Starting my first project.

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

i have been a mute spectator and been reading the posts by others for the most part of my membership in this forum.
i decided to finally go ahead and build (or Replace) my first speakers.
I plan to build these speakers as an experiment to test whether i have the inclination and the knowledge to build my own speakers. so, i don't mind bombing out on my first try(obviously i'd be diappointed if that happened). But, i plan to make this a hobby and gain knowledge from the experience and continue to build more.

OK, here are the drivers that i have bought.

LF Driver-

i know that for my first project a 3-way may be too much work but i don't mind cause from June 4th onwards i am pretty much free for 3 whole months. with the help of this forum i'm confident that i will be able to suceed in getting decent levels of SQ.

here is the problem.
i presently have a pair of old(30 years old) 3-way 4 speaker system sealed monitors. the cabinet is made of teak and filled with glass wool.

as i see it i have 3 possibilites.
1> Replace the existing drivers with the new ones that i have bought and concentrate all effort on designign a new crossover for the 3 way and keep it sealed.
2>ditto,except make it a vented enclosure.

3>design and build a complete speaker system from the ground up ,including the enclosure and decide as to sealed or ported along the way.

the dimensions of the old speakers is 20'x9'x11.5'(LxDxW).
The speakers are shown in the picture that i have attached. they are heavier than the wharfedale diamonds that are next to them.
Looking forward to ur views.

p.s am having the drivers flown in from US and will have them by the 1st week of June.


  • img_0044.jpg
    50.2 KB · Views: 1,621
You can calculate the internal volume of your old cabinets and use a simulation program to find out what the response will look like, but really, it's better to build new cabs than to re-use existing ones and match them to drivers.

Either way, you will need to design your own crossover for your new drivers (good choices, BTW) and that can be really tough.

Madisound has a system whereby you pay them something like $50 and they design the crossover for you, using their special simulation software. Then all you have to do is actually assemble the box and crossover.
Hey Nappy,
the volume of the enclosure is about 1.4 cubic feet.
Yeah, i will draw up some plots for the different configs before deciding.
anyone who can help me in this regard??
am using WinIsd.
Regarding getting crossovers designed, i think it would be better if i try to design it myself so i can gain some knowledge(even if it means making mistakes and more importantly learn from others)

Vivek, the towers are Wharfedale Diamond 8.4s.

In that case, when you build your new cabinets, you should have a binding post and wires going straight to each speaker--that way, you can have your crossover outside the cabinet, and if you decide you want to try a different design, you can do so without having to invade the box.

This also gives you the option of going with an active crossover and bi- or tri-amping, which is very desirable.
okay, i have decided to build a new cabinet and go in for a passive crossover first.
as u rightly point out, i would like to keep the crossover outside and expweriment till i find the sound to my liking. but due to some reasons i will not be able to indulge in these apeakers for a long time as i will be leaving my hoemtown later this year.
hence, i think it would be better if i finish it once and for all with a passive crossover and get it over with. since this is going to be my first experience i donot want to over complicate things and leave things hanging. i want to be able to fire them up before i leave my hoimetown.

Btw, am having problems with partsexpress. it seems the Vifa midrange and the Daytons have been back ordered.only hope they make it in time so that they reach my friend before he leaves the US.
Can anyone with previous experience with parts express tell me if it is possible to change the shipping address and also does anyone know how long they take to get the drivers out on the road and give me a tracking code when they say a item is backordered.
i don't get it. they advertise that the product is in -stock. why is it back ordered?

Got the Drivers

Hello all,
I just got word that the drivers were delivered to my friend's place.
now that that is done. i am expecting the drivers to be here in India by the 1st week of June.
Now i have some basic design questions which i would like to get clarified.

I have been using the Winisd for designing a vented enclosure with the following specs.
the f3=27.15Hz
fb=27.45Hz and the Vb=2.7ft^3.
the group delay is abt 23.24ms (but the freq is off the f3 by 1 Hz,26.15Hz).
the vent dimensions are as follows.
and vent mach=0.09

i have the following questions.
1> what is the relevance of group delay and the criteria for vented and closed.

2> i listten to all genres o music but i need my speakers to play Rock and metal well. so i guess the transients should be good. How do i obtain this.

3>Does a flat response mean a well designed speaker? what happens if there is a bump in the freq response and the response exceeds the reference level(0db) , call it a bump if u may.

