Looking to build cheap, efficient, good sounding floorstanding speakers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Ap said:
So an 8ohm tweeter and an 8ohm woofer results in a system nominal 8ohm load
I would go for a 2way MTM, bass reflex for efficiency - as the drivers are only 86db you can get up to 92db by parralleling the bass drivers ith a usable midband of around 88-90 - means it's easy to match the tweeter.
I wouldnt series bass drivers - from my experience it just doesnt sound right, partly due to the huge increase of impedance at resonance (peak in the range of 60ohms), also the inductor size has to double (so cost & losses increase).

an 8 ohm woofer and 8 ohm tweeter will have a nomial of 8 ohms but impedance can drop as low as 4 ohms in the pass band. this is ok but once you have impedance dropiing below 3 ohms in any band most AV amps and such dont like it. if you have a monster high current power amp like a krell it is ok to have a 2 ohm speaker.

the good thing about 2.5 ways is that the second woofer is completely out fo teh circuit by the time the tweeter comes in so the impedance is usually about 4 ohms.

yes 2 woofers in series is not a nice thing to do. the inductance of the voice coils combines to create a large inductance hence it will sound like a woofer with double voice coil inductance. however these woofer are working in a narrrow band (below 300Hz or so) so i dont think it would be much of a bother. the only disadvatnage is that uour inductor for these 2 woofers (in series) will be double ( think I mentioned this before). I would recomend trying inductors in the 2-4mh range. the super huge inductors (10mh or so) others recomend are expensive. besides i think that too much BSC is not good. 4 db is about right. 6 db is too much.

lastly when you go MTM you wold have to add a BSC ckt to both woofers. I dont like this. why? well I feel the BSC ckt usually done usig a large inductor and a resistor damps the dynamics of music. now the most critical area of music in 300-3k. the ear is hyper sensitive to changes in this area. by having a BSC ckt that would be is in ckt from 100/200Hz to 1.5/2k the BSC ckt might deterioate the qualtity of the sound. assuming you have 2 86db midbass the combined sens would be 86+6 = 92db. then you add the BSC ckt and loose 4db so you are back to 88db. Unless you like the MTM cylindircal soundwave I would prefer a 2.5 way.


in a 2.5 way the large inductor is only in series with the woofer not the midbass hence the large inductor is not in ckt in the critical part of the music band. now consider a 2.5 way ckt using an 86db midbass and 86 db woofer. the 86 db is a nominal sens. if you look at teh sens curve you will notice that in midband the sens and be as high as 88db. since you dont have BSCk ckt in the midband this is what you get: 88db. the 86db woofer then brings up the sens in the lower bass (below 300Hz) so that the total system sens is 88db. Yes ithe woofer has a large inductor and will loose about 1db in sens due to this but there is enough available sens to compensate. the diffcult part of 2.5 wyas is integrating teh woofer to the midbass. you will have to tinker withvalue and DCR of the series inductor used for the woofer.
 
Your approach seems like a good one.
I hadn' t really thought about the 'loss of dynamics' with BSC - I guess the larger inductor would impare the sound some how.
I was going on a simpler approach (2 way) but I guess a 2.5way isnt that much more complicated, and will yield the same sensitivity.
 
Well guys, as much as I appreciate all those efforts you put trying to help me, I regret to inform you, that your level is quite high for me... In clear, I don't get much of what you are saying...

For example, I always taught that a 2.5 Way simply meant that the woofer was doubled, henced no real deal other than two woofer and one tweeter configured in a "two way" solution either MTM or TMM. But now your starting to talk about some Baffle Step Circuit which would attenuate one of the two driver but not the other? I'm really comfused.

Since I have enough MDF, I do not mind fabricating two or three different enclosures and to really try out different possibilities. But for the cross over, I'm really baffled (pun intended). I would need something more tangible so as to get started on the design of the required enclosures.

Usekgb> Since you seem to have been sucessful with your design, would you mind sharing more deatails with me (us?) about the enclosure, the shape, the cross-over etc...

