MTM TL speaker project questions

[TLS Guy]

Taking apart and packing your audio system must be a record-breaking pain in the butt.

Yes, it will be a lot of work.

 

[Swerd]

Book is on order. Thanks!

Good. I hope you find it interesting and useful. This book is good at explaining how woofers and cabinets interact at the lower frequencies. It's OK about explaining some of the basics about crossovers, but it doesn't go into enough detail on that subject.

Once you understand designing a speaker cabinet, you'll begin to realize that the overall design process has to follow this general order:

1. Come up with a overall design goal, such as a 2-way floor-standing MTM speaker.
2. Select woofers based on their T/S parameters, maximum usable upper frequency, and sensitivity. Obviously there's much more to woofer selection, such as overall sound quality.
3. Design a suitable cabinet based on the woofers' T/S parameters. You can use the values provided by the manufacturer, but it's better to directly measure the woofers you have. Build a test cabinet, install woofers, and measure their loudness vs. frequency, and impedance vs. frequency. This set of data will later be directly used by software to design the crossover. Also measure the off-axis performance at the woofer's upper usable frequencies. At what frequency is the woofers' 30° off-axis performance within 3 dB of its on-axis performance? This will be the maximum usable upper frequency of the woofers. This is essential to know before you select a tweeter. Don't be disappointed to learn that it's a lot lower than the manufacturer's published upper frequency limit of a woofer.
4. Select a tweeter based on the maximum usable upper frequency of the woofers, and the woofers' sensitivity. Generally the lower in frequency a tweeter can go without distortion (its resonance frequency, Fs),the greater the cost. A good rule of thumb for tweeter selection says that for a crossover with a 2nd order slope (12 dB/octave),use a crossover frequency 4× higher than the Fs of the tweeter. A crossover frequency of 2,000 Hz would call for a tweeter with an Fs of 500 Hz. For a 4th order slope (24 dB/octave),cross over at 2× the Fs of the tweeter. That crossover frequency of 2,000 Hz would call for a tweeter with an Fs of 1,000 Hz. There is no reason to pay for a tweeter that can go low if you plan to cross it over higher. As in woofer selection, there are complex decisions about tweeter sound quality.
5. Start designing the crossover. As TLS Guy had said, computer-aided crossover design software can do about 80-90% of the design process for you. The rest you have to do by listening, testing, trial & error, and educated guesswork. If you have the high-priced design software, you can simulate numerous crossover variations by creating virtual crossovers in the software while listening to the results with your test speaker. Otherwise, you have to build a test crossover with real parts and monkey with the components. If you have Dennis Murphy's experience and judgement this can take half an hour to an afternoon. The rest of us need quite a lot longer.

I went through this exercise to make it clear to you that going backwards from an existing cabinet is so hard as to be impossible.

 

[Swerd]

With your main speakers which are basically built for that specific room, I guess time for new ones?

You underestimate the good doctor. TLS Guy designed the theater room in his new house around his custom made speaker system .

 

[KenM10759]

Good. I hope you find it interesting and useful. This book is good at explaining how woofers and cabinets interact at the lower frequencies. It's OK about explaining some of the basics about crossovers, but it doesn't go into enough detail on that subject.

Once you understand designing a speaker cabinet, you'll begin to realize that the overall design process has to follow this general order:

<snip>

I went through this exercise to make it clear to you that going backwards from an existing cabinet is so hard as to be impossible.

OK, you've convinced me I'm wasting my time. I'll go through the book and decide if I want to continue.

 

[Swerd]

OK, you've convinced me I'm wasting my time. I'll go through the book and decide if I want to continue.

Late last night as I wrote my long design post, I forgot to mention baffle step compensation (BSC) and tweeter diffraction. Woofers and tweeters mounted in a speaker cabinet actually perform different from the frequency response curves published their manufacturer. The cabinet's front baffle width directly affects both of these performance aspects, and both must be known before you design the crossover. In DIY, you can actually measure their responses in your test cabinet and see what must be done to correct things. If needed, you can build a new test cabinet with a different front baffle width. Most DIY builders I know simply choose the narrowest baffle width possible that allows mounting the woofers they chose. For example, a 6" wide woofer frame allows an 8" wide front baffle.

Go see the various topics on Salk's Blog Topics Page. Definitely read these:

BSC Made Simple by Jim Salk​

Crossovers 101 by Dennis Murphy​

Understanding Cabinet Diffraction by Jeff Bagby​

 

[Swerd]

@ryanosaur – I see you're reading this thread. When I first got into DIY speaker building, I had eagerly planned to design and build my own. I wanted to get all the measurement and design hardware and software. In the DC/Maryland/Virginia area there was group of DIY speaker builders who usually met once a year. That's how I first met Dennis Murphy. He often hosted these meetings at his house, and was treated by most as the honored & revered Grand Master. More than a few other DIYers told me, "Do yourself a big favor and just build what Dennis designs. You can spend lots of money on a design & test rig, spend lots of time learning how best to use it, but you'll never do better than what he designs." I took that advice and never regretted it.

Since then, Dennis designed most of Jim Salk's speakers, and has launched his own company Philharmonic Audio. By now, you are getting familiar with how similar sounding Dennis's speakers can be, especially in the mid range. I think its fun to understand what makes good speakers sound good – as well as what doesn't matter. But I no longer think I can design anything that does it better than what Dennis Murphy has already designed.

Now is the time to state the usual disclaimers… What do you expect from an acknowledged DM Fan Boy? YMMV, yadda-yadda, etc.

 

[ryanosaur]

I see you're reading this thread.

You bet! I'm truly fascinated by the topic. As well, the prospect of learning how to do this is kind of exciting. I know I could never 'build' my own saxophone, for example... but this is well within the realm of possibility.

I'm grateful for the knowledge that's been shared with me so far! ...And eager for more!

 

[KenM10759]

Reviving an old thread, but not the original idea.

I was listening to these speakers last night and just had to turn them off. They actually now sound quite horrible. The woofers in the right hand one, or one of them, is now failing. It makes a weird "flatulent" kind of sound with anything below about 100hz, as though the voice coil is hitting something.

If I pull these woofers and get whatever numbers I can off them, maybe measure things like the bolt pattern, flange diameters, depths, and whatever else I can get, such as resistance, is there any hope of getting a working replacement that's at least close? I don't wish to try and re-engineer a great speaker out of an attractive but poorly-executed cabinet, but if I could at least get rid of the objectionable noises I'd be happy.

I also found that the tweeters both have a fair bit of distortion, sound quite dull, and have a noticeable amount of metal dust/chips around the base of the silk dome. The magnet acting upon them. I believe I have found the exact same tweeter on Parts Express. The original is labeled as Vifa, but if I recall correctly they're part of the Peerless/Tympany company now. These measure and look EXACTLY like what's in the speakers:

https://www.parts-express.com/peerless-by-tymphany-h26tg45-06-1-silk-dome-tweeter-with-waveguide-6-ohm--264-1386

I wouldn't cry about trying them for $60 unless someone declares this a stupid move.