Natalie P First Build - A more conventional approach

[voltage23]

Thanks to everyone who gave input into my curved Natalie P design. Alas, priorities have shifted (as they are wont to do) and now time has become the main driver. I'm not going to have the time available to put into these that it was going to take to make the translam design. I've also decided to drop the hardwood front baffle and stick with an all-MDF enclosure to save time. So, back to the drawing board. To help myself feel a little better about not getting to build my beauty, I'm thinking about incorporating a subwoofer into my design. The new plan is to build a sealed NatP atop a side-mounted subwoofer all in one box. See pics below:





The NatP portion will be the same general dimensions as the original (9" wide, 22" tall, 14" deep, 30L) leaving me with a 9"x14"x14" (17.5L, 0.6 cu. ft.) subwoofer enclosure. I'm looking for recommendations for which subwoofer to use. What would incorporate well with the NatPs and also do well in a 0.6 cu. ft. enclosure? I won't have a lot of room to incorporate a large amp, but the Dayton SA100 would fit. That would give me 100W @ 4 ohms or 75W @ 8 ohms to power it with. I'm open to other suggestions on powering the subs, too.

Thanks!

[JonMarsh]

That's a pretty tiny space for a subwoofer- how low are you hoping for it to go? And how loud? Couldn't you make it a bit taller? Get at least a cubic foot for the subwoofer?

[JonMarsh]

If you can't go bigger, go for a driver optimized for small boxes, like the RS265HO-4. You can probably do OK with 0.6 to 0.75 cu ft.

And as to the MDF, be sure to go with cabinet grade MDF, no big box store stuff.

[TEK]

Not sure about your budget, but you could go for one driver on each side - so that they will cancel out the vibrations from each other. That might or might not work better for a small space sub-design.

[voltage23]

Not sure about your budget, but you could go for one driver on each side - so that they will cancel out the vibrations from each other. That might or might not work better for a small space sub-design.

Are you referring to an iso-barik alignment? Or did you have something else in mind? I don't think I'd have the room to do an iso-barik.

[dar47]

I would be incline to do the regular mtm monitor and just build to sealed base bins to put under the mtm. If you can swing it make a solid front baffle. I have glued up poplar 3/4" x 1 1/2" on edge plane flat then wrap with and wrap with 1/8" hard board. A little more time but an awesome stiff baffle.

[jwanck11]

I'd brace the sub area - doubling up all the way around - and port it.

[JonMarsh]

I'd brace the sub area - doubling up all the way around - and port it.

With a very small volume like you're talking about, you may be better off sealed. Small sealed drivers can have a fair amount of excursion, but putting a port in a small box that is tuned low and is large enough to support a sub driver AND tune it low enough will be a problem- if it's the right diameter, it will be too long to fit.

[voltage23]

Ok, so the integrated subwoofer is probably a bad idea in that small of a volume. That leaves me with a ported tower. Here are some renders:


External dimensions are 36x9x14. Internal volume is ~ 1.6 cu. ft or 45L. The port is 3 inches in diameter by 14 inches in total length for a tuning of 28hz.
Questions:
Is that enough bracing?
Should I consider a top-firing port? How much clearance would be necessary between the port and the cabinet above it? As a reminder, here is the niche it will be going in:

[JonMarsh]

You could consider a bottom firing port, and use spikes or other feet that raise it 2-3" from the floor.

[wkhanna]

That is what i did for my Nat Ps:

https://www.htguide.com/forum/showthread.php?22639-My-Natalie-P-Journey-Begins&highlight=nat+journey+begins

you get a little extra bump in LF output for free this way, too.:W

[voltage23]

Ok, so sitting on my couch watching TV today, I realized maybe what I should do is flip the whole design upside down. My ear height will be much closer to the bottom of the cabinet than the top. Should I align the drivers at the bottom of the cabinet? If so, it would be better to have the port at the top than the bottom since it will be further from the drivers, right? I can make sure there is 3" clearance to the upper cabinets.

[JonMarsh]

Unless you're lying on the floor, I don't think that would work out- and then you'd still have floor bounce to deal with. The closer you are to EITHER boundary, the more interaction...

