Help with Adire 281 TL design

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello all!

I've been lurking here for a while and recently registered. Of course, now I am hoping to lean on the experience of the folks here to help mw with a problem I've encountered.

I recently built the Adire Audio 281 speakers in a Transmission line version. The Adire 281's have been noted for their strong bass performance and have been considered a strong speaker for home theater as noted in this audioxpress review . Unfortunately, the pair I built have very poor bass response below 100hz, dropping off to -12db at 50hz. I've taken some measurements using RoomEq and my Radioshack SPL meter and they are shown below.

I'm not smart enough to diagnose the problem and was hoping to provide the forum with the relevant info to see if there's anything I can check / do. Any help or ideas would be greatly appreciated.

Thanks,

Russell

I've included the cabinet and crossover plans below.

281 cabinet
TL line stuffed with 24oz polyfill in 4.5" section, 18oz in 3.5" section, 6 oz in 3" section.

An externally hosted image should be here but it was not working when we last tested it.


281 XO

An externally hosted image should be here but it was not working when we last tested it.


281 TL (right channel measured in room ~3' away)
Solid Black line is RS SPL calibration line
Dotted black line is sound card calibration line
Blue line is corrected speaker line


An externally hosted image should be here but it was not working when we last tested it.


Previous speaker (wharfedale modus 1.6 right channel measured in room ~3' away)
Solid Black line is RS SPL calibration line
Dotted black is sound card calibration line
Purple is corrected speaker line

An externally hosted image should be here but it was not working when we last tested it.
 
Dan Wiggins... OK, check that. But what do you mean by "classical" methodology? Recently, that has become a swear word around these parts. Check the Clarity on Seas Thor Kit* thread, and you'll understand. Martin's spreadsheets come closest to predicting final as-built response.

One thing to inspect is the amount of stuffing inserted. Too much can overdamp the back wave. Try experimenting with removing some (or a lot).

Edit: *the action starts at post #10
 
frugal-phile™
Joined 2001
Paid Member
Shaun said:
But what do you mean by "classical" methodology? Recently, that has become a swear word around these parts.

And i hope that i played a role in making that so :) Classic is what we were doing before Martin (and George) came along and showed how much more could be gotten out of a QW box.

MJK should always be consulted when doing a TL.

I've been chatting with Russell by eMail, and have made a few suggestions already in the hopes of saving him from having to start over on the boxes.

One of the problems is that they are very heavily stuffed (i haven't checked Russell's density calcs, been busy) and with them all closed up he has no way of pulling it out and playing.

dave
 

GM

Member
Joined 2003
trpltongue said:

I recently built the Adire Audio 281 speakers in a Transmission line version. The Adire 281's have been noted for their strong bass performance and have been considered a strong speaker for home theater as noted in this audioxpress review .[/IMG]

Greets!

Well, after browsing the assembly manual and review as well as doing an approximate sim using MJK's software, you got exactly what DW said you were getting, a ~anechoic response down to 35 Hz, i.e. a ~IB response in a relatively small cab. If you wanted the Kit281's potential strong bass response, then you should have built the vented version that was reviewed.

Note that this isn't a TL per se, but a Daline like the Seas Thor.

GM
 
Thanks to all for the great replies!

I have been working with Dave and Dan via email and as mentioned, I have no real ability to remove all the stuffing. Right now I have ~ 1.8lb/cuft of stuffing as per the 281 design manual. From speaking with Dave and Dan, that seems a bit overstuffed.


GM,

I am admittedly very new to TL design, but based on the descriptions of the vented and TL design, I thought the TL design would reach somewhat lower, flatter than the vented design:

Vented:
anechoic response to 45Hz with a box tuning of 28Hz
typical in-room extension into the 30hz range

TL:
anechoic response to 35hz with a line tuning of 37Hz
typical in-room extension to 28hz

Either way, I didn't expect such a sharp falloff at 100hz.
 
*edit to above

I just spoke with Dan and he informed me of where I was misunderstanding :) I thought that the TL would have a -3db response at the F3 of the driver, but Dan has informed me that the typical TL response is -10db at the F3 of the driver, which is spot on what I'm getting.

I will still try removing the stuffing from the first line section as much as possible and listening, but if all else fails I will try blocking the line terminus and porting the rear to see how it sounds.

Oh the joys of being a rookie :)
 
Seems a rather strange design to me. Leaving aside the discontinuities, it seems over-long & lacking sufficient volume to get much useable LF gain. And that kind of stuffing density is excessive in the extreme IMO for almost any TL.

This is the cabinet with no damping applied, in the TL sections sheet & the drivers paralleled for the sake of the sim.
 

