Ares pestering the DIY community

[ARES24]

Periodically I think of questions that I have difficulty finding direct answers for. I will just post them here;

1. Does the orientation of a port make any difference to the sound of a speaker? example; I have the ER18's, they have a round rear port. If I changed the port (same height centered) to a square port with the same area and same length, front or back, what would be the difference if any?

2. Using the same vertical dimensions of said ER18s (examples I can see allow for better brain matter!), but change the shape of the towers, what worries are caused?

 

[TLS Guy]

Periodically I think of questions that I have difficulty finding direct answers for. I will just post them here;

1. Does the orientation of a port make any difference to the sound of a speaker? example; I have the ER18's, they have a round rear port. If I changed the port (same height centered) to a square port with the same area and same length, front or back, what would be the difference if any?

2. Using the same vertical dimensions of said ER18s (examples I can see allow for better brain matter!), but change the shape of the towers, what worries are caused?

Actually the front of the speaker is the best place to put the port. If properly designed, there should be phase inversion in the cabinet and the port in phase with the woofer. The different shape will alter port length slightly as the end correction will be different.

Changing the shape of the cabinet will change reflected resonance modes. The dimensions are best different. A cube would be the worst. Air volume inside must be the same. Non parallel surfaces help also.

 

[Swerd]

1. Does the orientation of a port make any difference to the sound of a speaker? example; I have the ER18's, they have a round rear port. If I changed the port (same height centered) to a square port with the same area and same length, front or back, what would be the difference if any?

The only thing I might add is that if the port creates audible "chuffing" it will be more easily heard if the port is in the front. In the ER18 MTM cabinet, the port diameter is large enough to avoid that, so I doubt if putting it on the front would be harmful.

TLS's comments are interesting and I hadn't known that before.

2. Using the same vertical dimensions of said ER18s (examples I can see allow for better brain matter!), but change the shape of the towers, what worries are caused?

In the MLTL design, cabinet vertical dimensions (pipe length) obviously matter. So does the width of the front baffle. It affects the baffle step frequency. You can't make the front much narrower than 9" because the woofers couldn't fit. If you make it wider, the amount of baffle step compensation built into the crossover might have to be also change – essentially requiring a crossover redesign.

If you can't change the vertical or horizontal dimensions of the cabinet, how do you change the shape of the towers? Changing the depth while keeping the height and width the same will adversely affect the bass tuning.

 

[ARES24]

If you can't change the vertical or horizontal dimensions of the cabinet, how do you change the shape of the towers? Changing the depth while keeping the height and width the same will adversely affect the bass tuning.

I am thinking about non-rectangular designs. Instead of seeing a rectangle when you look straight down from the top, maybe a trapezoid. I haven't some across 'baffle step compensation' in my generic reading yet. I have seen people suggest a couple books for loudspeaker design but have never been smart enough to actually remember them, suggestions?

 

[Swerd]

I am thinking about non-rectangular designs. Instead of seeing a rectangle when you look straight down from the top, maybe a trapezoid. I haven't some across 'baffle step compensation' in my generic reading yet. I have seen people suggest a couple books for loudspeaker design but have never been smart enough to actually remember them, suggestions?

I don't know of a DIY speaker building book that discusses baffle step compensation (BSC). Here are two online links that describe what baffle step is and how to compensate for it:

BSC made simple (and why it may be important to you) — Audioblog

http://www.quarter-wave.com/General/BSC_Sizing.pdf

Narrow speakers such as your ER18 MTMs, will have a more noticeable baffle step than speakers in wider cabinets. Dennis Murphy included BSC in the crossover assuming the front baffle of the speakers will be roughly 1 to 2 feet away from the wall behind them. It adjusts the bass and mid range to be flat through that range when the speakers are about that far from the wall behind them. If closer to the wall, less compensation is needed, and if further away, more is needed.

Once I learned what speakers sound like with and without proper BSC, I was amazed to see how few commercially available speakers have it done correctly.

Here is one unidentified but well known 2-way bookshelf speaker that lacks enough BSC (in my opinion). See the rise in loudness starting just above 500 Hz that dips back down above 2 kHz?

It can over emphasize voices and brass, but in a way that actually smears the sound. You will lose clarity and detail, not emphasize it. Depending on the music, it can make the mid range sound more crisp and detailed than was actually in the recording. At its worst it sounds honky and fatiguing.

The down side to BSC is that it makes a speaker less sensitive. Manufacturers are aware that many buyers look at how loud a speakers can get, but ignore the frequency response across the critical mid range.

 

[TLS Guy]

The only thing I might add is that if the port creates audible "chuffing" it will be more easily heard if the port is in the front. In the ER18 MTM cabinet, the port diameter is large enough to avoid that, so I doubt if putting it on the front would be harmful.

TLS's comments are interesting and I hadn't known that before.

In the MLTL design, cabinet vertical dimensions (pipe length) obviously matter. So does the width of the front baffle. It affects the baffle step frequency. You can't make the front much narrower than 9" because the woofers couldn't fit. If you make it wider, the amount of baffle step compensation built into the crossover might have to be also change – essentially requiring a crossover redesign.

If you can't change the vertical or horizontal dimensions of the cabinet, how do you change the shape of the towers? Changing the depth while keeping the height and width the same will adversely affect the bass tuning.

If the port chuffs at a normal listening volume, then that is not a speaker worth owning.

I should have mentioned BSC. You need to keep the front as narrow as possible.