I find that a seated listening height is generally between 28" and 36" from the floor. Go look at the Ardent design and see where the midrange and tweeter are. That's a good reference point- put one of your mid woofers at the same level as the midrange, and the tweeter at the same level, period. Then the other mid woofer will have to be above it. Do NOT change the arrangement of the mid woofers and tweeter or it will mess up the response in the transition region.

Do NOT put the drivers close to the floor, and don't fire the port up into the air from the top of the cabinet, or I'll have to call in Evil Twin, the original designer of this system... :W

[voltage23]

Ok, point taken. These speakers are going to be sitting atop an entertainment center, so the bottom of the boxes will already be 36" off the ground. The boxes are 36" high, so with the standard driver spacing at the top (tweeter 11" from top edge), the tweeter will be 61" above the floor. Should I consider placing the drivers somewhere in the middle?

[fbov]

I suspect Jon was visualizing a floor-standing placement, where his points are well taken.

If the speakers sit on a 36" high shelf, you'd want the drivers as low in the box as you can manage. I did a sealed pair that has the minimum 12" tweeter height (sitting on 24" stands in my case).

Have fun,
Frank

[TEK]

It the speakers are going to be positioned on top of something that is 36" tall(approx 1 meter), would'nt it be better to build them as two-ways bookshelf speakers?

[Steve Manning]

Might I suggest doing a drawing of what you have in mind in it's entirety. Does not have to be detailed, but that way you get all the pertinent details out there and that way you won't have to keep changing your plans each time you reveal a new tidbit of information.

[JonMarsh]

Might I suggest doing a drawing of what you have in mind in it's entirety. Does not have to be detailed, but that way you get all the pertinent details out there and that way you won't have to keep changing your plans each time you reveal a new tidbit of information.

Bravomissimo!


You could still retain the volume necessary for the low tuning, IF you can build them big enough, by going deep with the enclosure. If that is not possible, than just build what you've drawn, and turn them upside down. It's all relative, after all...

My conservative esthetic sensibilities still rebel at the notion of a port exiting the top of the cabinet, though a straight through port will usually work better. One might as well actually put the port completely external to the cabinet, and save the internal volume and outside dimensions? :W (you are subtracting for port and driver volume in calculating your net internal volume, right?)

Sorry if I seem to be channeling ET a bit- it's been one of those weeks at work, and more of the same coming up today, Sunday, with a boss who things that I can make a 20% time effort and get an 80% result and that will be good enough for everyone...

[DFAL Minion]

if one is to implement a build at the high performance level a DFAL design delivers, one should plan thoughtfully in order to realize its full potential.

to achieve this result it is wise to plan for the tweeter to be at the listener's ear level.

[Carl V]

it's been one of those weeks at work, and more of the same coming up today, Sunday, with a boss who things that I can make a 20% time effort and get an 80% result and that will be good enough for everyone...

oh my, Dilbert comes to life.
life imitates art (comix)

[voltage23]

Thanks for taking time out of your busy week to help, Jon!

Here is a rough sketch of what I'm talking about:



I'm pretty constrained in my volume. Right now I have the top of the box aligned with the top of the TV, though I could go a couple inches taller. Depth is pretty constrained too: the front of the TV is 15.5" from the back wall. I was planning to make it 14" deep and setting it flush with the TV, leaving 1.5" between the back and the wall (is that enough? can I go tighter?) If I decided to make it flush with the front of the base cabinets, I could go deeper, but I think it would look funny.

What do you think is the best port location? If I did do a top port, no one would ever see it.

BTW, the reason I'm not doing a simple bookshelf is because I'm looking for more low end output as there are no current plans for a subwoofer.

Thanks for everyone's input!

[wkhanna]

i recommend placing the speakers at the front edge of the base cabinets for a few reasons....all pertaining to getting best sound quality......

....you do not want anything on the sides of the speakers, such the cove sides, to cause reflection of the sound wave.
....it will also allow you to use the maximum depth dimension utilizing all available space for your speaker cabinet volume. you could even put the wire terminals on the side, in this case, to maximize every last square inch possible.

also, you may be able to use a front-facing slotted port at the top of the front baffle? slotted ports are not typically a first-choice. but for aesthetics in this case, it may provide a viable option....?

.....but i will let the more experienced designer-members comment on that idea first, however....