4> considering the driver that i have bought and the kind of music i would listen to would a sealed enclosure or a vented enclosure be in order.

thanks for ur patience and looking forward to ur guidance.
well i have been doing some reading of the archives regarding the group delay and there seems to be a consensus that the group delay that i have reported is okay.
i am presently leaning towards the vented enclosure.
also my friend has promised to help me calculate the driver parameters using MLSSA when he gets here. also i will be re-checking my design with the Leap software that he has and using the new parameters as measured by MLSSA to re-design the enclosure.
in light of this should i defer my designs till i measure the parameters ?

i want to start work on the crossover soon , but first want some assurances from the more experienced members of this forum that i am on the right track regarding the present design that i have posted. any modifications that i will have to make?

more importantly, i want to know if the woofer is good enough to work in a enclosure such as this.(since it has a linear Xmax of only 4.5mm).


p.s teh box dimensions according to winisd(not optimum) is L=32.98inches
Thank you Nappy.
Can u please suggest what i should do regarding the T/S parameters testing using Mlssa. My friend says that he has the test equipment. a pink noise generator and a signal generator.he said he could burn them in and test their parameters for me.
how many hours of burn in should i ask him to do before testing the parameters.
or is this design good enough and should i forget abt measuring the t/s parameters.
btw the LEAP that he has is leap 3. it is a DOS version.
Can u give me more insights as to what kind of crossover frequencies i should be looking at and also the order and type for a 3-way speaker system.
thanks for the sustained support.
How long to burn in the drivers before measuring? That's a complex question.

The longer you burn your drivers in, the less the parameters change with more burning in; but they never totally stop changing. (It's like asking how long it takes a note on a piano to decay... it never decays fully, but it will fall below your noisefloor eventually.) You might try measuring the drivers *right* when you get them, then measure after every hour of burning in, and decide when they've burned in long enough.

Please understand that your box volume will be off by some amount; maybe 1%, maybe 10%. Most of the time, no one worries about it very much, because most of the time, in-room response contributes more error to the speakers response curve than errors in the box.

In short: shoot the engineer and build the damned thing. :smash:
Oh yeah... you asked about crossovers.

This is the hardest part about designing a speaker, IMO. Building boxes is easy; buying parts is easy; wiring things is easy. But actually designing a passive crossover... it's almost as much art as science.

For those drivers, I think crossover values of 500hz and 5000hz would be about ideal. You might cross over to the tweeter higher, or to the woofer lower; everyone really does it their own way.

For a first project, I would really, really recommend that if you have the money, you buy a nice active crossover (the Behringer DCX2496 gets a lot of good press) and amps to drive the speakers. It is sooooo much easier to just pick a crossover frequency on the box, than to pick values and calculate impedances and such in order to design the right crossover for your speakers.

There are also people out there who will design the passive crossover for you if you pay them money. While it costs money just to get the design, it's still a lot less expensive than going fully active. The drawback is, it's still not as good as active, and you don't have the option of tweaking crossover frequencies later. But it is a lot less expensive.

If you are very knowledgeable with electronics modeling software, like SPICE, you might do better to design a passive crossover yourself...
Hey Nappy,
as much as i would love to go the active way am in no financial condition to spend 500-600$ for an active system. for now atleast, i will go the passive way. I was doing a bit of reading and it does seem a bit daunting and will demand that i get back to my basics and do a lot more calculations and circuit optimisations.
btw, i was wondering if i will need an Lpad attenuator for the tweeter.
the sensitivty of the tweeter =90dB: midrange=88dB: Woofer=89dB.
will an L-pad attenuator be need for the woofer too to get an overall sensitivty of 88dB or should i consider something else.
will a 12db/octave be enough for both the high pass and low pass crossovers?
i am yet to decide between a butterworth or bessel filter.
I can see taht even this passive crossover is going to cost me quite a bundle but will still be a lot more cheaper than an active system.
i do have some skills in Spice and Orcad but limited.

also will Mlssa be able to give me the actual nominal impedance and the inductance of the voice coil? if not then how do i go abt determining this? Do i need to calculate the nominal impedance of the woofer too or will the midrange and tweeter's be enough?

coming to the box. can i substitute an oval shaped port instead of a circular port.correct me if i am wrong, cause i understand that an oval port helps air flow better.
also what is the relation for spped of sound in mach number to m/s.

Hey can anyone give me links or info abt amplifiers for an active system ..say about 35 watts a piece for the midrange and the tweeter and one amp for the woofer rated at abt 50-60 watts.
are their any affordable plate amps in the power range that i am seeking or even amp modules that can be directly placed inside the cabinet...
i still have'nt abbandoned the idea of an active system..

another thing is i was fooling arnd with the vent dimesions on WinIsd and i came up with a two port design with
the port length=6 inches
diameter=2 inches
the thing is that the vent mach dropped from being .09(30 m/s) from the earlier design to .05(17 m/s)
i know that a high mach number is bad but is there such a thing as a low mach number that is bad?

appreciate the help.

Lower mach is better, universally.

For amplifier stuff for the inside of cabinets, check out the "Chip Amps" forum on this same site.

About the L-pad thing... there's a good chance that your woofer will lose 4-6 dB of sensitivity due to the "baffle step". Download a program called "The Edge" for simulating baffle step and figure out exactly how much it will affect you. If you still want to go passive, you could build your cabinet with a narrow baffle so that the baffle step occurs above your crossover frequency; then you will only need an L-pad on the midrange and tweeter to match levels. (Using an L-pad on a woofer is considered a bad idea.)

If you go active, then no L-pads are required. Just change the gain of the amplifier to match sensitivities...
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.