on tweeter levels post-prototyping.
I used largish alligator clips and 16awg wire to make mine. I also use a lot of bare alligator clips, especially on the tweeter circuit. Gets me closer to point-to-point.
Parts have been ordered.
I have plenty of alligator clips and Wago connectors on hand, so I am good there.
Thanks for everyone's help, I will pick this back once I have the rest of my components in hand.
After the break for Halloween festivities, we now join our regularly scheduled programming.
Got one crossover wired up today.
Pretty happy with this, going to get the other one wired up and have a listen for a bit.
Where do you bury your dead until next year?
If you are asking about the Halloween decorations, they all fit in these bins in the garage
As far as the actual bodies go, are you familiar with the Yellowstone zone of death?
No, but I am eager to learn as there are these people who . . .
F350 fits in the garage?
Crew cab long bed
Here is where I ended up with these (the canyon just under 2k fills in a bit off axis):
I think that's where these will end up.
They sound pretty good for what they are.
I always wanted to redo a vintage pair, and now I have.
Solder 'em up, seal 'em up and on to the next.
Thanks to everyone that helped out on this thread.
Solid build ready for another 40 years of service! Good job, I enjoyed watching this one come together.