Dayton 3 way made with 1" thick, native white oak

[poorwill]

Two new speakers will be used as the main fronts in surround sound system. An 18" sub woofer provides strong base.

[poorwill]

More pictures

[poorwill]

XSim

[Chris Roemer]

So ...
a 3 cu.ft. 8" w/NO port? How much baffle step are you shooting for?

[poorwill]

Some of the volume at the bottom will be boxed off for the cross over. Cross over was designed with the infinite baffle measurements from part express. I do have a microphone but decided to not build a prototype. I do not know how to account for baffle step, may be you can point me in the right direction? I did model it with BoxSim 2.0 with the baffle locations and port but I am not sure I can trust the results or my inputs.

[Chris Roemer]

This design does not look like it will give satisfactory results.
The mid and tweeter will likely run quite hot (unless this is an "in-wall" design).
There's a "sticky" (near the beginning of this forum) by Paul Carmody giving an outline of the "process" to design a speaker using "factory" measurements (which are mostly "infinite baffle").
Did you model your box design (with the woofer in mind)?

[poorwill]

Chris,
I understand your point that I could have incorporated more or larger woofers into this box. I guess I will be padding the tweeters down to match whatever I get from the woofer. Do you see any thing wrong with 2.0 cuft for the vented volume? The sub woofer cross over is 60 or 80 Hz. Would tuning the port higher than 44 Hz help any?
Thanks for your thought so far.
Greg

[Chris Roemer]

(updated component specs around 2 a.m. on Thurs)

It's not a function of woofer size, or "count". The problem is that the tweeter and mid will send more of their energy "forward", whereas a lot of the bass (which is less directional - based on freq.) will "wrap around" and run behind your cabs. This effect is a fairly gradual decrease in perceived bass output (up to -6dB) that occurs from roughly 1000Hz downward (based on baffle width).

I see your box model now. For your tuning I'd probably go with a 4"id port that's 6" long. I like to vent out the back.
A 3cf (external) box is pretty big, so we assume they'll be floorstanding, pulled out a couple ft. from the wall. Most try to get the tweeter at "seated ear level" (around 30" ?). Maybe you can lower you cross point to your sub w/these, like 40-50Hz?

Try this for your XO (components laid out amp to speaker, L to R - like in your schematic):
2nd order HighPass w/L-pad (for attenuation) - a 6.8uF series cap, then a 0.30mH shunt coil (to gnd), then an L-pad: Series Resistor = 2.5ohms, Parallel Resistor = 2ohms.
2nd order BP (mid) w/L-pad: a 0.70mH series coil followed by a 20uF series cap, then 2 shunts; a 5uF cap and a 5.0mH coil, lastly (next to the mid driver) the L-pad: SR = 3ohms / PR = 6ohms.
"Reverse polarity" is indicated for the mid only, which is pretty common for 2nd order 3-ways.
2nd order LP: a 6.0mH "low DCR" series coil (usually an "iron core" type), then a 20uF shunt cap.

This should get you a lot closer to "fine tuning" range than what you had.

The big (5mH) shunt coil on the mid does NOT have to be low DCR. An 18 or 20ga aircore should be okay.

[poorwill]

The bottom will be removable to adjust the cross over and the back will be removable to adjust the vent. I am expecting a bit of testing based on Chris's comments...

[turn2]

I have used this tweeter and mid and am well familiar with the woofer. You would be well advised to listen to Chris. Your original xo and design work would be screamin' shreiky if used as is. Don't be too attached to it. But don't be disheartened. You can get to a good place with these drivers.

[poorwill]

I am still working on the crossover but it is probable time to move these out of the basement and into the living room. They look and sound good enough for me. The baffle step, diffraction, and basement concrete all are doing there thing and then some.

[DeZZar]

Really need to get these up off the floor for measurements for crossover design.

Are the measurements gated? If not, the measurement needs to be gated to within the earliest boundary reflection to really see what the speaker itself is doing. If you can get 3-4 feet clear in every direction (360) around the speaker and the mic positioned 3x the baffle width away - you'll get much more accurate speaker measurements.

https://www.audiosciencereview.com/forum/index.php?threads/how-to-make-quasi-anechoic-speaker-measurements-spinoramas-with-rew-and-vituixcad.21860/

[poorwill]

I thought about taking them outside, but I am done lifting these 90 lbs beast.

[poorwill]

I did measure these outside on the way to the living room. Attached are the measurements from the two speakers (A & B), 43" away. They did not have identical crossovers. I do not think the two woofers had identical frequency responses either. The smoothest curves have a +/- 3 ms window. The modulated curves were either auto or +/- 10 ms. I measured the bare drivers at 5" and 43" also but I have not finishes implementing any changes.

[DrewsBrews]

This is what I use to give me a rough idea what to start with for baffle step compensation then tweak from there. Gives a decent explanation of the phenomenon and what to look for in the graph. The "BSC" circuit is only needed if you can't get the lowpass of the crossover to do this for you.

Though if it sounds good to you that is all that matters.

http://mh-audio.nl/Calculators/BDL.html