Enclosure resonances, not a big deal?

[thadman]

Last I heard, B&H Photo had the best price on Earthworks mics, something like $440 for the M23 and $550 for the M30. You have to call them to get the good price.

Thanks for the tip :T I'll give them a call.

What do you use for enclosure stuffing? Corning Owens 703 appears to be an obvious choice, but I read somewhere that fiberglass is optimally suited for high velocity absorption whereas acoustic foam is optimally suited for high pressure absorption. Should we use a combination of fiberglass and foam or is one sufficient?

[brianpowers27]

When possible, use rigid fiberglass. Hands down, it is a better solution.

[thadman]

When possible, use rigid fiberglass. Hands down, it is a better solution.

For the high frequency transmission line, wouldn't progressive stuffing be preferable over traditional stuffing? I've seen long fiber wool used before, but where can you source it and would it be better than rigid fiberglass for the first several inches of attenuation?

[ThomasW]

http://www.rhlindsaywool.com/products/index.htm

[thadman]

http://www.rhlindsaywool.com/products/index.htm

excellent find

Would you recommend the Scoured Carded New Zealand White Romney, Product #61S?

description: Originally created in response to audio equipment manufacturers, this product has been used for felting and stuffing as well. Specifications: scoured carded fleece, white, semi-lustrous. Staple length 3-4 inches (75-100 mm). Fiber count 46-48s (31-33 micron).

Or is another fiber better suited?

[ThomasW]

excellent find

I first bought from them 20yrs ago.

Would you recommend the Scoured Carded New Zealand White Romney, Product #61S?

That's fine

[thadman]

Are there rules of thumb for stuffing density? I'm not sure how much to purchase.

[thadman]

Btw, congrats thad. You've changed the format of the forum .
Got a question. I see you are in Indiana. Will these be ready for Dayton DIY?

Does Dayton DIY (ie PE) provide equipment for the active crowd?

[brianpowers27]

Thadman,

I sourced mine at Momper's insulation in fort wayne. They were a little challenging to work with but I ended up with a greaat deal. I would have spent 4x more to have it shipped in.

I ended up with Rockwool 60. THis is 6.0pcf insulation and it useful for a variety of things. I purchased one case (4'x2'x1.5"x12 sheets) for $30. I have built some bass traps and lined several speakers.

Many insulation suppliers don't work with this stuff often. It is important to be patient since they can usually find what you need if you use the correct vocabulary.

I couldn't tell you about the progressive stuffing requirements of t-lines. The only t-line I have built is an OB/mini-hframe design. I didn't use any stuffing there. I will tell you experientially that this works wonders.

I believe you can find all of the NRC coefficients you need here. These will help you to see the relative efficacy of each material. The only problem is the charts stop at 100hz. I can't cite a reference but I will tell that for sub 100hz 6.0pcf fibers/mineral wool is the kind when the thickness is >=4".

http://www.bobgolds.com/AbsorptionCoefficients.htm

[ThomasW]

Are there rules of thumb for stuffing density? I'm not sure how much to purchase.

I'm sure there are, I stopped building TL's in the early 1980's so I'm not up on what the current thinking is......

How much you need depends on the size of the line.

People have their favorite damping 'recipes' Some go heavy to light, others light to heavy, some the same amount through the entire line, some a dense wad behind the driver and a consistent amount in the rest of the line. The variations are almost endless.

With a mid it's not the same as for a woofer. I think 3 lbs would get you started

[thadman]

Thadman,

I sourced mine at Momper's insulation in fort wayne. They were a little challenging to work with but I ended up with a greaat deal. I would have spent 4x more to have it shipped in.

I ended up with Rockwool 60. THis is 6.0pcf insulation and it useful for a variety of things. I purchased one case (4'x2'x1.5"x12 sheets) for $30. I have built some bass traps and lined several speakers.

Many insulation suppliers don't work with this stuff often. It is important to be patient since they can usually find what you need if you use the correct vocabulary.

I couldn't tell you about the progressive stuffing requirements of t-lines. The only t-line I have built is an OB/mini-hframe design. I didn't use any stuffing there. I will tell you experientially that this works wonders.