BSC is one of the biggest arguments in favor of active crossovers. You can only get BSC right under artificial test conditions with a passive crossover. To really get BSC right, and that is essential for truly satisfying reproduction, I believe and active solution is mandatory. To get this essential parameter right under any room/environment conditions mandates an active design.

 

[BoredSysAdmin]

Here is one unidentified but well known 2-way bookshelf speaker that lacks enough BSC (in my opinion). See the rise in loudness starting just above 500 Hz that dips back down above 2 kHz?

"Unidentified" ? ..... Riiight....

Spoiler

SoundStage! Measurements - Ascend Acoustics Sierra-1 Loudspeakers (9/2007)

 

[Swerd]

Narrow speakers such as your ER18 MTMs, will have a more noticeable baffle step than speakers in wider cabinets. Dennis Murphy included BSC in the crossover assuming the front baffle of the speakers will be roughly 1 to 2 feet away from the wall behind them. It adjusts the bass and mid range to be flat through that range when the speakers are about that far from the wall behind them. If closer to the wall, less compensation is needed, and if further away, more is needed.

The down side to BSC is that it makes a speaker less sensitive. Manufacturers are aware that many buyers look at how loud a speakers can get, but ignore the frequency response across the critical mid range.

I could have explained this more clearly:

BSC lowers the stepped-up portion of the mid range to the loudness level of the bass below the baffle step frequency. Because of that, it lowers the overall sensitivity of the speaker.

"Unidentified" ? ..... Riiight....

My only point was to illustrate what the (apparently) uncorrected baffle step looks like in a frequency response curve from a speaker in a narrow cabinet. That graph is a good example. It is not my intention to name the speaker or criticize it. It is widely sold and liked by many. And, I'm told that more recent versions have corrected that.

 

[BoredSysAdmin]

My only point was to illustrate what the (apparently) uncorrected baffle step looks like in a frequency response curve from a speaker in a narrow cabinet. That graph is a good example. It is not my intention to name the speaker or criticize it. It is widely sold and liked by many. And, I'm told that more recent versions have corrected that.

No arguments from me at all. It's the curious person in me decided to "expose" that unnamed manufacturer Spoiler tag is there to show that
Sorry, I don't have anything useful to add to this conversation... sorry ... going away now

p.s: It's good to know that such acclaimed speakers is not perfect

 

[Swerd]

I am thinking about non-rectangular designs. Instead of seeing a rectangle when you look straight down from the top, maybe a trapezoid. I haven't some across 'baffle step compensation' in my generic reading yet. I have seen people suggest a couple books for loudspeaker design but have never been smart enough to actually remember them, suggestions?

To answer your original question, I think that could work. If you maintain these features of the original design the speaker should sound the same as the original design without requiring a new crossover:

• Original 9" wide front baffle
• Locations of the drivers
• Height of the cabinet (pipe length)
• Total interior volume of the cabinet

If the height is the same, all you have to do is adjust the trapezoid dimensions so the cross-sectional area of the pipe matches that of original cabinet (interior depth × interior width).

And yes, if you build a rectangular port, it will work as long as it's cross-sectional area and length are the same as with the original cylindrical tube. You should be able to mount it on the front baffle without audible chuffing. You could use a roundover bit on a router to make mildly flared ends of the port.

That trapezoid cabinet will be harder to build, but it might have less resonant vibration than a rectangular shaped cabinet with parallel walls.

 

[lsiberian]

1. I always believe in porting to the front, but I can do the back in a pinch. My reasoning is I want the port in phase with my

2. I don't believe in porting mains unless they are doing subwoofer duty. If I must port them I insist on a 40hz tune or lower so the porting doesn't interfere with your subwoofer main integration.

The beauty of DIY audio is you decide what you want to do and how to implement it. You can make a design that works for you. Find a woofer that meets your goals and the rest is pretty straight forward IMO.

 

[lsiberian]

To answer your original question, I think that could work. If you maintain these features of the original design the speaker should sound the same as the original design without requiring a new crossover:

• Original 9" wide front baffle
• Locations of the drivers
• Height of the cabinet (pipe length)
• Total interior volume of the cabinet

If the height is the same, all you have to do is adjust the trapezoid dimensions so the cross-sectional area of the pipe matches that of original cabinet (interior depth × interior width).

And yes, if you build a rectangular port, it will work as long as it's cross-sectional area and length are the same as with the original cylindrical tube. You should be able to mount it on the front baffle without audible chuffing. You could use a roundover bit on a router to make mildly flared ends of the port.

That trapezoid cabinet will be harder to build, but it might have less resonant vibration than a rectangular shaped cabinet with parallel walls.

That would be a tough glue up IMO. Definitely use something to line up the boards like dowels or a notch of some kind. To me that's probably the biggest challenge in construction assuming you have a good table saw.

 

[ARES24]

That would be a tough glue up IMO. Definitely use something to line up the boards like dowels or a notch of some kind. To me that's probably the biggest challenge in construction assuming you have a good table saw.

I have two good table saw's and a biscuit jointer.. lets get silly!

 

[ARES24]

To get this essential parameter right under any room/environment conditions mandates an active design.

But this is beyond my current abilities .... Back to the internets!!!!

 

[ARES24]

Active crossovers makes me feel like I will need a lot more amps... This is not a con, just a timetable thing......

Is there a 'recommend' vs 'avoid' list, I was just looking at a few models on PE and realized I know .. nothing.... What are important factors to look at?

 

[ARES24]

I was asking about strange designs because a friend asked about being able to mount a pair (upside down) on basement posts. BSC will not work apparently if the speakers are not within appropriate distance of the wall. That idea fell through anyway