My first OB project: Fast, cheap, and great-sounding!

[mstap42]

Some background and a word of thanks...

A while back, I asked y'all for thoughts on some Magnavox console drivers. (See the thread Advice on Magnavox console speakers ... )

Everyone who responded offered something interesting and encouraging. The bottom line was these are good speakers. Worth keeping. Better than the little bit of cash I could get for them. Plus, I can do something pretty easily, starting from scratch and learning as I go with help from AKers.

How true that is, and how grateful I am for all the help and encouragement.

As I write, I'm listening to Pat Metheny Offramp on my new OBs. Earlier this evening, I auditioned Peter Gabriel, Holly Cole Trio, Kate Rusby, Dire Straits, Shannon Curfman, and Rickie Lee Jones.

My teenage sons like them. Even my wife does...as long as the music isn't jazz. She's being very patient about the optics--these prototypes are raggedy-looking.

But the sound is really, really nice. All I can say is thanks!

Design and rationale...

Goals: Cheap. Fast. And did I mention cheap?

Baffle material: 1" Styrofoam sheeting
Fasteners: zip ties
Configuration: One OB and one H-frame (I'm already forming opinions on those. More about that below.)
Crossover: 2nd order Linkwitz-Riley @ 3,000 Hz.

Cost: Less than $40
Drivers: Gift from a friend's basement cleanout
Sheeting: Dumpster dive
Fasteners: Zip ties, $3.00
XO components: $30 - PE buyouts, including shipping
Crimp disconnects $3.00
Wire (red, black), stranded, 20 AWG: On hand
Plywood mounting blocks: On hand
Hot glue: Two sticks, one on hand. Bought a bunch more for $6.

Rationale
Baffles: Easy to build and reconfigure
Configuration: OB can quickly become U-frame. H-frame can become U-frame or OB.
Crossover frequency:
I thought on this quite a while, taking into account the advice and comments in the earlier thread. Plus some, in a new thread, that got vaporized a couple of weekends ago when AK was down a couple of weekends ago.

Stretch goal, but can't afford it: 3-way, active XO, tri-amp. I can't afford active, nor at the moment the bandpass components. (That would double the XO outlay, leaving aside the need to buy some mids. I have some 6-1/2" mid/bass that might work, but they're much less efficient than the horn and alnico woofer.)

I considered 800 Hz. But then there are the limitations of the horns. So iI looked at 2,000 Hz or above. The woofers are OK at least to there. And the horns don't honk too badly at that point.

Having decided on 2,000 Hz as the minimum, I measured the DC resistance on the drivers. Horns - 12 Ohm nominal, 13.2 actual. Woofers - 4 Ohm nominal, 3.8 actual. (Lucky me!. The LF XO components for 8-Ohm would have cost a good bit more.)

2nd order was a given. Butterworth or Linkwitz-Riley? Butterworth was cheaper, but crossing above 2,000 Hz -- the higher end of the woofers' range and near the lower end of the horns' audibly pleasing range -- I decided to spend a little more for the L-W for its flatter curve at the XO point. No point calling attention to the weaknesses of the design any more than necessary.

All that remained was to shop for components. I thought of DIY -- unwinding motor coils -- for the inductors. I got 15 yards out of an old vacuum cleaner motor. Not enough.

Luckily, PE has some buyout inductors. Ferrite core. Cheap cheap cheap. Ferrite core -- another compromise, especially for the high pass filter.

Optimal crossover point given budget and available parts from PE: What XO point would give me inductor values AND poly cap values close to those available? To get this, I mapped the component values for every 500 Hz from 2,000 to 4,500. Then I crosswalked those to the PE availability.

Happily, the L-W values matched up better than the Butterworth values. At 3,000 Hz, I could pair inductors in series for L1 and L2 and caps in parallel for C2 to get within 0.02 of all calculated values. (Realizing that not all calculators figure things the same way.) No worries, with 5% tolerance caps and something like 10% on the inductors.

Sound? Sweet, detailed highs. A little forward, but not annoying. No ringing. Maybe a little "round" in the mids, but my tone controls compensate well enough. Bass? As predicted, the H-frame extends the bass. But off-axis is important because my room isn't large. I will end up chopping the front off and coverting to U-baffle.

Sorry no pics, but I will post some. Stay tuned!

Pics added ...

 

[darg]

Hey, that's cool. Keep it simple and easy to modify, that's the way to go.
That way, when you think you have the best setup it's easier to get it transferred into wood. Just make sure to have the H or U air/sound "tight". That way it will give you best results. Let us know how the different variations sound.

Enjoy

 

[nfafan]

You have a very forgiving wife. I can't even get KLH-32s or Walsh 2s out in the public eye - has to be hidden in the man-cave.

 

[BrocLuno]

I guess your wife understands trial and error

 

[mstap42]

You have a very forgiving wife. I can't even get KLH-32s or Walsh 2s out in the public eye - has to be hidden in the man-cave.

