my prototype towers

[MacManNM]

Well after 2 months of intense design here it is....... well almost one of them. I still need to mount the 10" sub (which should get here monday) tomorrow I am going to finish the other tower and do some prelim listening tests. My first measurements indicate that my code was pretty much on the money. This speaker is pretty flat from 36Hz-20kHz. Listening tests confirm that they should sound real good as a pair.

Here is the prototype. The bottom is the space for the front firing 10" sub.

Here is a pic of the prototype's crossover.

Here is the internal baffel for the sub and the woofers.

All in all I think it is coming out quite well. I listened (to one) for a bit tonight and have noticed that these are very accurate and dynamic. I can't wait for tomorrow so I can hear them both!

 

[Sheep]

Thats quite the "crossover". How many watts do these need?

SheepStar

 

[mike c]

wow. whats the specs on these? driver size? ohm requirement?
do you know the sensitivity?

 

[HookedOnSound]

Very nice, congrats!

 

[MacManNM]

The drivers are as follows, tweeter 1" vifa, midrange 5.25, woofer 7".
There will be a 10" sub added to the bottom of the cabinet as soon as it arives.

Impedane is around 5 ohms. and sensitivity is about 86 or so. They require a lot of power but they are very detailed and dynamic.

 

[jaxvon]

Very cool, John. Too bad I can't sit down and have a listen to them.

 

[STRONGBADF1]

Looking good MacMan,

Timaaaay will love them once the subs are in!!!

Is there more bracing than the picture show or is the box pretty solid as is?

What finnish are you going to use?

Please keep us posted,
SBF1

 

[WmAx]

It's not clear what kind of acoustic dampening, or what degree of that you are using internally. Of course that is tricky with a ported system or you have a high risk of decreasing the efficiency of the port output. I recommend a 2" rigid, 6lb/ft^3 density fiberglass boards attached to the side walls, and 2 layers of 2" to equal at least 4" on the rear and front walls(greater distance=longer wavelengths) as opposed to the normal low density fiberglass. This rigid fiberglass will still result in some port output loss. But the probable increase in sound quality due to eliminating these resonances is fair trade IMO. A high grade acoustic foam will also work, but since it is usually only available convoluted, you will need about 3" to compensate for the lower overall mass. Cushion or packing foams are not a suitable substitution, as they have much lower co-efficient at the frequencies of interest. The materials I recommend should result in basically zero internal volume resonant effects related to the wall spacing dimensions. For the actual walls, you should attach a mass loaded damping material similar to Dynamat(some roofing materials are available that approximate it), applied in layers to equal about 1/3 the thickness of the wall thickness. This will severely reduce the magnitude of panel resonances.

-Chris

 

[MacManNM]

Looking good MacMan,

Timaaaay will love them once the subs are in!!!

Is there more bracing than the picture show or is the box pretty solid as is?

What finnish are you going to use?

Please keep us posted,
SBF1

There is a couple of more pieces of bracing in there, I was too excited to get them together to take more pics, although I should have. The outside is going to be covered in 3/4" Knotty pine to match the new room.

 

[MacManNM]

It's not clear what kind of acoustic dampening, or what degree of that you are using internally. Of course that is tricky with a ported system or you have a high risk of decreasing the efficiency of the port output. I recommend a 2" rigid, 6lb/ft^3 density fiberglass boards attached to the side walls, and 2 layers of 2" to equal at least 4" on the rear and front walls(greater distance=longer wavelengths) as opposed to the normal low density fiberglass. This rigid fiberglass will still result in some port output loss. But the probable increase in sound quality due to eliminating these resonances is fair trade IMO. A high grade acoustic foam will also work, but since it is usually only available convoluted, you will need about 3" to compensate for the lower overall mass. Cushion or packing foams are not a suitable substitution, as they have much lower co-efficient at the frequencies of interest. The materials I recommend should result in basically zero internal volume resonant effects related to the wall spacing dimensions. For the actual walls, you should attach a mass loaded damping material similar to Dynamat(some roofing materials are available that approximate it), applied in layers to equal about 1/3 the thickness of the wall thickness. This will severely reduce the magnitude of panel resonances.

-Chris

I use fiberglass in most of my enclosures, I really dont care too much about slowing down the port too much, the box is tuned well below where I'm going to be using it. I calculated the enclosure sixe based on this, so I'm not worried about taking up space in the enclosure. I really wanted it dead. The only reason I ported it is because I wanted to flatten out the low end where I am cutting it. Also I will be adding another 3/4" of material to the outside, so that will further stiffen the unit. As it is now when you knock on the box it sounds like you are knocking on concrete.

 

[WmAx]

I use fiberglass in most of my enclosures, I really dont care too much about slowing down the port too much, the box is tuned well below where I'm going to be using it.

The fiberglass high density material I suggested has a much higher coefficient as compared to normal fiberglass insulation. Which specific fiberglass, and at what density are you using?

I calculated the enclosure sixe based on this, so I'm not worried about taking up space in the enclosure.

I was not referring to the internal volume. The problem is that any considerable amount of acoustic damping material will cause the energy transfer between the driver and port to reduce in efficiency.

