A little help building some nice cabinets?

[Char93]

Hello everyone!!! 4-5 years ago, when I had no idea of how to create a loudspeaker, I just created a sketch and gave it to my father, who is amazingly good at creating things. We built together a couple of cabinets which ended up being a disaster. 18 mm thick pinewood was used in the construction, and the result ended up weighing more than 50 kg per speaker. The speakers are over 1,30 m high, 0,31 wide and around 0,6 m deep. Inside, a kind of labyrinth/transmission line was built, but not only was it NOT calculated, but also it was futile, since there was no opening (they're basically sealed boxes with a labyrinth inside). The cabinet is 4 ways, but it has a 3-way filter for a 10" bass woofer, a 6,5" mid woofer and a 3" tangband driver for the highs, and then it has a separate high-pass filter with a tweeter hooked on it. The result is a very heavy, not good sounding speaker that, on top of which it has two "outs" (one for the tweeter crossover, and one for the 3-way filter).

The speakers were built, tested and then kept in a storage room for a while since I didn't have an appropriate space for them. Now that I do, I've been hearing them and they sound extremely colorful on the mids, veeery lacking on the lows, and OK on the highs. The sound is not "bad", just too dissimilar.

I was thinking of creating an appropriate 4-way filter that cuts at 150 hz, then at 500 hz, and then again at 6000 hz, so that I get effective 4-way filtering. But then again I have no idea on how to do it (ok, I've estimated what kind of capacitors and inducters I'm going to use, but I have no idea on the resistances, how to "build" an appropriate inductance, etc.

The other problem I've encountered is the cabinet. It is too big and I don't think the sound is as big as it should be, so I've been thinking of reducing its depth and removing the labyrinth inside. Then, I kind of really want to build a transmission line, but I've been told the qts of the woofer is too low to fit such a construction, and that I would be better off building a sealed cabinet (or a bass reflex for that matter). Is it so inappropriate to build a transmission line? The technical specifications of the drivers are as follow:

FPW-100 (6,5") first, FPW-250 (10") second:

(RMS)
50 W
125 W
Frequency range
100-5500hz
48-4500hz
Impedance
8 Ohm
8 Ohm
SPL
90 Db 1W/1m
92 Db 1W/1m
(Fs)
91 Hz
50 Hz
(Re)
7.7 Ω
7.4 Ω
(Qms)
6.34
4.28
(Qes)
1.30
0.88
(Vas)
10.35 l
57.83 l
(Cms)
0.37 mm/N
0.32 mm/N
(Xmax)
±3.65 mm
±4.3 mm

These would be the drivers for the low and mid-low frequencies. For the high-low frequencies I would be using a TangBand 3" W3-593SF and for the real highs, a Beyma T2010 tweeter.

In the attachments you will be able to see the front panel of the cabinet when I was building it, its current state and a sketch of how I would it to be, in the event a transmission line was doable (neck and mouth measures need to properly be implemented in the sketch)

I would need a hand on what type of cabinet would suit best the drivers, some cabinet calculator that is accurate enough (I'd like to maintain height and width, modify depth) and then, the most important thing, I'd like to create a 4-way crossover appropriate for the drivers.

Thank you so much!!!

PS: Sorry if there are some English mistakes, I'm Spaniard.

 

[pkitt]

From your listed T/S values, I calculated Qts of the 10" woofer at 0.73. This is quite high and usually does not bode well for TL use. In addition, the resistance of a series crossover inductor will raise Qts further, compounding the problem. The woofer's Fs is rather high, too, but that would actually allow (require) an optimum system tuning frequency to be significantly lower than 50 Hz due to the high Qts. Last, this woofer's Xmax is rather underwhelming. I'd be willing to take a stab at modeling a TL for this woofer if you also provide values for Sd and Le, but I can't promise a good outcome. I presume the T/S values you posted are those published by the manufacturer, which may or may not be reasonably representative of your woofers.
Paul

 

[Char93]

I take it that it would be rather difficult to model a TL with a good outcome. Would you recommend any other type of cabinet? I really would like a TL, but if it's not feasible, I understand.

I can only give you an approximated sd, taking into account it is a 10" driver. It should be around 346,19 cm^2. As for the Le, I have no idea. I could ask the manufacturer for these data, since they provided me with most of the data they had not gathered in the technical data available at the website.

If you think a TL is suitable, then I could ask the manufacturer for such numbers.

Thank you so much!

 

[Sonce]

The other problem I've encountered is the cabinet. It is too big and I don't think the sound is as big as it should be, so I've been thinking of reducing its depth and removing the labyrinth inside.

