Small Full Range Driver in TL for office

[adwilk]

Hello, all. I acquired a pair of W4-1337 SDF Tang Band 4 “full range drivers some time ago and I want to put them to use. I have a laser cutter and intend to make a design similar to the picture shown to put in my office. The issue it seems, is that the basis of design uses a flared line, by my best math its no where the 1/4 wavelength, and damping looks to be placed conveniently, but probably not adequately. Am I right in thinking that the flared design will raise f3? That doesn't seem to be desirable- especially in a small package. A tapered design should lower f3, right? Do you calculate the 1/4 wavelength on the actual f3 of the driver or the anticipated f3 based on the alignment? I'll be laminating the whole thing together- the middle panels could "house" the damping material hiding it from view? I mostly want to play with the new laser cutter and put the drivers to use. Thoughts, concerns, or should I go back into hiding?

 

[adwilk]

Ok, it just occurred to me that the speaker shown is probably a horn design. Is this the way to go, or should I work out a TL?

 

[ski2xblack]

I think that driver would lend itself to blh, as in the picture, but let's hear what @TLS Guy has to say.

I'm not sure how much the smaller blh types actually act like horns rather than some permutation of tl or reflex alignments, so I'm curious about whatever results you come up with.

Are you considering plexiglass side panels too? Its a cool look.

 

[Pogre]

@TLS Guy is who you want for this one! I just summoned him.

 

[Pogre]

I think that driver would lend itself to blh, as in the picture, but let's hear what @TLS Guy has to say.

I'm not sure how much the smaller blh types actually act like horns rather than some permutation of tl or reflex alignments, so I'm curious about whatever results you come up with.

Are you considering plexiglass side panels too?

Beat me to it!

 

[adwilk]

Ok, fine. I’ll wait for the scolding from Mark!
Yes, I want to use clear material on the sides.

 

[ryanosaur]

My limited knowledge seconds the Rear-Horn moniker the the above design.

As I understand the Quarter-Wave tubes, the Tapered Line (larger at closed end, smaller at the Terminus) will slightly raise the F3 and will make the harmonic resonances easier to tame while being the shortest option in terms of Line Length. The opposite Flared Line (like a Voigt Pipe) will be the longest line length with a lower F3 and more Harmonic distortion to dampen. A straight line (neither tapered or flared) is in between in terms of those criteria.

As an aside, Voigt (as I understand) didn't even think that his Flared design was actually worth building!

Looking at the minutia details, a true transmission line should have the driver placed along the length of the line rather than at the end. That latter, when flaring wider towards the mouth, would be the textbook definition of Rear Horn Loading. In terms of the Driver Placement on a true transmission line, my understanding is that a single driver should be placed ~1/3 along the length of the Line, at the null of the Third Harmonic. Though I haven't seen this specifically stated, if employing a second matching Driver, it would seem that it should likely be placed ~1/5 along the length of the line at the Null of the Fifth Harmonic. Such placements will help cancel out these more destructive acoustic distortions leaving the remaining higher Odd Harmonics to deal with through the addition of the damping material.

 

[Gmoney]

Ok, fine. I’ll wait for the scolding from Mark!
Yes, I want to use clear material on the sides.

lolo, yeah ole Doc can be blunt lol, but he sure does know speaker’s. I for one enjoy reading his post even if I get Chastised a bit .

 

[annunaki]

If you do a design like this find a nice looking way to incorporate a gasket. I see there was no gasket used at all in the picture above. Chances are the pressure is low enough and the path of least resistance is the horn but could possibly cause an issue, likely at high output levels.

 

[adwilk]

If you do a design like this find a nice looking way to incorporate a gasket. I see there was no gasket used at all in the picture above. Chances are the pressure is low enough and the path of least resistance is the horn but could possibly cause an issue, likely at high output levels.

Thanks for the response. Can you provide more info on said gasket? Where and what?

 

[adwilk]

lolo, yeah ole Doc can be blunt lol, but he sure does know speaker’s. I for one enjoy reading his post even if I get Chastised a bit .

Haha true. Doc and I have worked on stuff in the past. I’ve never left the experience having not learned something. He was scheduled to visit Mayo today. My thoughts are with him and the looming pandemic.

 

[Gmoney]

Haha true. Doc and I have worked on stuff in the past. I’ve never left the experience having not learned something. He was scheduled to visit Mayo today. My thoughts are with him and the looming pandemic.