In return for tolerance, she allows me a man-corner.

I guess your wife understands trial and error

I sure hope so. I'm surely trying her limits these days. Sooner or later I may be shown the error of my ways!

Jokes aside, something has changed over the last year. Maybe she sees the pleasure this audio hobby gives me, our kids, and even our friends -- including a bunch of teenagers who think all my retro gear that's older than most of their grandparents is hip and cool.

 

[PR Audio]

mstap42, glad it worked out! Told you those components would make a great OB system. When you transfer your final design to wood, take your time with designing the baffle, don't be afraid to experiment with shapes, though some electronic compensation maybe needed. Also, go for the WAF by properly finishing the wood, weather its painted, veneered, or stained; makes life easier at home. For claiming to be a newbie you know quite a bit!

 

[mstap42]

Hey, PR Audio, if you and a few others like CopperWizard and darg hadn't intervened, these would have been long gone and I'd not have had such a blast the last couple of weeks.

Even in their unrefined state, the drivers are impressive. I have a dozen or so test tracks and albums that I know like the back of my hand. Records bought it in the '70s and CDs of the same. For example, Joni Mitchell's Court and Spark. Always I have suspected that there was more to it than I could hear. Little hints and allegations of subtleties hidden in the mix. Rhythm section, strings, horns, John Guerin's guitar riffs, etc.

My suspicion was confirmed right away when I played the CD. All kinds of little stuff, just on the edge of audibility, magically appeared. Same with Steely Dan's Aja and the few others I have had time to audition. These are very, very nice indeed. Unfortunately, the verdict for my beloved TimeWindows, that do many things so very well, isn't great. They hide many details. (Now I understand what so many have said about those Philips tweets.)

You said I know a lot more than I let on ... well, I didn't at first but I have been reading and reading and reading. Now I'm up to speed enough to banish the fear and let the fun begin!

Thanks for the counsel on home life. You got it. WAF is always in mind. Even she hears the difference, and she's a visual person. And ... she's getting some skin in the game. Recently she said, "You know those old-growth timbers you salvaged from that barn? You could make your baffles from those. They'd look great!" Both of my sons have suggested baffle tweaks. They may even make some mods themselves. It just keeps getting better and better!

My house is a busy place with no dedicated listening space. Hard to find even 15 or 20 minutes to listen. But with the family involved, I think even that may improve.

Stay tuned!

 

[mstap42]

The date code on the 15" woofs is 1056 142.
Looks like Gefco Manufacturing Company, Grayslake, IL , October 1961.

 

[PR Audio]

Glad it is working out! If you get the family involved it gets easier. With a 13 yr old son I know, and the Gestapo likes to hire an interior designer friend of hers every 3 yrs or so, so I know everything about WAF...Though I have a media/HT room, recently I claimed a room for my audio hobby, small at 9'x9' but mine.

Looks like you have definitely reading and doing research, that is how I have been learning. That, plus trial and error...after error, after error, after error...

 

[mstap42]

Further progress...

Today I modified both baffles. I changed both the H-frame and the OB into U's. Pics attached. The change accommodates these speakers to the small space I have for them. Happily, and unexpectedly, the change also tightened up the bass a bit.

Another change was to reverse the horn polarity on the XO. I had missed that detail about 2nd order L-W crossovers when wiring them up the first time. Props to sregor for that suggestion in another thread. The difference was quickly and obviously apparent: the treble is considerably more detailed. [The attached pic shows the original XO wiring. Now the horn's + (red wire) is connected to XO ground.]

Finally, I reduced the amount of baffle around the horns (three sides of one horn and only the top on the other.) Based on some reading about baffle configuration, I hope that change will improve the soundstage. (My beloved Time Windows do soundstage so very, very well.) I have not had time yet to listen. Will report back here with results.

 

[PR Audio]

Sounds like you are making progress...Remember that a foam prototype like yours is bound to have leaks. When you build the real ones with the W frame properly sealed, bass will tighten/deepen further and be somewhat "faster". Yes, when using 2nd and 4th order XO's one driver needs to have polarity reversed; though there might be exceptions that I haven't encountered yet.

 

[mstap42]

Project update:

The drivers are now in plywood baffles: 3/4" X 19" W X 30" H, canted back at 3 degrees.

Big sonic improvement over the foam sheeting prototypes. I hear more detail and punch in the low end. The horns (crossed at 3 KHz) don't "stick out" as much as they did before.

Last night I cued up some of records -- Dave Brubeck (1954), Sarah Vaughan (1955), and Modern Jazz Quartet (1961). Then I played a couple of tracks from Natalie Merchant's Tigerlily CD (1995) We had company, so I couldn't sit and just listen. Our guest said it was "amazing" even in the next room. I found a way to wander in front of the speakers a few times and that was even better.