Do not reduce the cabinet depth!
10" and 6.5" share the same enclosure volume and that is not good. You must put divider inside just below 6.5 woofer, to create sealed box for it.

 

[Char93]

Do not reduce the cabinet depth!
10" and 6.5" share the same enclosure volume and that is not good. You must put divider inside just below 6.5 woofer, to create sealed box for it.

Hello. The sketch you're looking at is not the actual building of the cabinet. The cabinet is 2x deeper. Sorry, I don't think I attached a picture of it. The sketch I uploaded is of a design that I would use to modify it, in case I could/would.

The two woofers are separated, so there is no interference among them. The current volume of the woofer enclosure is of around 240 liters, it is sealed, and the low frequencies are not very overwhelming (they're actually rather underwhelming).

Thanks.

 

[giralfino]

Hello everyone!!! 4-5 years ago, when I had no idea of how to create a loudspeaker, I just created a sketch and gave it to my father, who is amazingly good at creating things.
...

IMHO 4-5 years later it seems you still have no idea how to design a speaker.
There are so many problems in what you want to do, I will point you some here in no special order.

Material: solid wood is not an appropriate material for an enclosure, especially a large one, because of its mechanical properties. MDF or birch ply are far better options.

Kind of enclosure: the choice of the enclosure depends on the T/S parameters of a driver. It seems that your woofer is optimized for a somewhat compact sealed box, and you should build that. A TL has to be carefully designed with real T/S parameters.

Frequency Response: above 200-300Hz you need the FR of the drivers in order to create a crossover, using at least simulation tools. This is because you cannot assume the FR and impedance are flat over a great range.

Complexity: it not really easy to design a 2 way crossover, because you need either measure in box FR of all drivers, or the ability to simulate what happens on your baffle with a FR provided by the manufacturer. A 3 way or worse a 4 way are far, very far more difficult to design. I would never attempt to design a 3 or 4 way as a your first design.

3 vs 4 way: if you still think you can succeed in designing a multi-way speaker, probably a 3 way is enough with the drivers you have: the woofer up to the baffle step frequency, and a mid up to a tweeter. The TB fullrange can sustain easily a 300-500 Hz crossover.

Choice of drivers: frankly, your driver choice is poor. The 10" and 6.5" drivers are PA stuff, lot of SPL, unknown FR, no deep bass. Besides that the TB mid has a too low sensitivity for being mated to your 10" driver. If you want to use the 10" driver, pick up a mid with a sensitivity of 90dB, and retain the tweeter.

Ralf

Is seems you are only seeing the box problem, but the bigger one is designing an appropriate crossover. Without measurement or simulation capabilities you are far better in building a reputable design instead of trying to put together some drivers without knowing how to design a tailored crossover.

 

[Char93]

IMHO 4-5 years later it seems you still have no idea how to design a speaker.

I know. Not that I have seriously researched within these years. This is why I'm here. To learn from you guys.

Material: solid wood is not an appropriate material for an enclosure, especially a large one, because of its mechanical properties. MDF or birch ply are far better options.

I know what material is best now, I didn't know a few years back. I'm going to use the wood I have because I would like for it to be a mod which didn't involve buying new material (until completely necessary, for example, a new inductor for a crossover, or something of the like).

Kind of enclosure: the choice of the enclosure depends on the T/S parameters of a driver. It seems that your woofer is optimized for a somewhat compact sealed box, and you should build that. A TL has to be carefully designed with real T/S parameters.

OK, I just would've liked a TL, but if it's not feasible then I shall modify it into a smaller sealed enclosure. If there's any chance of a TL that would sound equally good or better than a sealed box, then I'll go with it.

Frequency Response: above 200-300Hz you need the FR of the drivers in order to create a crossover, using at least simulation tools. This is because you cannot assume the FR and impedance are flat over a great range.

I'd like to cut the frequency for the woofer at the lowest I can. I assumed 150-200 hz would be fine. If it is not, I'd be more than happy to cut it at a higher frequency. I know nothing of crossovers or simulation tools, other than the fact that crossovers use inductors, capacitors and resistances.

Complexity: it not really easy to design a 2 way crossover, because you need either measure in box FR of all drivers, or the ability to simulate what happens on your baffle with a FR provided by the manufacturer. A 3 way or worse a 4 way are far, very far more difficult to design. I would never attempt to design a 3 or 4 way as a your first design.

I created a cabinet for a project at school when I was 16, and then I created this disaster at 18. It would be my third design, but I understand that it's far too complex, this is why I'm asking for help here.