[Steve Manning]

also, you may be able to use a front-facing slotted port at the top of the front baffle? slotted ports are not typically a first-choice. but for aesthetics in this case, it may provide a viable option....?

Bill ...... I was thinking the same thing on the slotted port. I've seen Wilson Audio going that route of late.

[fbov]

Thanks for taking time out of your busy week to help, Jon!

Here is a rough sketch of what I'm talking about:
[ATTACH=CONFIG]26907[/ATTACH]

What I was seeing in my mind's eye as well.

Your application doesn't allow for ideal tweeter height, but as long as you're far enough away (big room), and things are as flush as possible, you should still be in the response sweet spot. I have four NatP's; my mains tweeters are a foot low, the front-wide tweeters at ear level, at ~9 foot listening distance. There's no obvious sonic difference due to slightly above-axis listening location for one pair. Room placement, and getting the baffle and TV flush with the rest of the furniture will be a bigger effect.

Have fun,
Frank

[voltage23]

I'd prefer to avoid a slot port as it will be more difficult to construct and paint. Any opinions on what would be the best location to put a flared round port?

BTW, listening position is 18 feet away (it's a big room!)

[voltage23]

So I've decided to go with a front port. Now I just need to decide what size to make the cabinet. I will take your advice and align the front baffle with the front edge of the cabinet. That gives me more depth to play with. If I go 22" of depth, I end up with a volume of about 2.6 cu ft or 75 litres. Is that too big? If so, why? What box size and port tuning would you suggest given that I won't have a sub and the room is a large open-concept so there won't be a lot of room gain?

Frank, I understand you went with 65 litres tuned to 32 hz. Why did you decide to go with a larger box with a higher tuning than the recommended 50 litres @ 28 hz? If you're going to go bigger, why not tune the 65 litres to 28 Hz as well?

[JonMarsh]

Unibox is your friend... (runs in Excel) You are the Y in DIY- they/we can get you an answer, but you might learn fun things playing with it yourself...

Keep in mind that enclosure volume numbers are always based on net internal volume- if you increase the box beyond the optimum point for the driver, what happens is that you wind up with what is called an EBS alignment- yes, it will go lower, but the output level is shelved down at some point above the box Fb.

[voltage23]

I modeled a couple options in WinISD (Is this ok, or do I need to get Unibox?)


Yellow is the original recommendation: 50L @ 28 hz
Blue is fbov's 65L @ 32 hz
Pink is 65L but tuned to 28 hz instead
Green is the max volume I can fit: 75L tuned to 24 hz for really deep f6

Are there any disadvantages to the green option? I think I read something about xmax might become a problem in EBS enclosures? Is lower tuning always better, or is there a point of diminishing returns where there isn't much content anyway?

[BobEllis]

You can model excursion in WinISD. Select the excursion view then in the project window, click the signal tab, increase power until you run out of excursion. Now flip over to SPL and see if the result is acceptable.

[voltage23]

Ah, I was using WinISD, not WinISD Pro. I'm confused by the excursion plots in WinISD Pro. At even low watts, all of the enclosures exceed xmax at low Hz. Is there some sort of highpass filter that needs to be applied to prevent that, or do we just count on there not being any content in that range? How do we protect the drivers from damage?

I'm assuming that since many people have built the 50L @ 28 Hz version that it's ok. Looks like 65L @ 28 Hz has just a hair more excursion, so I'm going to assume it's ok too. That's probably what I'll go with since the 75L @ 24 Hz has a fair bit more excursion and seems like a step too far?

Changing direction for a moment: Looks like there are two options for the L4 inductor:
1) 18AWG 0.44mH 0.31 DCR (PE #255-226)
2) 15AWG 0.47mH 0.20 DCR (PE #255-408)

I don't have a lot of experience working with electronics. How hard is it to reduce the inductance of option 2 to the necessary 0.44? Am I better off just sticking with option 1 despite the increased resistance?

[cochinada]

Is there some sort of highpass filter that needs to be applied to prevent that, or do we just count on there not being any content in that range? How do we protect the drivers from damage?

It's always recommended to apply a HP filter on a subwoofer whenever you see those problems of over excursion. That's also what I did on mine.