Yep Great guy for sure! When he post up I always read his post. I enjoyed the photos he posted up on his new home. Especially the one where he’s standing outside in that long black Trenchcoat and black hat. He reminded me of one of them dudes from that movie the Hateful Eight! Him standing in the snow with that walking cane in one hand looked cool man smooth!

 

[Swerd]

@adwilk

To my eyeball, that photo in your original post looks like a horn-loaded design, not a TL. But it may work well.
A couple of other thoughts:

• That Tang-Band W4 driver has been available in a number of models, with titanium or paper cones with added bamboo fibers, over the years. In the past, models such as the paper/bamboo W4-1320 SB, were known to have an audible peak in the treble region in the 6 kHz range. See this write up of a DIY design called The Bandit.
  
  
• I don't know what the performance of your W4 driver is, but you should be aware that Tang-Band has been known in the past to publish optimistic curves of it's drivers.
  
  To tame that peak in the Bandit design, a trap filter was required. It's the parallel LCR filter on the right side of the schematic. (See the 12 µF cap in parallel with a 0.05 mH coil & 16 ohm resistor. Read the Bandit web page to learn if there is really a 0.09 ohm resistor or if that's the DC resistance of the small coil.)
  
  
  
  
  The other, the LR circuit on the left side of the schematic, is meant as baffle step compensation (BFC) for that 7" wide cabinet. I don't know what BFC might be needed, if any, for that horn loaded design.
  
  Here's a photo (with the LCR filter on the left side and the BFC filter on the right)
  
  
• Drilling holes in plexiglass can result in visible chipped edges. The photo in your OP looks like it may have countersunk washers that cover any chipped edges in the screw holes. They're not the typical flat washers. See them also at Amazon.
  
  Get some small pieces of plexiglass to practice drilling holes at low speed.
  
  
• You may, or may not, have air leaks where the plexiglass attaches to the white material inside (styrofoam?). Practice using white or clear silicone sealer (such as GE Silicone II). Or look in a hardware store for white colored weather strip gasket material.

 

[adwilk]

Thanks, Swerd. Interesting read and results on "The Bandit" project. I appreciate the tips on sealing the enclosure. The weather strips make a lot of sense. I'm using a laser cutter (glowforge) to make a router jig to cut the pieces that will be laminated. It should make easy work of creating the appropriate pilot holes for attaching the plexi panels.

I have a woofer tester to get the parameters of each driver. I suppose I should start there. I'll double check the series, but I believe this is the titanium model, and I had uncovered some evidence that there was some "ringing" at certain frequencies and spl's. I don't ever foresee playing these loudly, but that always changes.

I think I'll implement minidsp to use as both the dac and active correction for this project. It won't require any additional amplification and I should be able to experiment with flattening the response. I was planning to use an outboard dac with a music server anyway.

 

[slipperybidness]

@adwilk

To my eyeball, that photo in your original post looks like a horn-loaded design, not a TL. But it may work well.
A couple of other thoughts:

• That Tang-Band W4 driver has been available in a number of models, with titanium or paper cones with added bamboo fibers, over the years. In the past, models such as the paper/bamboo W4-1320 SB, were known to have an audible peak in the treble region in the 6 kHz range. See this write up of a DIY design called The Bandit.
  View attachment 34712
• I don't know what the performance of your W4 driver is, but you should be aware that Tang-Band has been known in the past to publish optimistic curves of it's drivers.
  
  To tame that peak in the Bandit design, a trap filter was required. It's the parallel LCR filter on the right side of the schematic. (See the 12 µF cap in parallel with a 0.05 mH coil & 16 ohm resistor. Read the Bandit web page to learn if there is really a 0.09 ohm resistor or if that's the DC resistance of the small coil.)
  View attachment 34713
  
  The other, the LR circuit on the left side of the schematic, is meant as baffle step compensation (BFC) for that 7" wide cabinet. I don't know what BFC might be needed, if any, for that horn loaded design.
  
  Here's a photo (with the LCR filter on the left side and the BFC filter on the right)
  View attachment 34714
• Drilling holes in plexiglass can result in visible chipped edges. The photo in your OP looks like it may have countersunk washers that cover any chipped edges in the screw holes. They're not the typical flat washers. See them also at Amazon.
  
  Get some small pieces of plexiglass to practice drilling holes at low speed.
  
  
• You may, or may not, have air leaks where the plexiglass attaches to the white material inside (styrofoam?). Practice using white or clear silicone sealer (such as GE Silicone II). Or look in a hardware store for white colored weather strip gasket material.

Good call, plexiglass can be challenging to work with!