3 vs 4 way: if you still think you can succeed in designing a multi-way speaker, probably a 3 way is enough with the drivers you have: the woofer up to the baffle step frequency, and a mid up to a tweeter. The TB fullrange can sustain easily a 300-500 Hz crossover.

As long as choosing between a 3 or 4 way crossover is just a matter of complexity, I'm fine with the 4 way one. The reason is simple: The cabinet is already adapted for 4 different drivers. Reducing the depth of the cabinet is one thing, but changing the design/height/other things would be much more challenging. Only if the sound would be wronged because of the use of a 4-way filter and 4 drivers, then I would consider going back to 3 ways.

Choice of drivers: frankly, your driver choice is poor. The 10" and 6.5" drivers are PA stuff, lot of SPL, unknown FR, no deep bass. Besides that the TB mid has a too low sensitivity for being mated to your 10" driver. If you want to use the 10" driver, pick up a mid with a sensitivity of 90dB, and retain the tweeter.

You don't need to tell me... I unfortunately was broke when I decided to do this. For me, a budget of 1000€ was extremely high already. Unfortunately, again, it was a mistake spending most of these 1000€ in "high quality wood" which turned out to be "low quality speaker material". I considered the wrong sources back then, which said that real wood was the best way to go. Now I'd like to make the most of what I have. The woofer is 92 dB and the mid-woofer is 90 dB. If you reeeally think the best way to go is invest some extra 200€ at most and use two decent / good 10" drivers, forget about the 6,5" and use 2 3" fullrange tangband units, I could consider that as well... Although modifying the front panels would be such a pain.

Is seems you are only seeing the box problem, but the bigger one is designing an appropriate crossover. Without measurement or simulation capabilities you are far better in building a reputable design instead of trying to put together some drivers without knowing how to design a tailored crossover.

I consider the crossover problem to be as important or more important than the box itself. This is why I would love to learn how to design a tailored crossover.

Thanks.

 

[pkitt]

Okay, here's what we can do just to see if a TL has any possibility of working. I will model with an Sd of 346 cm2. The value of Le is not particularly critical, mostly affecting sensitivity. The model will be just an example of the best I can optimize for this 10" woofer, even though it may not be a particularly good choice, and may not be a perfect "fit" for your cabinet. The results may be a total bust or may be decent. If it looks decent, then you could decide to go further by getting someone to measure the actual T/S values of your woofers.
Paul

I take it that it would be rather difficult to model a TL with a good outcome. Would you recommend any other type of cabinet? I really would like a TL, but if it's not feasible, I understand.

I can only give you an approximated sd, taking into account it is a 10" driver. It should be around 346,19 cm^2. As for the Le, I have no idea. I could ask the manufacturer for these data, since they provided me with most of the data they had not gathered in the technical data available at the website.

If you think a TL is suitable, then I could ask the manufacturer for such numbers.

Thank you so much!

 

[Char93]

Thanks a lot, pkitt. Could you maintain the height and width of the current box for the TS line, though? These are 131,9 cm and 34,2 cm respectively. Thanks.

 

[pkitt]

I did some exploratory modeling. While this woofer will "work" in a TL there is no benefit from doing so, and you might as well just stick it in a sealed box. Other TL designers, like "GM", might get different/better results, however. I did maintain the cabinet's width and height, BTW.
Paul

Thanks a lot, pkitt. Could you maintain the height and width of the current box for the TS line, though? These are 131,9 cm and 34,2 cm respectively. Thanks.

 

[Sonce]

The sketch you're looking at is not the actual building of the cabinet. The cabinet is 2x deeper.
The two woofers are separated, so there is no interference among them. The current volume of the woofer enclosure is of around 240 liters, it is sealed,

OK, this explains everything.
Volume for the 10" woofer enclosure should be around 80 liters, sealed and well stuffed.

 

[Char93]

Alright, thank you pkitt. Then I shall as well just use a sealed box. Could you tell me what an appropriate range of volume for the woofer box would be (in liters)? Anything I should bear in mind when remodeling the cabinet?

Also, if I were to build a 3 way system, what would you choose for the mids range, two 3" of my tangbands or 1 6,5" of my fonestar?

Any help at all building a tailored 3 way crossover?

Thanks.

 

[Char93]

Thank you, pkitt and sonce. Then I shall stick to a sealed box of around 80 liters. I posted earlier but something must have happened. In the other post I said that, in the event I keep the 3 ways, would you recommend me to use 2 of the 3" tang band drivers, or 1 of the 6,5" fonestar for the mid frequencies?