I believe you can find all of the NRC coefficients you need here. These will help you to see the relative efficacy of each material. The only problem is the charts stop at 100hz. I can't cite a reference but I will tell that for sub 100hz 6.0pcf fibers/mineral wool is the kind when the thickness is >=4".

http://www.bobgolds.com/AbsorptionCoefficients.htm

I currently have 5 4' x 2' panels of Corning Owens 703 that I use for wall treatments ... but it was bloody expensive! I tried sourcing it in Fort Wayne, but to no avail.

You mentioned using the "correct vocabulary". If I was interested in purchasing some Rockwool 60, what should I attempt to convey to the salesman?

[brianpowers27]

You mentioned using the "correct vocabulary". If I was interested in purchasing some Rockwool 60, what should I attempt to convey to the salesman?

6.o pcf.. mineral wool... etc.. It is very hard since they sell so little...

[AJINFLA]

When / Where is Dayton DIY? I will be in Munich from July 13th to August 12th.

I'm perfectly satisfied with my current monitors on most content, but the single midwoofers strain at higher amplitudes (>100dB), especially if the signal contains a lot of LF content. Supermodified by Amon Tobin is a supreme example ... my midwoofers aren't a big fan of this record at higher amplitudes.

For the most recent build, I picked up a pair of TD15M Apollo drivers as well as a pair of Beyma TPL-150 AMTs. I'd like to do a 2-way, but I have some reservations. A PHL 1120 (or AE TD6.5M if its EVER released) could potentially be added as a transition driver if synergy is not achieved between the TD15M and TPL150.

The TD15M with its unique cone profile and corresponding modal behavior makes polar response calculations quite difficult. John has mentioned that it is flat to 4khz and suitable off-axis to 2khz. We'll soon see how true this claim is. Polar response measurements will have to be taken to see how suitable it is as a midrange driver and how high it can really reproduce off-axis.

StigErik (over on DIYAudio) uses a pair of TPL-150s in his current design and has them crossed to a Seas W22 around 1400hz without issues. I've got a few ideas for potential tweaks that may benefit performance.

I plan on removing the rear enclosure of the TPL-150 and instead of operating them dipole, I'll place them in a 16" stuffed transmission line. The attenuation should be significant above 200hz, as 16" is roughly 1/4 wavelength of 200hz.

I'm also interested in placing the TPL-150 in a DIY waveguide. Due to the drivers dimensions, construction of the waveguide should be relatively straightforward and painless. With 90* horizontal and 30* or so vertical, a 1khz crossover shouldn't be a problem. The commercial waveguides (TPL-150H) could be ordered if the DIY version doesn't measure up.

Fuzzmeasure Pro software has been purchased for measurements along with an Apogee Duet. The microphone however has not. I'd like to purchase an Earthworks product, but haven't found a good deal yet.

Any input?

I would only use the TPL-150 as a dipole. I would not use any 15" driver above 1k. I would use a 5-6" midrange between both.
Last years Dayton had a pre/power combo, so yes, active is possible. My entries had powered subs, but I had to run max gain, as my speakers were 97db/2.83v and MikeV had the 2 amps mono bridged @ 400w per channel.
I might e-mail RichT to see what equipment is to be used this year and offer some suggestions.

cheers,

AJ

[thadman]

I would only use the TPL-150 as a dipole.

For what reason? Other than polar response, the transmission line should behave similarly to the dipole alignment.

I would not use any 15" driver above 1k. I would use a 5-6" midrange between both.

That is my thinking too, especially since CTC spacing is so large at such high crossover points, making lobes unavoidable. We'll see though. I still need to do measurements and have left available funds to purchase a transition midrange if necessary.

A 5-6" transition midrange however is not without its own flaws. It will have much lower sensitivity and heat sinking ability at the lower end of its passband compared (~93-95 for the midrange, ~97-100 for midbass/tweeter) to the TD15M and TPL150 whilst substantially increasing the cost of the project by requiring a third amplifier and more processing power. There also aren't any high efficiency midranges on the market that appear to deserve attention. Sure there is the PHL 1120, but thats hard to source and has a terrible frequency response. There also that Accuton beast of a midrange, but it costs ~$600/driver. It's something I'd like to avoid if possible.

[AJINFLA]

For what reason?

Audibility

Other than polar response

That's the audible one

the transmission line should behave similarly to the dipole alignment.

