tell me where I'm wrong.

[raypalmer]

Okay so my Mk.1 speaker was a bit of a bust. But that's why pencils have erasers right?


I've decided on a bookshelf for Mk.2, my original notion was to clone the Totem Model One but have decided to try something else.

I'm going to extricate my drivers from the Mk.1 attempt as I feel they're pretty good kit. I'm dealing with:

1 Peerless 830657 mid-woofer per speaker
1 Peerless 812978 textile tweeter per speaker

So a 2-way. For xover. I'll probably go with a first order butterworth straight, no chaser. Adhering to my "lite is rite" design ethic. Although I'm wide open to any and all opinions on xo design as well.


Here's my preliminary design. A more concise draft to follow...

An externally hosted image should be here but it was not working when we last tested it.

I admit it's a pretty ambitious little TL but I have high hopes. Let me know why it won't work!

 

[fwater]

It will certainly work, to what level of performance is to be seen. How did the geometry of the TL come to life?

 

[leadbelly]

1) IMHO 1st order crossovers are a bad choice for those drivers. 1st is really only appropriate for extended range drivers. I think you will be hard pressed to find a spot where either the tweeter isn't over excursion or the woofer isn't beaming under power.

2) The cross section on the TL looks wrong. Are the sides flat? How did you calculate?

 

[raypalmer]

It's mostly the product of a creative eye. I'm lacking the software and/or formulae necessary to work it out proper. Although I think I may need to narrow the channel more, at the ratios shown I don't think there'll be enough resistance. Only problem is that I will definitely need a baffle-step if I do that, and I hate to(am unable to) work those out.

 

[raypalmer]

1) IMHO 1st order crossovers are a bad choice for those drivers. 1st is really only appropriate for extended range drivers. I think you will be hard pressed to find a spot where either the tweeter isn't over excursion or the woofer isn't beaming under power.

2) The cross section on the TL looks wrong. Are the sides flat? How did you calculate?

Good! I need xover ideas.

The sides are flat, and the geometry needs a tweak. Like I say, it was worked out rather ad hoc.

 

[Cloth Ears]

I admit it's a pretty ambitious little TL but I have high hopes. Let me know why it won't work!

I'll have a go, but I won't tell you why it won't work, just why it may not work.

I'm assuming that you were looking at a TL tuning of around 70-80Hz, as it appears that is the length of your quarter wave (ie. about 1.15 metres).

Generally, TLs are tapered pipes - either expanding (in the case of TQWT) or contracting (in the case of a more 'normal' TL). Yours has a large (expanding) driver enclosure followed by a 65cm(?) constant area tube. I would suspect that this might act more like a bass reflex enclosure than a transmission line. How did you model the design?

The design is also wasteful of space. All that sand is not going to really have much effect except after a few weeks of use. It will have settled by then and will start to 'rattle' a bit unless you keep it topped up. If you want it heavier, why not try thicker pieces of wood in the cabinet sides?

Without working out the measurements, have you thought of something more like the attached picture?

 

[raypalmer]

Well I must say I like the look of that. Would I have to increase the length of the transmission path to tune closer to 60-55Hz?

 

[fwater]

One place I can give advice is the XO. 1st order won't cut it. Crossed high enough to prevent distortion or damage, the tweeter will never be able to mate up with the woofer. Crossed low enough to avoid beaming and any breakup modes, likewise for the low end. A carefully planned out 2nd order for the woofer can build in impedance compensation and BSC. A 2nd order for the tweeter will be easier. If you can get me the .frd files (SPL trace, or if somone already has them), I'll take a shot at the XO and let the more experienced guys critique it...

 

[planet10]

cloth ears TL makes more sense.... to do a proper TL you need to at least plug the driver parameters into Martin's design tables.

dave

 

[raypalmer]

Fair enough... Where are the tables?



fwater; The TS Parameters are:
http://www.solen.ca/pdf/peerless/830657.pdf

and

http://www.solen.ca/pdf/peerless/811978.pdf

Your input is greatly appreciated mate.

 

[fwater]

Alright, there will be a bit of work involved whipping something up. Post the width of the baffle you're going for, and give me 24 hours.

Pleased to be of help in the same selfless way it was given to me here long ago (and still is).

 

[raypalmer]

Alright, there will be a bit of work involved whipping something up. Post the width of the baffle you're going for, and give me 24 hours.

Pleased to be of help in the same selfless way it was given to me here long ago (and still is).

fwater you're a gentleman and a scholar and there aren't many of us left 22cm is the number.

 

[raypalmer]

What are the common numeric values for stuffing density and speed of sound in stuffing??

 

[Moondog55]

Ignore speed of sound; these days it is considered irrelevant, damping is to absorb HF, volume and length are important factors.
I just used up all of my knowledge on TL's so I hope some-one else can assist further.
Regards
Ted

 

[planet10]

Fair enough... Where are the tables?

http://www.quarter-wave.com/TLs/TL_Theory.html

1st article in the unnumbered section...

the whole site is worth a read, as is the t-linespeakers site URL below.

dave

 

[raypalmer]

My (defunct apparently) TL seems like a fail. Any thoughts on this design? WinISD seems to think it will work.

EDIT: I put some time into this design a while ago. But I became muddled about whether this Isobaric-ish ported speaker would really work. Obviously there will be 10-15cm standoffs from the ground for the floor-facing driver. I know neat makes something like this but their's is strictly isobaric to my understanding.

 

[planet10]

WinISD doesn't take into account the tall aspect ratio which causes a TL effect -- if the port was at the bottom you would have mass-loaed TL (althou not nescessarily be tuned properly)

dave

 

[raypalmer]

I thought the phase-inverted driver in the base would alleviate that?

 

[planet10]

I thought the phase-inverted driver in the base would alleviate that?

Didn't even see that... if the phasre is inverted you'll just get (mostly) no bass.

Just be patient ans see what fwater comes back with...

dave

 

[Cloth Ears]

My (defunct apparently) TL seems like a fail. Any thoughts on this design? WinISD seems to think it will work.

EDIT: I put some time into this design a while ago. But I became muddled about whether this Isobaric-ish ported speaker would really work. Obviously there will be 10-15cm standoffs from the ground for the floor-facing driver. I know neat makes something like this but their's is strictly isobaric to my understanding.

Don't give up on the TL. I modelled what you had (which looked like a 10 litre box with a 36cm long port of 5x20cm dimensions. It had a big peak at 115Hz. But you may be able to do much better in almost the same sized box with a TL. I agree with p10, wait for fwater to do the 'shining knight' thing.

The other design isn't Isobaric (as far as I can see), it's just a dual woofer bass reflex. Isobaric is the type of mounting shown in my primitive attached drawing. For the box you have designed, you would probably need a 100 litre enclosure (and maybe a different sized port). Currently it would have a small peak at 110Hz and be "rolling off" (amplitude would be reducing by) at about 12dB/octave. And, you would probably need a bigger port to handle anything over 5W of input power.

Again, don't give up!