It is likely a good idea to tape up the area before you drill or cut plexiglass.

 

[TLS Guy]

Hello, all. I acquired a pair of W4-1337 SDF Tang Band 4 “full range drivers some time ago and I want to put them to use. I have a laser cutter and intend to make a design similar to the picture shown to put in my office. The issue it seems, is that the basis of design uses a flared line, by my best math its no where the 1/4 wavelength, and damping looks to be placed conveniently, but probably not adequately. Am I right in thinking that the flared design will raise f3? That doesn't seem to be desirable- especially in a small package. A tapered design should lower f3, right? Do you calculate the 1/4 wavelength on the actual f3 of the driver or the anticipated f3 based on the alignment? I'll be laminating the whole thing together- the middle panels could "house" the damping material hiding it from view? I mostly want to play with the new laser cutter and put the drivers to use. Thoughts, concerns, or should I go back into hiding?

You are correct. That is a horn loading the speaker. It is classic, with the expansion chamber behind the cone, a throat followed by exponential expansion. And as is correct it has no damping.

This is the most efficient way to load a driver with suitable parameters. High flux density and very low Q work best.

The math is totally different than for a TL. F3 is set largely by the dimensions of the opening. However placing the speaker in a corner lowers f3 and greatly extends bass output.

The classic horn designs are by Lowther, the world's oldest speaker manufacturer, since the landmark Voight corner horn of the early thirties. This is widely recognized as the first high fidelity speaker.

 

[adwilk]

Alright I cut up some stuff. I used hornresp to get an idea of how the driver would perform with the published t/s. My math is challenged to its very limit trying to determine how to achieve the appropriate volume/progression of the horn but the mouth and throat seem to be appropriate and the overall volume has to be uhhh... close. I'm hoping active dsp can help smooth out the response. I'm obviously interested in the aesthetic, but sound is important too. Anyway, here is the cut up version. I made the part slightly oversized to account for the bearing on the router. I could cut/laminate all the pieces like this but.. well.. I'd rather do some work. Total dimensions will be about 12" deep and 21" tall, and 7" wide. I ordered some plexi panels today, but I'm not concerned/optimistic about a speedy delivery. Other fish are frying. I am looking forward to working on this as I think my family is about to go on lockdown.

 

[adwilk]

So this is going to be more of a woodworking question- thoughts on the best way to use the above pattern? I'm not having initial success. The plan is to cut the pattern into 6 panels per speaker using 3/4" MDF. At this point, tracing it on each panel and cutting with a jig saw is an option I guess, but I'd like to use my router for better results- I'm just not exactly sure how to proceed.

 

[Swerd]

So this is going to be more of a woodworking question- thoughts on the best way to use the above pattern? I'm not having initial success. The plan is to cut the pattern into 6 panels per speaker using 3/4" MDF. At this point, tracing it on each panel and cutting with a jig saw is an option I guess, but I'd like to use my router for better results- I'm just not exactly sure how to proceed.

Cutting six ¾" MDF panels where the outside and inside cut patterns are identical will be tough.

Cutting those 6 pieces of MDF to the exact same outside dimension will be tough enough, but it probably can be done with a saw of some kind, but I doubt if jig saws work well with ¾" MDF. I personally would go to a Home Depot, Lowes, or a decent lumber yard where there's a large panel saw plus someone who knows how to use it. Get them to cut six panels that are 12" × 21". Take the time to figure out how many 12×21 panels you can get from one 4×8 sheet of MDF. Get more than six. You'll probably have to sand or use a router with a bearing bit to make them exactly the same outer dimensions.

The irregular inside cut pattern looks like it will require making a router template. In addition to a router, you'll need ¾" template collar bushing, plus a ¼" (or ½"?) spiral upcut bit. The collar bushing should work better for this than a bearing bit. The patterns you now have should be useful for the first step to make a router template.

See the PDF I attached for directions on how to flushmount an irregular shaped tweeter faceplate. Or see this: https://www.reddit.com/r/diyaudio/comments/4ysmms/tutorial_ how_to_flushmount_irregular_shaped/

I say all this like I've done it before, but I haven't. I watched while someone else did it, so it looked deceptively easy . I don't know how much router experience you have, but it's going to take practice. Be prepared for that idea with extra MDF or Masonite boards to use.

 

[TLS Guy]

I did not know you were going to use MDF.

Horns are different. Donald Chave of Lowther taught me as a kid, that the panels of horns must actually not be too rigid and have internal damping. Lowther horns are still built from good quality 3/8" plywood!