AP> I do like your simple idea mentioned previously, and I hadn't seen the cross over as a band splitter before, this really opened my eyes! Thanks.

Navin> I really dig your input, but it is rather hard for an uninitiated like me to get a hold of what all this means in practical terms. Perhaps a little drawing would help to ease my pain.

Thanks to all, I can't wait to start working on these speakers!
Sébastien
 
drawing...ok i dont know how to draw using a computer. i can draw by hand and fax it to you. oddly enough i manage (the operations of) a software CRM company.

i dont have auto cad or corel draw.

i will try to put in words and maybe someone else can help...

1. consider your speaker in 2 parts.

a) 1st part is a 2 way using a 18cm vifa uint that we will call teh midbass and a 1" tweeter. this part can employ a simple 2 way XO either of parallel type or series type. A simple search of series and parallel Loudspeaker crossovers will give oy a lot of information. Series XO require drivers to be better behaved which the vifa using are. in fact the vifa unit are so well behaved that a simple XO using a 0.5mh inductor and 4.7uf or 6.8uf cap can suffice as a initial testing point.

b) the second part consists of 2 18cm vifa units we dub as the woofers wired in series (against wired in parallel). If you see your woofer you will see 2 terminals one red and one black. when you connect in parallel the reds of both woofers are connected together and so are the blacks. when you connect them in series the red of one unit is connected to the black of the other unit.

one of these 2 woofers is mounted on the rear of the sealed box containing the midbass. the magnets of the 2 woofers are connected using a wooden or plastic tube and some blu tak or silly putty.

the second wofoer is mounted in a cabinet below that of the 6" 2 way but the cabinet is tuned to bass reflex instead of sealed.

the 2 woofers are still electrically connected to each other although they are in 2 different boxes.

now i would need a drawing. email me in private so i can fax you what i am trying to say.
 
As for a picture this is my guess....

I think having the additional midbass at the rear of the box may complicate things - are they wired in reverse polarity?
If so what affect does that have on the front firing upper driver?
 

Attachments

  • monitor2.jpg
    monitor2.jpg
    1.5 KB · Views: 441
As for the 2 way vs 2.5 way see attached piccy..

The reason for the difference in a 2.5way is the lower woofer is only doing bass (ok some lower mid as well) by placing it closer to the floor results in a smoother bass reponse due to the increasing of the floor bounce frequency.
(this occurs due to the interaction of direct (from the woofer) & reflected (from the floor) sound having a partial destructive interference causing a dip in response.)

In a dual woofer 2 way the MTM is prefered so that each driver is the same distance from the tweeter.
 

Attachments

  • monitor3.jpg
    monitor3.jpg
    2.1 KB · Views: 523
Thanks a lot you guys for clarifying this up! I really appreciate it. And far from turning you down, I think that I'll do two boxes. One will be a simple MTM arrangement with a two way crossover.
The other will be a TMW with a 2.5W crossover. I finally came down to my senses, going with that third extra woofer would actually be too much for a first set of speakers. I've ordered some extra drivers to test this other stuff that was suggested later on.

The precise drivers beeing these:
VIFA TC26T50-06
http://www.partsexpress.com/pe/show...St3=-34511624&DS_ID=3&Product_ID=118216&DID=7
VIFA TC18WG-15-08
http://www.partsexpress.com/pe/show...St3=-34511624&DS_ID=3&Product_ID=121783&DID=7

I modeled them in BassBox and WinISD and both gave me some pretty huge boxes. 140L is not really acceptable for my soon to be two piece appartment, I'll probably be needing that space to sleep. So it's either the speakers or a bed. I guess I wouldn't need a bed, after I build those speakers.

I've tweaked the parameters a bit and found something more acceptable. In fact I found three more acceptable solutions.