In inaudible ways? Maybe. So? Btw, a "transmission line" for frequencies above 1k is unnecessary and absurd. It may help sell B&W's, but for a dose of reality, go listen to Earls Summas. You're not far. Then tell me how the treble from his >900k tweeters sound with that chintzy little back cup vs a "transmission line" B&W (which I assume you have heard).

That is my thinking too, especially since CTC spacing is so large at such high crossover points, making lobes unavoidable.

Using non-coincident drivers makes lobes (per speaker). Inevitably.

We'll see though. I still need to do measurements and have left available funds to purchase a transition midrange if necessary.
A 5-6" transition midrange however is not without its own flaws. It will have much lower sensitivity and heat sinking ability at the lower end of its passband compared (~93-95 for the midrange, ~97-100 for midbass/tweeter) to the TD15M and TPL150

All designs have flaws. Again, a focus on the inane and ignorance of the basics. A 93-95 mid on the correctly shaped dipole baffle will achieve both the required sensitivity to match the woofer and tweeter and provide better thermal dissipation. Which on a pro mid, will not be an issue until way after your ears beg for mercy.

whilst substantially increasing the cost of the project by requiring a third amplifier and more processing power. There also aren't any high efficiency midranges on the market that appear to deserve attention. Sure there is the PHL 1120, but thats hard to source and has a terrible frequency response. There also that Accuton beast of a midrange, but it costs ~$600/driver. It's something I'd like to avoid if possible.

Good mids can be had for less than $100 and amplification for such a narrow passband not much more. Plus the savings from not building such ridiculous enclosures pictured above should more than offset this.

cheers,

AJ

[thadman]

Audibility

That's the audible one

I'm aware of the differences in polar response between a monopole and a dipole, as I had previously stated. I'm not interested in a dipole at the moment because of the same issue, polar response. A dipole only emulates the figure-eight radiation pattern over the bandwidth where wavelengths > baffle width. Above this bandwidth, it does not behave accordingly. I'm aware that the gradual narrowing of power response due to beaming may come close to this radiation pattern ... but there are simply too many factors at play here. I'd rather try a monopole. Thank You.

In inaudible ways? Maybe. So? Btw, a "transmission line" for frequencies above 1k is unnecessary and absurd. It may help sell B&W's, but for a dose of reality, go listen to Earls Summas. You're not far. Then tell me how the treble from his >900k tweeters sound with that chintzy little back cup vs a "transmission line" B&W (which I assume you have heard).

I'm not interested in augmenting the drivers output with the rear wave. If that is your argument, than yes, it doesn't make sense to use a transmission line for that application. I'm more interested in avoiding air-spring non-linearity as Dennis H previously mentioned in this thread.

Let's assume we wish to reproduce 120dB transients (Overture of 1812 anyone?) @ 2m above 1khz. We will assume the baffle width is of appreciable width to provide 6dB boundary gain, this drops our displacement requirements to 120dB @ 1m. 8 cubic centimeters of displacement are required to meet this output level.

Going by Dennis H's provided equation:

Distortion % = 140*(one-way driver displacement)*(1/enclosure volume)

If we wish to remove any non-linearity present in the system, we'll choose an arbitrarily low distortion level. Let's say .1%.

Required Enclosure Volume (.1% air non-linearity) = 10*140*8(displacement required to achieve our output goal)

We now have an enclosure volume of 11,200 cubic centimeters (or 112 liters) that contributes .1% distortion at 120db @ 2m. This is obviously ridiculously large for a tweeter enclosure. What is then reasonable for a user? I've determined a line whose height and width matched the dimensions of the AMT and whose depth matched the depth of the woofer enclosure (16") to be reasonable for my uses.

I'm too lazy to look up the dimensions of the AMT, but to assume 1" width and 6" height isn't unreasonable. A "reasonable" enclosure volume was found to be ~16 liters.

Now, let's determine the maximum SPL attainable by the system within an arbitrary distortion threshold (for this argument, let's assume 1%). We can manipulate the equation to provide us with the maximum displacement below our distortion threshold.

Maximum displacement = Enclosure Volume / 140

1600/140=11.43 cubic centimeters, >8 cubic centimeters. Therefore our output goals are able to be reached without the influence of air non-linearity on the system.

