DIY OBs
Per brook's request: A thread specifically about these. Probably best to let the DIYaudio crew chit chat before I say anything beyond what's in the article.
Jeff's project, i hope, will get more people trying FAST OB. Martin's article certainly goosed the idea, and it has a great following here. Hopefully Jeff's article will give the concept wider exposure and we can have more diy addicts to help grow our hobby.
dave
dave
Badman, do you have any polar response graphs on the speaker? I built a similar speaker not long ago and the polars left a lot to be desired. I ended up liking them best facing nearly straight forward and listening right between them 2 meters away. There was certainly a unique to OB sense of space with the design, but I could never get the image I had hoped they'd produce. I was getting a lot of unilateral (guess that would be a good word for it) reflections in the 1-4kHz region. At first I attributed this as clarity, but w/i days it wore on me and became coloration.
Now I've gone to a 2-way box with a waveguide and I can't look back. I could get an OK stereo image and a decent tonal character with that style OB, but nothing like the imagine, space, tone and dynamics I'm getting now. Nothing seems to sound hot or muffled ever.
Maybe that specific design does something better than mine? That little fostex may have better dispersion in the upper octaves? From the description it sounds like it does, but since there aren't any graphs I can't interpret for myself. IOW, I'd sure like to see some graphs as I still have my baffles and hate to destroy them after all the work I put into them. It would be nice to pop in that little Fostex and enjoy all my effort.
Thanks,
Dan
Now I've gone to a 2-way box with a waveguide and I can't look back. I could get an OK stereo image and a decent tonal character with that style OB, but nothing like the imagine, space, tone and dynamics I'm getting now. Nothing seems to sound hot or muffled ever.
Maybe that specific design does something better than mine? That little fostex may have better dispersion in the upper octaves? From the description it sounds like it does, but since there aren't any graphs I can't interpret for myself. IOW, I'd sure like to see some graphs as I still have my baffles and hate to destroy them after all the work I put into them. It would be nice to pop in that little Fostex and enjoy all my effort.
Thanks,
Dan
Hi Dan:
My polar response testing is very rough- RTA and a mic. I don't have graphical polar abilities at this time. I need to get HolmImpulse up and running, but for my purposes, I'm satisfied with my method, it's just not good for publishing. I use a variety of 'fixed' locations with the mic, as well as simply free-handing it to check for narrow range anomalies to focus on.
The close CTC spacing is actually about equivalent to a typical 8" 2 way with a 4" flange tweeter, and the felt around the tweeter in conjunction with the absorbtion on the woofer really made for a much more consistent behavior, though there's still a little increase in power response at XO. The first order XO also helps average this out somewhat.
My polar response testing is very rough- RTA and a mic. I don't have graphical polar abilities at this time. I need to get HolmImpulse up and running, but for my purposes, I'm satisfied with my method, it's just not good for publishing. I use a variety of 'fixed' locations with the mic, as well as simply free-handing it to check for narrow range anomalies to focus on.
The close CTC spacing is actually about equivalent to a typical 8" 2 way with a 4" flange tweeter, and the felt around the tweeter in conjunction with the absorbtion on the woofer really made for a much more consistent behavior, though there's still a little increase in power response at XO. The first order XO also helps average this out somewhat.
I did similar tricks and graphed it every way. The felt definitely helps off axis. I didn't get any increased power response around the crossover though. Mine power response issue was much higher in frequency and I used a 4th order LR around 200Hz.
no felt:
11.25 dregrees off axis:
45 degrees off axis:
felt added:
45 degrees with less smoothing:
you can see the concentration of in room reflection with this graph:
Hope that helps,
Dan
no felt:
11.25 dregrees off axis:
45 degrees off axis:
felt added:
45 degrees with less smoothing:
you can see the concentration of in room reflection with this graph:
Hope that helps,
Dan
Last edited:
I've got the drivers for this project here, and, somewhat concerned by Jeff's comments about the possibility of the demagnetizing of the Alnico, I just this morning tested the JBLs with WT3. Fs was identical at 25.57Hz, and Qts was close, so I'm good to go.
I have no clue on the crossover, however, so if someone could 'splain it, I'd appreciate it. TIA - Pat
I have no clue on the crossover, however, so if someone could 'splain it, I'd appreciate it. TIA - Pat
Hi Tubesguy:
What's the question on the crossover? It's a notch filter and a series capacitor on the tweeter. The woofer (aside from the notch) is allowed to run full-range, but firing into the felt absorbtive pad to suck up some of the highs, where it's very beamy. By using the acoustic treatment rather than an electrical filter, the 123a acts very much like an ideal driver going through a first order electrical filter.
Hope this helps!
Helps quite a bit, and thanks! Now I just have to save up for that big honkin' inductor, especially if I follow the advice regarding the gauge of the wire.
By the way, the part number for the felt is actually 87415K53, with the number mentioned in the article having omitted the "1".
Thanks for posting this first of all.😉
What ended up being the final crossover point, somewhere between 500hz and 2000hz?
What ended up being the final crossover point, somewhere between 500hz and 2000hz?
It's essentially a 2kHz textbook first order. Not really possible with most drivers because of bandwidth limitations, but the 123A can play fairly cleanly to 5-6k and the FF85k is well within its limitations with such a high high-pass.
- Status
- Not open for further replies.