A little help for a tailored crossover?

Thank you so much!

 

[pkitt]

In the tapered TL I modeled, the line's net volume was ~85 liters and F3 was about 50 Hz. The modeled response, however, was virtually what you'd probably achieve with a sealed box of about the same volume since the rate of rolloff below F3 appeared to be about 2nd-order. I can't help you with your other needs, however.
Paul

Alright, thank you pkitt. Then I shall as well just use a sealed box. Could you tell me what an appropriate range of volume for the woofer box would be (in liters)? Anything I should bear in mind when remodeling the cabinet?

Also, if I were to build a 3 way system, what would you choose for the mids range, two 3" of my tangbands or 1 6,5" of my fonestar?

Any help at all building a tailored 3 way crossover?

Thanks.

 

[system7]

This is not a LITTLE help needed, but a LOT of help IMO!

Let's look at the drivers:
Fonestar - Sonido, Megafona y Comunicacin
Fonestar - Sonido, Megafona y Comunicacin
Tang Band W3-593SF 3" Full Range Silver Cone Driver
Beyma T2010 - Beyma T2010 soft dome tweeter. Beyma speaker components available at US Speaker.

10" PA 92dB bass, 6.5" PA 90dB bass, Tangband 3" 86dB fullrange, Beyma 92dB 2kHz soft dome.

I think that bass is not dissimilar to an Eminence Alpha10A:
Speaker Detail | Eminence Speaker

So about 50L reflex IMO. Which is what most 10" PA speakers get anyway.
Tekton Design Enzo XL loudspeaker | Stereophile.com

Now you'd often do this sort of speaker as a two way with a horn tweeter crossing around 2-2.5kHz. But people seem to have luck with a sensitive tweeter, which you may have there.

Eminence would probably suggest 2mH and 10uF on the bass, 3.3uF and 0.3mH with attenuator on the tweeter. A three way is a bit more complex.

I would guess you'd end up with something like this, using the 10", 6" and tweeter:
HW 3/130 NG - 8 Ohm
Just hope the tweeter is loud enough.

 

[GM]

Other TL designers, like "GM", might get different/better results, however. I did maintain the cabinet's width and height, BTW.
Paul

Yeah, with specs like these, it's a 'crapshoot' between a MLTL, inverse tapered TQWT and sealed alignment. If a faster LF roll-off below tuning is acceptable, then the MLTL at least yields a bit more mid-bass 'punch' or acoustic BSC depending on the room, etc., though without modeling it, don't know if it needs to be bigger than the existing cab to get enough worthwhile gain.

Did you scope out a MLTL?

GM

 

[pkitt]

No, I just modeled a tapered TL primarily because the OP was hoping it would work and had created a cabinet for it.
Paul

Yeah, with specs like these, it's a 'crapshoot' between a MLTL, inverse tapered TQWT and sealed alignment. If a faster LF roll-off below tuning is acceptable, then the MLTL at least yields a bit more mid-bass 'punch' or acoustic BSC depending on the room, etc., though without modeling it, don't know if it needs to be bigger than the existing cab to get enough worthwhile gain.

Did you scope out a MLTL?

GM

 

[system7]

I think this whole cabinet discussion is a red herring.

You can do cabinets almost any way with a 10" bass PA speaker.

Google images for: 2Pi speaker

Big boxes, big baffles for sure. 50L reflex up to as big as you like closed box.

But actually Wayne Parham seems to add a smallish vent with whatever sized box. Small aperture vent and quite short because the Fs is highish. Truth is, that high Qts bass isn't very fussy from what I've seen.

Char93 really needs to rethink the drivers and driver layout IMO, because the tangband has no place in this design as mentioned earlier. I think it starts with a good crossover here into a two or three way.

An efficient 10" PA bass with a corrugated surround is never going to give really deep bass, which is just the physics of it, but ought to satisfy. And high Qms seems to be a good thing for detail, which is why people like the whole PA driver approach.

 

[Char93]

Here's a video of my testing with the XO/3ways. I think the best sound in the mid frequencies is given by one TB, not two TBs or the Fonestar 6". I think I'll stick to one TB. Then I'll only need to pad down the tweeter a bit.

Thanks.

https://www.youtube.com/watch?v=VR3zHhfkY8Y

 

[system7]

The benefit of series wiring a couple of closely spaced 3" tangbands is you get 6dB more SPL level at the same power level on each driver, so they ought to distort less at high volume. Polarity is +-+.

If you attenuate the mids with a 15R resistor across them, they behave like 8 ohm units anyway. Just a thought. It's your project.