Now you might argue that this is excessive and that a drivers own non-linearity would greatly exceed it. This may be true, however if we can remove a source of distortion ... why not? From my understanding, it shouldn't have a negative influence on the system.

Also, to suggest a straight 16" transmission line is "ridiculous" is pretty ridiculous on its own accord. Constructing a straight, rectangular transmission line is trivial. In fact, I believe the highly regarded "Statements" utilize a straight, rectangular line and I haven't seen any comments on here accusing it of being excessive.

BTW Why would you assume I have heard the B&W transmission line?

Using non-coincident drivers makes lobes (per speaker). Inevitably.

What are you exactly trying to convey here? As long as the drivers centers are within 1/4 wavelength of each other, power response aberrations (may be a narrowing of power response or lobes) should be audibly insignificant.

All designs have flaws. Again, a focus on the inane and ignorance of the basics. A 93-95 mid on the correctly shaped dipole baffle will achieve both the required sensitivity to match the woofer and tweeter and provide better thermal dissipation. Which on a pro mid, will not be an issue until way after your ears beg for mercy.

I'm aware that all designs have flaws. I stated that I was going to "attempt" to create a 2-way using the TD15M and TPL150. I didn't claim it was possible or even likely, I stated that I would complete measurements and make judgements based on that information. If the attempt failed or the system was not found to be satisfactory, I would add a transition midrange.

I'm going active, which was stated previously with regard to Dayton DIY. I don't have to match sensitivities. Based on what information have you deduced that 93-95dB is sufficient for my application? I have not spoken at all about my application, so I'm not sure what you're basing that assumption on.

I'm curious as to how a 5-6" pro midrange (93-95dB, lets assume 94dB) would have better thermal dissipation than the TD15M with the Apollo upgrade (lets assume 97dB) which is known for its massive heatsinking ability. Let's isolate mechanical compression and consider only thermal compression, 500hz should place both drivers well within their mechanical limits. 103dB @ 2m (109dB @ 1m) is fairly loud, but not unrealistic.

The Pro Midrange would require 32 watts

The TD15M would require 16 watts

I fail to understand how the Pro Midrange would be an improvement (with regards to heatsinking ability) over the TD15M.

Good mids can be had for less than $100 and amplification for such a narrow passband not much more. Plus the savings from not building such ridiculous enclosures pictured above should more than offset this.

cheers,

AJ

What midranges? The only well-documented (measurements, personal experience) Pro Midrange which also possess high sound quality that I'm aware of is the PHL 1120, and it costs ~$180/each. I'd also have to construct a separate midrange enclosure, just another thing to build.

I fail to see how any of the enclosures I've suggested are ridiculous. The statements utilize conceptually the same thing.

[Dennis H]

We now have an enclosure volume of 11,200 cubic centimeters (or 112 liters)

I didn't check the rest of your math but that's 11 liters. And that whole box spring thing really doesn't apply to a ribbon that is leaky around the edges. And expecting a ribbon to play down to 1K at 120dB at 2m is unrealistic unless you think a ribbon has an Xmax of 10mm. Bottom line, a few inches of fiberglass will absorb any reasonable backwave from a ribbon and you don't really need to analyze it more than that.

[thadman]

I didn't check the rest of your math but that's 11 liters. And that whole box spring thing really doesn't apply to a ribbon that is leaky around the edges. And expecting a ribbon to play down to 1K at 120dB at 2m is unrealistic unless you think a ribbon has an Xmax of 10mm. Bottom line, a few inches of fiberglass will absorb any reasonable backwave from a ribbon and you don't really need to analyze it more than that.

You're correct. It is 11.2 liters. However that is still traditionally considered excessively large for a tweeter enclosure.

The TPL-150 is not a ribbon, it is an air motion transformer. It has significantly higher surface area than a ribbon of the same dimensions.

A 6" x 1" ribbon (dimensions we assumed the TPL150 to be) has ~39cm^2 of surface area. The ribbon would need 2mm of xmax to do 120dB at 2m. An AMT of the same dimensions would have significantly higher surface area wrt the ribbon, further lowering the xmax requirements.

[AJINFLA]

I'm more interested in avoiding air-spring non-linearity as Dennis H previously mentioned in this thread.

Right, because that is so audible.