Also I will be adding another 3/4" of material to the outside, so that will further stiffen the unit. As it is now when you knock on the box it sounds like you are knocking on concrete.

Increasing the thickness raises the relative resonant frequencies, but does not dampen them. You can reduce audibility by raising resonances, dependant upon the specific frequency and Q of those resonances as related to human audibility, but this requires complex analysis and correlation with perceptual research to be certain the modification is of relevant benefit. The mass loaded damping material converts the vibrational resonances into thermal energy, thus reducing resonances, not just shifting the frequency of those resonances.

-Chris

 

[MacManNM]

The fiberglass high density material I suggested has a much higher coefficient as compared to normal fiberglass insulation. Which specific fiberglass, and at what density are you using?

I was not referring to the internal volume. The problem is that any considerable amount of acoustic damping material will cause the energy transfer between the driver and port to reduce in efficiency.

Again I don't really care much about efficiency that much. The resonant frequency should be high enough that the amount of energy in music is not that great. Hence, it should not be a factor.

Increasing the thickness raises the relative resonant frequencies, but does not dampen them. You can reduce audibility by raising resonances, dependant upon the specific frequency and Q of those resonances as related to human audibility, but this requires complex analysis and correlation with perceptual research to be certain the modification is of relevant benefit. The mass loaded damping material converts the vibrational resonances into thermal energy, thus reducing resonances, not just shifting the frequency of those resonances.

-Chris

All I need to analize them is an impulse source, a digitizer, and a calibrated microphone. Which I have, and will perform measurements at a later time. These are prototypes, they are going to end up in the scrap heap in 6 months anyhow. Just a fun project for me to waste time and money, and have an excuse to drink beer and use power tools at the same time!

 

[WmAx]

All I need to analize them is an impulse source, a digitizer, and a calibrated microphone. Which I have, and will perform measurements at a later time. These are prototypes, they are going to end up in the scrap heap in 6 months anyhow. Just a fun project for me to waste time and money, and have an excuse to drink beer and use power tools at the same time!

Unfortunately, it's no easy matter to measure/analyze audibility magnitude of enclosure wall resonances.

1. The majority of the output is on the sides and rear, which will be attenuated and hard to recognize in the front field response. In addition, the effective reflection distances in a normal environment will not allow for sufficient gate time for adequate frequency range analysis.

2. If you do full CSD analysis of multiple angles in fine increments from from front to rear, you will need a large turntable and you will need a true anechoic environment or mount the speaker suspended in the air with a sufficient distance from reflective surfaces, because in a normal set up, the gate time vs. frequency wavelength will not allow sufficient time window width to get proper resolution of the frequencies of about <1000-1500 Hz{1}. An even more accurate method would remove direct radiation of the driver all together, measuring only cabinet acoustic output, and this is possible, but adds great complexity to the situation. I can expand if requested.

-Chris

{1} If you shift resonances to a very high frequency, such as >1500Hz, then you can analyze them in a normal environment. This assumes an insanely rigid/high density cabinet; but to insure the cabinet resonance(s) are above the required range, an accelerometer must be used to measure the impulse response of the cabinet at multiple points.

 

[MacManNM]

Unfortunately, it's no easy matter to measure/analyze audiblity magnitude of enclosure wall resonances.

1. The majority of the output is on the sides and rear, which will be attenuated and hard to recognize in the front field response.

2. If you do full CSD analysis of multiple angles in fine increments from from front to rear. You will need a large turntable. You will need a true anechoic environment or mount the speaker in the air many feet, because in a normal set up, the gate time vs. frequency wavelength will not allow sufficient opening to get proper resolution of the frequencies <1000 Hz. An even more accurate method would remove direct radiation of the driver all together, measuring only cabinet acoustic output, and this is possible, but adds great complexity to the situation. I can expand if requested.

-Chris

I have an anechoic chamber, and a large rotary stage, along with a full compliment of automated diagnostics. But i'm not going to make those types of measurements on these speakers. This is just a fun project.

 

[WmAx]

But i'm not going to make those types of measurements on these speakers. This is just a fun project.

Okay. I was just trying to be of use.

-Chris

 

[MacManNM]

Okay. I was just trying to be of use.

-Chris

Thank you, it is good advice. Now that these prototypes are done, I am closing in on the final design for a set of towers. I believe that my next generation will incorporate those features, and a couple of others.

Here are a couple of pics with all the drivers installed. I still need to put the wood covering over them, and flush mount all of the drivers. The sound from these is quite pleasing. I had to knock down the mids ~2dB to make my ears happy. It was a compromise, for rock music they were pretty harsh, I believe this is partially due to how bad some of the recordings are. For classical, jazz, and easy listening (Sinatra Duets was fabulous) the detail and clarity was stunning. I have a Flamenco guitar cd that is an exceptional recording, I thought the guy was standing in front of me.

 

[j_garcia]

They look great Nice work.

 

[Swerd]

Yes, nice work and thanks for the photos.

For all us curious types, what are those drivers and do you have schematics of your crossovers?

 

[MacManNM]

They are Dayton alum drivers and a Vifa tweet. Crossovers are standard APC 3rd order, with some level compensation on the midrange and tweeter.

 

[Matt34]

A little late to the game but just wanted to say job well done!