What's a 3.5 Way Speaker? Designing My Main Speakers 

I've missed something, 400hz to 1000hz is only just over 1 octave. 10hz to 400hz is 6ish octaves, 1000hz to 3000hz is 3 and 3000hz to 20,00 is 4. Edit, Sentence deleted... I missed something didn't I?

OK, I re-watched, my question would be, why not run the 8" down to 150hz? ish I've done a fair amount of work with PA and 150hz is a common choice for crossover and it means a lot of attenuation of breakup and 1kz resonance that is common in 15" drivers, in fact EV used to put in a deep filter to get rid of it. Edited due to dyslexia. Not only can I not spell but I can't see the words I missed, or should that be, I can't see I've missed words..?

Having a 1k crossover puts it in the key vocal range. If the midrange will handle 500hz then that might be a better option and push the lower xover down as well as was mentioned

And it IS a problem! Like an addiction...

@@IBuildIt Not a bad one.

@@IBuildIt I feel that pain. I've always enjoyed making things more than actually using them xD

Ultratweeter sounds like someone with over a million followers on Twitter! Lol

I would think you are looking to torture someone with that much tweeter

Unfortunately, I don't have the space for horns. Or at least not the kind of horns I'd like to build.

The 8" has a cone diameter of ~17cm, so that's 34300/17=2017Hz. The 15" has a cone diameter of around 32cm, so the beaming frequency for that one would be 1kHz. I've done some preliminary testing on the 15", and it doesn't look bad. Distortion starts to creep up below 100Hz when it goes over ~90db, but nothing that would stop me from using it.

@@IBuildIt You are not considering 1/2 wavelength and how you have a reflection across the cone surround internal diameter . Your math is solid. 1/2 wavelength is where you begin to generate reflections that create the nulls that make for the loss of hemispherical output. That 15 is not that bad. RadioShack had some decent drivers. It may have been made by Tongen in Japan that far back. I had some 15 inch driver from RadioShack that had 11mm of Xmax in the early 90's. Nobody was making them like that back then. Same look, same tiny magnet. Had a big old plastic cap to make it look impressive. Foam surround though.

I see your point, but I take the full wavelength as the absolute limit, and crossing below that will give acceptable results. So, crossing the midwoofer to the midrange at around 1kHz would be well below that absolute limit. After all, most 2-ways are well beyond the 1/2 wavelength for their crossover point. BUT, I'm just getting started and there's nothing that's written in stone, yet. I want to get the individual drivers measured today and that'll give me something more concrete to work with.

@@IBuildIt I truly believe you have a good handle on what you are doing. What I have learned in studying and listening to many loudspeakers is that many loudspeaker designers are woefully contradicting basic good design principles. The few that really are working within the physics that we are presented with are the ones that make the magic. A little tidbit of information is that having a wide horizontal and a wide vertical dispersion is not always advantageous. Floor and ceiling bounce of the tweeter output is not your friend. You know the range that is most critical for spatiality. Getting secondary bounce effects is including your rooms reflections in a manner that they are louder than the reflective cues in the recording. You loose the best effects of the available soundstage in the recording. Your room treatment helps a lot. But the floor bounce is nearly impossible to mitigate the same way. A planar is one way to lick that problem easily. The vertical spread is usually less than 10 degrees. Food for thought buddy.

Well there's a lot of leeway in speaker design and some will say that beaming is good for improved directionality and less room interaction. It comes down to what the listener likes and the majority of them are in rooms that are about as far from optimal as possible. I've never used a ribbon or planer so I may order a pair to try out. Any recommendations?

Agreed. I wish I had a mind for remembering titles to musical pieces, but that part of my memory is shit.

@@macedindu829 J.S. Bach-Prelude in C Major.

@@RonaldJS you da real MVP :D

I said all of that in my pinned comment and also put a caption on the screen during the video, so yes it makes sense to me :)

@@IBuildIt Ah that's funny. That will teach me to read the comment after I watch the video! Best of luck on the build!

The better ones are also available under the ScanSpeak brand.

Honestly, I don't think the driver arrangement makes much of a difference, as long as they are close enough together.