-113L box, tuned at 34Hz, F3 at 30Hz, 0.55db of boost at 50Hz
-85L box, tuned at 36Hz, F3 at 33Hz, 1.01db of boost at 65Hz
-70L box, tuned at 39Hz, F3 at 36Hz, 1.62db of boost at 65Hz

Altough that boost might not be unpleasant (I do listen to a lot of techno music), I could always cut it later on with some eq.

My first attemp would definitely be the 2.5way TMW arrangement. Every driver in the ported box, no separations, braces for the mid and woofer and one cross vertical brace. I don't know which box to choose from the three precendent ones, any suggestions?
In WinISD, I simulated the box dimensions and got something rather weird, can I play with those dimensions quite freely or would it create too much problems? (i.e. would the 113L box be ok designed such that the internal dimensions are 9x18.25x42 inches (WxDxH) + 10% for braces and port? )

For the crossover, I was thinking about something along the line of what can be found here:
http://www.soundlabsgroup.com.au/lcaudio/lc_audio_crossovers.htm

Would that sound right? (of course values would probably need to be adjusted)

Well thanks!
Sébastien
 
What exactly would that do? I mean the .qts is of 0.52, is that way too high to allow it to be used in a ported box?

I tried modelling a few sealed boxes and the F3 get up to 55-59Hz depending on the size of the box. As one of my main objective is not to need a sub with these speakers, this F3 is way to high to my tastes... What would be your suggestion to counter this?

Thanks,
Sébastien
 
Exactly - I get about 35l per driver.
As for dimensions the best in minimising the internal resonances are the 'golden ratio or golden section' which is something like 0.618, 1, 1.618 . Try a google search for more info.
- so that might be width of 0.6, depth of 1 & height of 1.6,
I often go for an aesthetic solution based on the diameter of the driver.
I might do width of 236mm based on 180mm (driver) +36mm (wood) + 10mm x each side - I like to recess baffles & use felt to fill-in.
I strongly recommend multiple shelf braces every 200mm or so.
 
Yes of course I can make a 30L box, I will try that as soon as the drivers come in (should be anyday now).

I'm still wandering on the internet for that golden ratio thing. I've known Tau (or Phi) for a number of years, and never taught it could apply to speaker design :boggled:

Thanks,
Sébastien
 
Did you mean a 60l box? - I get 30l each driver.

As for xover I havent played with series xovers & I dont think one xover will work for all speakers (which is what soundlabsgroup suggest).
I did some preliminary modelling & I find that a 2.5 way has really punishing impedance of 3ohms for most of the bass - I would strongly advise adding a series cap (100mfd) to the mid driver.
I would also go for separate volumes for each driver - this allows you to play with different damping/stuffing per enclosure.
 
>Is a 2.5way MTM impossible?

====

Of course not, just roll off one of the Ms with an inductor. This is no different than a 2.5 TMM on the same baffle, or as a TM on the front with another on the back in a bipolar layout. Of course you lose most/all of the benefits of a D'Appolito's polar response control of the tweeter, but it has the advantage of keeping the driver's acoustic centers closer than a TMM, which is more important IMO. The tweeter's off axis response can be controlled with strips of foam or felt.

====

>The Qts of the Vifa woofer is 0.5+. I would not recomend a bass reflex box for this.

====

Why not? Regardless, all it means is that Vb will be large WRT Vas. A T/S max flat dual driver design needs to be ~4.59ft^3 (130L)/32.8Hz Fb minimum. In a ML-TL sim it does this ruler flat at ~96dB/2.83V sensitivity. Really, unless it's well away from any boundaries in a fairly large room it will sound underdamped (boomy), great for movies, not so great for music.

Maintaining the same height and vent, but lowering its cross sectional area to a 3.69ft^3 (104.5L) Vb maintains a ruler flat alignment to ~35Hz at ~97dB/2.83V sensitivity and has the benefit of being able to be rolled off to better blend with the room without any Fb peaking by just lengthening the vent.

====

>0.5+ Qts, wouldn't it be good for a TL?

====

Or my preference, a ML-TL.

GM
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.