Attachments

  • tl undamped.gif
    tl undamped.gif
    7 KB · Views: 198
Scottmoose,

Thanks for taking the time to build those! Perhaps you can check my density calculations to make sure I'm not messing it up. 48 total ounces of stuffing into ~1.7 cuft of volume.

Interesting that your second graph with 1lb/ft^3 shows a -10db point of ~45-50hz, similar to what I'm seeing, though my measurements do have a bit steeper falloff.
 
Well, 48oz / 16 = 3lbs. 3lbs / 1.7 cubic feet = an average density of 1.76lbs ft^3 of material.

There will be a degree of variation in practice from the simulation, especially given that different materials require different packing densities for similar damping performance, but they get you in the right zone. In this case, your 75% great quantity of damping is going to kill some more of the LF gain.

People often get the wrong idea about TLs, primarily because the blasted phrase has come to be used for so many different cabinets, with completely dissimilar objectives. In the case of this cabinet, it's a near-aperiodic line, closer to a proper transmission line than a lot. These cabinets tend to offer a damped bass, with a shallower roll-off (starting at a higher frequency) than a vented enclosure; 18 or 12db / octave compared to 24db / octave is common. Their real advantage lies here, and in presenting a much flatter impedence to the amplifier (the mythical 'ideal' load), with superior cone control to an equivalent sealed box.

Ripping out all the damping from the last part of the line might help a bit. Looking at what MathCad's telling us about their performance without any damping at all, I wouldn't expect a huge difference -they simply don't have enough volume & the line is too long for there to be much LF gain. It appears that Dan hit his design objectives well enough, albeit arguably going a bit far in some areas, but those objectives do not match your taste, which isn't entirely surprising given the confusion the name Transmission Line invariable engenders.
 
Scott,

You lost me on the calculation there.

3lbs / 1.7 cuft = 1.8lbs/cuft

You calculated 1.7cuft / 3lbs = 0.566cuft/lb (isn't that the inverse of density?)

I'm so confused :confused:

Regarding the benefits of TL vs sealed boxes, wouldn't that smoother rolloff equate to more bass than the equivalent sealed box? Perhaps not as "boomy" as a ported box though, depending on tuning. The TL's should certainly have impactful bass though right?

Sorry for all the quick-fire questions :)
 
My bad. Corrected above. 1.76lbs ft^3, average. Which is very high.

Depends. Speaking in general terms, the equivalent sealed box would usually be larger than a well executed aperiodic TL, and its impedance curve & exursion would be higher too. Well damped rather than impactful would be my phrase for this type of box, generally. Aperiodic lines go low, smoothly, with that slow, shallow roll-off. But they won't have the punch of something with more gain higher up, especially in the midbass regions. Suits some, not others.

If a vented box booms, then it's badly designed IMO. A vented cabinet should not boom if done properly. Trouble is, not that many are. :rolleyes:
 
That make a lot of sense. Thanks for the lessons :)

Yeah, I was hoping to get a flatter response down to the mid 50's then rolling off after that. I guess I just didn't understand the explanation of line tuning and anechoic response well enough. I'd love to have gotten a response similar to the "Fat thor", darn near linear down to 50hz, then rolling off after that.

I don't know if I want to take the time to rebuild this whole box. So if removing the stuffing doesn't help much, which I doubt it will, I'll probably build it into the ported version by removing the front baffle, drilling a bunch of holes in the TL and using it for bracing rather than a TL, and rebuilding the front baffle with the 3" flared vent kit that I've got at the house.

It's a bummer because I was hoping to test out the sound of the TL's.

Perhaps in another project I can utilize the TL design.
 
trpltongue said:
Thanks to all for the great replies!

GM,

I am admittedly very new to TL design, but based on the descriptions of the vented and TL design, I thought the TL design would reach somewhat lower, flatter than the vented design:

TL:
anechoic response.......

Greets!

You're welcome!

Understood. Well it does, but these are relative terms that have to be further defined, which DW did with the term 'anechoic'. This tells us that the TL will have basically the same response as the driver in an infinite baffle (IB), so the driver's Qts will tell us how rolled off it is since there's no box gain. For an anechoic TL to have ~the same half space response as its vented counterpart would require a Qts = 1, so assuming the drivers are wired in parallel you can flatten them by adding a 5 ohm non-inductive power resistor in series with your speakers, though at a ~4.5 dB drop in efficiency.

If you want to remove some of the stuffing at the driver and port end, one of these will come in handy: http://www.sears.com/shc/s/p_10153_...hanics+Tools&sName=Automotive+Specialty+Tools

GM
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.