What are you exactly trying to convey here? As long as the drivers centers are within 1/4 wavelength of each other, power response aberrations (may be a narrowing of power response or lobes) should be audibly insignificant.

Really? You are going to cross a 15" to the AMT, which is what, about 6", 1/4 wavelength apart? So around 340hz? Interesting. And what happens as we go up in frequency above XO? The drivers still remain correlated?

I'm going active, which was stated previously with regard to Dayton DIY. I don't have to match sensitivities.

A 5-6" transition midrange however is not without its own flaws. It will have much lower sensitivity

:roll:

I'd rather try a monopole. Thank You.

Best of luck.

cheers,

AJ

[cjd]

I fail to see how any of the enclosures I've suggested are ridiculous.

I suspect he was referring to the ones pictured a bit ago which went crazy on bracing and such.

A TL will have its own issues WRT introducing nonlinearities - it's still about picking your poison. And since it's frequency dependent it could be less linear in some ways than air. Hopefully your "sealed" is still lossy.

By the way, how far down on the crossover before you consider it no longer worth considering a driver as contributing (and thus its 1/4 wavelength worth noting)? I'd assume -50dB or more based on the other issues you worry about, which means you'll need TONS of overlap to keep your center to center spacing working.

C

[thadman]

Right, because that is so audible.

The audibility of distortion has been debated to death, with Zaph arguing for the audibility and significance of non-linear distortion and Dr. Gedess arguing the insignificance of it related to other factors. I'm not sure how much of a factor it plays, so if I have the opportunity to remove its influence on the system without much inconvenience, why not?

Really? You are going to cross a 15" to the AMT, which is what, about 6", 1/4 wavelength apart? So around 340hz? Interesting. And what happens as we go up in frequency above XO? The drivers still remain correlated?

No. Power response narrows, this is elementary. Each filter / spatial relationship has its own lobe pattern. I'm aware of this and have decided to proceed. Dr. Geddes crosses a 15" to his OS waveguide with great success in his Summa loudspeaker. The TD15M and TPL150 would have an even smaller CTC distance. If results are found to be unsatisfactory, a transition midrange could be added between them without much hassle. This has been stated several times.

[thadman]

I suspect he was referring to the ones pictured a bit ago which went crazy on bracing and such.

A TL will have its own issues WRT introducing nonlinearities - it's still about picking your poison. And since it's frequency dependent it could be less linear in some ways than air. Hopefully your "sealed" is still lossy.

By the way, how far down on the crossover before you consider it no longer worth considering a driver as contributing (and thus its 1/4 wavelength worth noting)? I'd assume -50dB or more based on the other issues you worry about, which means you'll need TONS of overlap to keep your center to center spacing working.

C

Why is my previous enclosure being referenced? We are discussing the excessiveness of a completely different speaker alignment, namely an AMT transmission line. A 15" <--> AMT is quite dissimilar to a 5.5" <--> 1" tweeter.

[cjd]

It was a guess, and may not be correct.

Why are you so combative?

I have the opportunity to remove its influence on the system

Only the opportunity to move it elsewhere...

[thadman]

It was a guess, and may not be correct.

Why are you so combative?

If I came off as combative in my post, I am sorry as that was not my intention.

[brianpowers27]

I'm aware of the differences in polar response between a monopole and a dipole, as I had previously stated. I'm not interested in a dipole at the moment because of the same issue, polar response. A dipole only emulates the figure-eight radiation pattern over the bandwidth where wavelengths > baffle width. Above this bandwidth, it does not behave accordingly. I'm aware that the gradual narrowing of power response due to beaming may come close to this radiation pattern ... but there are simply too many factors at play here. I'd rather try a monopole. Thank You.

If you'd like you can check out my OB. I narrow the baffle for each component, in order to keep the figure 8 radiation pattern. I have it operational. Since you are in FW, you could always come see it.

Isn't the ratio of the backwave generally considered equal to the sealed box Q? I have generally understood that ratios between .5-.7 yield a good level of damping.

[cjd]

If I came off as combative in my post, I am sorry as that was not my intention.

Something in the general tenor of your posts does come across that way - once it starts, everyone starts, so it just gets worse. Guess we can all work on our internet manners.

[thadman]

TPL-150s have arrived, pics will follow. These are some serious tweeters!