How hard is it to reduce the inductance of option 2 to the necessary 0.44? Am I better off just sticking with option 1 despite the increased resistance?

It's not hard at all. You just need to unwind part of it until you get to the target mH. Of course you will need to have a multi meter that is capable of measuring inductance.
OTOH you need to simulate your crossover first and see what impact has the different resistance. BTW, by unwinding part of the coil you will also lower the resistance a bit.

[BobEllis]

Which drivers are you modeling? Don't expect the RS-180s to have a whole lot of output below 50 Hz. Yes, as noted above a high pass filter is a good idea even if you aren't crossing to a sub. It can be as simple as going into your receiver's settings and selecting speaker size small and then setting the crossover frequency as low as your speakers can handle. Try to get 40 Hz.

What part number is called out for the inductor? If it's the 18 ga. part, use that because the resistance was used in the design calculations. If not specified, what is the DCR called out for the part? If lower than the 18 ga. part unwind as Joaquim suggests. Sorry I can't be more specific but I'm on my phone and can't easily look at the schematic and respond.

[DFAL Minion]

this schematic is for the shielded version:

[BobEllis]

Thanks for that.

Looking at the BOM Evil Twin suggests a .45 mH sledgehammer inductor from Madisound. That value isn't available in the sledgehammer series but is available in the sidewinder series. It's a 15 ga. coil with 0R2 DCR. So, either unwind the Parts Express part until the value measures .44 mH or buy the Madisound part. Don't have an LCR meter? Shipping from Madisound is probably less expensive than buying a meter.

[voltage23]

Thanks for the inputs, all! Parts are ordered!

[fbov]

... What box size and port tuning would you suggest given that I won't have a sub and the room is a large open-concept so there won't be a lot of room gain?

Frank, I understand you went with 65 litres tuned to 32 hz. Why did you decide to go with a larger box with a higher tuning than the recommended 50 litres @ 28 hz? If you're going to go bigger, why not tune the 65 litres to 28 Hz as well?

Let's be clear. I built mine in 2008, so the 65L towers tuned to 32Hz use shielded drivers.

While modeling is very helpful, be careful in your comparisons. The prediction tells me you're modeling the unshielded drivers, which need less volume than the shielded drivers for similar response profiles. The 50L/28Hz is close to how my 65L/32Hz shielded towers modeled, but much less severe (F3 is 32Hz). I was looking or a good room match... ended up pretty good.


And there are no differences in XO circuit for shielded vs. unshielded drivers, just differences in bass alignment.

That gets back to "what bass alignment do you want?" The answer, in my mind, is part preference (your call), part application, which I can address. For flat bass, you'd like to match the bass roll-off to the room gain profile so you get extension, not boominess. The problem is the room gain threshold depends on room dimensions.

A rough estimate of the room gain is: Onset - 565/shortest room dimensions, maximum effect - 565/longest room dimension
(565 is 1/2 the speed of sound, equal wavelength x frequency, so this finds the 1/2 wave resonance for normal mode fundamental frequency.)

My room is 7.5' ceilings with a 22' long dimension. Room gain starts around 75Hz and is nearly full strength by 25Hz. My design choices give me an F3 of 32Hz, right where room gain is giving me 3-4dB of boost. I end up fairly flat response, room mode dips not withstanding.

You can't do this in a room 30' room with 12' ceilings. If room gain reinforcement cuts in below ~40Hz, I would suggest a slightly peaked alignment, sacrificing extension so the bass level is more linear as low as it goes.

And this gets into the cone excursion discussion.
- I design for rated excursion at 1/2 rated power.
- I ignore the predicted loss of cone control below tune; no infrasonic filters on a 10-octave design!
- But I also design so I'm close to flat to start...

... Because the danger of cone over-excursion is greatly increased if you use EQ. Tune really low, with no room gain, and EQ could boost 30Hz output by 4-6dB, where the speaker is vulnerable to excess power. With unshielded drivers, I like a 40L box tuned to 35-40Hz for flat to a slight peak.

And checking my BOM, my L4 is series 20uH and 25uH, 18 ga. coils. Note that designers frequently spec resistance for inductors when that resistance is important. That's why I didn't spend money on lower resistance.

Have fun,
Frank