The speakers didn't have crossovers and were designed to work with a digital crossover and 6 channel amplifier. I doubt that I could sell them for anything near what the drivers alone are worth.

@@IBuildIt active speaker system.

@@IBuildIt which digital crossover did you use?

Behringer DCX2496

Less is also less.

Speaker building and design is a technical subject.

Low cost doesn't always mean low quality, and that's especially true for bass drivers. The measurements I'll make (including distortion) will show exactly how good the woofer is and whether it's suitable for my needs.

I will add... that I used to buy many used speakers from the thrift shops. Among these were a few Radio Shack speakers. Id swap the RS drivers into other speakers, to compare the differences... and they were unbearable bad, compared to the original drivers. I also once tried to use a lower quality woofer to repair my foam rotted Phillips woox speakers, such as a SONY woofer. It was a serious downgrade... and so I had a person refoam the original drivers. There was no comparison between the two. The phillips woofers sounded 1000x better. There are a lot of variables that change the accuracy and potentials of these drivers. A large part is the magnet strength. But thats not the only factor. It can be wire gauge and amount of wraps, accurage of the coil wraps, distances between coil to magnet, excursion, accuracy of the mounting, cone material, cone shape, basket shape and or stiffness, spider flexibility, and more.

@@IBuildIt Let me ask you a question: Do you think a single frequency sweep, is an actual indication of a quality speaker sound? Its easy for a speaker to play a Single noise, at low to medium volumes... without any struggle / distortions. What happens when you play something like Disco... like Earth Wind and Fire? Where there in an entire orchestra worth of instruments all playing at different frequency ranges, and all at the same time? My non Audiophile headphones, for example... would completely distort such music... such as the lower end sounds, distorting the clarity of the highs. Even though that is a single driver comparison... the fact is, that even when you have a +3 way speaker setup... when you have complex music on each frequency or ... each "Driver"... the individual Drivers level of Control will be TAXED. If the drivers are of poor Precision and Design... then they will distort more quickly and easily, and will cause issues in musical clarity, image, and more. If the driver are highly precise, and very powerful in their control... they will not distort at all... even under maximum volume levels, and very complex dynamic music. So, unless you are going to be taking a songs worth of readings... rather than a Frequency Pitch reading... then you are not actually getting the picture.

You don't seem to understand how measurements work. Distortion is distortion, regardless of the test signal. The sweep just makes it so that it can be shown accurately.

@@IBuildIt The distortion that they frequency sweep measures, does NOT tell the full story, of how speakers actually operate under stress. Thats like comparing an RX-7 Car, that has a body stiffening kit installed in it... to one that does not. Under high G-Force curves.. the one Without the frame stiffening bar, will contort.. and lose traction / performance. That type of "Distortion" isnt being measured, by typical low speed traffic driving situations. My friend drove me in his RX-7 that had such a mod... and I believe he ended up taking a 45 mph curve at over 90mph, and it didnt show ANY sign of traction loss (slipping), nor any body tilting / contortions...etc. That thing was incredible. It also took off like a damn rocket, gluing you to the seat. :) Again... we are talking about testing the speakers every potential issue, under the highest levels of endured stresses. A weaker cone material, for example... might flex quite a bit, at higher volume levels + fast tempos. The more its trying to play at the same time... the more that cone distortion, would factor into the output sound. Yet, when just using a frequency sweep... it shows little to no such distortions. And as stated... if your magnets and coils are not very powerful... then the speaker can not stop and accelerate fast enough, to keep up with the sounds that its being asked to play, with any degree of accuracy. For example... if I tell you to drive 60 mph forwards, and need to do so in less than 3 seconds... and your vehicle isnt capable of getting there that fast... its not being accurate to what I asked of it. If you hit your brakes at 60 mph... and slid 70 ft... instead of instantly stopping on a dime... then again, you "driver" isnt doing what its being asked to do. When the cone cant accelerate fast enough, nor stop instantly and be able to reverse instantly... then you get micro-distortions. This gets worse, the more power you give these drivers... because you now have more forces in play. So you have speed issues, as well as material and construction related issues, that are not going to show up, ULTIL you have taxed them well. Until you have pushed them to realistic levels of usage.