The trade-offs with my DCM Time Windows are obvious. To get the OBs' detail in the bass, even at low volume, I give up the TWs' really good soundstage. The TWs' "sweet spot" is pretty forgiving. But I have to be seated to enjoy the OBs at a distance nearer than about 8'. It may help to cant them back another degree or two. The stands are deep enough to handle it without increasing the risk of tipping.

 

[darg]

That looks good. How happy are you with the performance of the 15" around 2khz? Mine got a little bit shouty with a considerable jump between 1500 and 2200?

 

[StimpyWan]

With an OB design, one thing that helps bass response, is to make the baffle as wide as possible. Less front to rear bass cancellation, the wider the panel. Typically as least 20 or more inches, and the wider the better. While I expect your wife wouldn't allow 36" wide speakers, it works just as well, to fold the baffle into your U shape, since the sides add and contribute to the baffle width.

Also, it won't hurt to make the U wings even deeper. And to help improve dispersion and imaging, the U shape can have unequal length sides. One U wing can be short, with the other longer. Usually, face the long side of the U wing to the outside wall, and the short U wing towards each other. Better dispersion towards the center, that way, to improve imaging.

And thanks for the pictures and write up. A fantastic project so far. I hope the experiments continue to produce sonic wonders!

 

[Poultrygeist]

Great job!

Listening to OB's in near field is always a joy.

 

[GD70]

Fantastic project and thread! Nice after dinner read!
Once you have the configuration you're happiest with, then you can use some nice ply and veneer, get that WAF higher!
On the side pieces, they look to be at 90 degrees to the front baffle. Did you experiment with angling the sides at all, and if so what did you learn sonically?
Cheers!
Glenn

And yeah, the TW's are. Rey nice sounding speakers too! Are yours early or later? Mine are early production.

 

[mstap42]

That looks good. How happy are you with the performance of the 15" around 2khz? Mine got a little bit shouty with a considerable jump between 1500 and 2200?

darg: In an earlier post, you clued me in about this. I don't know about 1,500 and 2,200...I haven't listened to enough to say for sure. It's not an issue with my Pat Metheny recordings and probably (guessing) not so much with lots of my other music.

But there's a definite bump from 1,500 down to around 700. A good example is Modern Jazz Quartet's The Sheriff. The higher notes of Milt Jackson's vibes solos are already very prominent in the mix, and the effect is even more pronounced when played through my OB rig. That may suggest similar issues with instrumental solos in my handful of other small-ensemble jazz recordings.

As a reality check, I looked up the frequency range for vibes. The standard setup is 3 octaves, F3 to F6, or approx 175 Hz - 1,400 Hz. The top octave, starting at F5, is at about 700 Hz, confirming what I think I'm hearing.

 

[Lavane]

I have those exact woofers and horns from a maggie console I plan to build OB's with. The amplifier from the console is the moddified 9303 in my avatar. im trying to finish up a maggie 175 thats on my bench now so I can get started on the OB's.

 

[mstap42]

With an OB design, one thing that helps bass response, is to make the baffle as wide as possible. Less front to rear bass cancellation, the wider the panel. Typically as least 20 or more inches, and the wider the better. While I expect your wife wouldn't allow 36" wide speakers, it works just as well, to fold the baffle into your U shape, since the sides add and contribute to the baffle width.

Also, it won't hurt to make the U wings even deeper. And to help improve dispersion and imaging, the U shape can have unequal length sides. One U wing can be short, with the other longer. Usually, face the long side of the U wing to the outside wall, and the short U wing towards each other. Better dispersion towards the center, that way, to improve imaging.

And thanks for the pictures and write up. A fantastic project so far. I hope the experiments continue to produce sonic wonders!

StimpyWan: Great suggestions, thanks! Interesting you should mention folding the baffle. I factored this in when laying out my cuts on a sheet of 4' X 8' B/C plywood. With space an issue, folding seemed like a possible answer. The plywood had some flaws, but I was able to lay out my cuts to allow for 7" wings on either side, with an effective width of about 33". I envision mounting them on 30" piano hinges. Haven't thought much yet about how to stabilize them in various "open" positions.

Your suggestion of unequal U wings is the first I've read. Mine are about 10" along the bottom edge and maybe 8.5" at the top edge (see the new pics added to the other ones above). It would be easy to add longer ones outside, keeping the 10" ones on the inside, maybe even cutting them to 8" or so at the base.

 

[StimpyWan]

Your suggestion of unequal U wings is the first I've read. Mine are about 10" along the bottom edge and maybe 8.5" at the top edge (see the new pics added to the other ones above). It would be easy to add longer ones outside, keeping the 10" ones on the inside, maybe even cutting them to 8" or so at the base.

Danny Richie, over at GR Research, is a big proponent of the unequal length wings. Many of his OB designs, have staggered wings. There's some good reading, on his forum, at Audio Circle. Worth a look.

http://www.audiocircle.com/index.php?board=16.0

http://gr-research.com/