[richidoo]

Florent, welcome to AudioCircle! Thanks for sharing your project with us.

I generally recommend people build a kit for their first DIY speaker for the main reason that they can have guaranteed success on the first try. It is difficult to avoid failure on the first try of any sport. But since you are so committed, hopefully you will persevere through the many challenges and find success. Speaker design is pretty difficult. I have built about 25 speakers in the last decade, and I am only now confident enough to do my own design from scratch and it is still based loosely on someone elses successful design that I know works.

Your drivers are good choice and your sim looks fine. Make sure your drivers are time aligned on Z axis when you are using symmetrical order crossover. You will like the flatter phase response. This may require a tilted baffle, but you can handle that.

Sorry I joined late so I won't jump on the fine advice you already got. But I'll pick up on your latest post.

Digital amps usually have flat or decreasing distortion up to max power. So the sound quality doesn't change much, and old rules about the first watt don't really apply to modern class D amps. Your amps are quadruple what you really need for residential listening, but there's no harm in having extra power. It will allow further DIY speaker experimenting in the future.

The best way to protecting your speakers from damage is using your ears. It will sound frightfully awful when they are in danger of damage. But you have to be paying attention and also willing to turn it down. If less responsible people will use the stereo besides you then automatic protection is prudent, but it's just you then it is not needed. 100dB is very very loud and that falls within the driver power rating. Jazz and other brain music is not usually played loud enough to damage. Even short peaks above the power rating are OK. It's the sustained compressed signal over power rating that gradually overheats or melts the voice coil, or less likely, the full scale transient pop at high gain that can damage the suspension. A little care and thought will prevent accidents.  ymmv   Yes you can put a fuse anywhere you want. Usually fuse is put before power supply. It adds resistance so between the PS and amp not ideal.

There is nothing you can do to prevent baffle step loss, except mount the drivers flat into the wall of your room, or make them a bipole speaker with identical woofers on the front AND back sides of the box. Aside from those options, the sound will diffract on the front baffle edges and the diffraction will attenuate the LF on-axis SPL. So you have to equalize the response to compensate. In a passive crossover speaker the compensation is made by attenuating the frequencies above the baffle step freq. In an active speaker you can boost the freqs below the baffle step. It looks like 6dB first order, but it is actually shallower than 1st order. You probably only need 3-4dB of compensation (instead of the full 6dB) since the diffracted LF energy is still in the room and is reflected back to your ears. 3-4 usually sounds best, depending on the size of the room.

Rounding the edges can smooth out the FR bumps caused by diffraction, but the smoothing effect of rounding is proportional to the radius of the roundover, and the wavelength being smoothed. The smoothing effect begins with radius 1/10 of the wavelength, but if you want a perfectly smooth baffle step you need a large enough spherical enclosure which is highly impractical! https://trueaudio.com/st_diff1.htm  A roundover of 1/2" radius will mostly affect the high treble. If you want audibly smoother upper midrange you need 2-3" radius. As an aside: Spherical speakers are hard to build, but if you ever hear one you'll remember the smooth response they have. The larger sphere the better!  Once at RMAF 2008 a guy brought a 3 foot diameter sphere speaker from Japan. It had a tiny 3" full range driver but it sounded really smooth and natural.

Be sure to post some pictures, especially of your wood shop!!

[Florent]

Hi richidoo !

I guess i didn't take a kit because i wanted to do it myself. Didn't see the point on buying this stuff if i can make it with scrap parts from my shop !
Now, a few weeks and dozens of hours re searshing on the subject later, I think kits are the way to go for noobs like me ! But still, i want to make it myself. Now that i am this far, let's go the extra mile right ? I guess there is a 50/50 chance it won't turn out good...

I bought the parts that i think i will need for the first step, drivers, simple crossover, and the amp with power supply. I will look up for a microphone tomorrow to make measurements as soon as i get one.

First step will be assembling everything and test it ! Looking forward for the soldering 

Second step will try to improve that... And if it reaaaally doesn't work, well, i still can buy a DSP right ?

Pictures of the wood shop is out of the question right now it is a complete MESS !!!   

Got the parts today. Here is where i am at ! Finishing the aesthetics (and gluing the enclosure) will be for when the sound is right ! And my brother will 3D print me a case for the amp and power supply...

[wushuliu]

I bought the parts that i think i will need for the first step, drivers, simple crossover, and the amp with power supply. I will look up for a microphone tomorrow to make measurements as soon as i get one.

First step will be assembling everything and test it ! Looking forward for the soldering 

Second step will try to improve that... And if it reaaaally doesn't work, well, i still can buy a DSP right ?

Pictures of the wood shop is out of the question right now it is a complete MESS !!!   

Got the parts today. Here is where i am at ! Finishing the aesthetics (and gluing the enclosure) will be for when the sound is right ! And my brother will 3D print me a case for the amp and power supply...

Too bad you bought crossover parts before understanding how to design one but I guess that will be a learning lesson for you. I can almost guarantee you will need larger .9-1.25mh inductors for the woofer. However, you do not need to do any soldering until you have it right. Just use alligator clips instead to connect the leads:

[S Clark]

Yep, you build your crossover with alligator jumper wires.  Get a pack from Amazon, because you'll need several. 
Take some time to get used to taking measurements- there's a learning curve.  Danny Richie of GR-Research
may have a video on his approach to measurements.  If so, watch it and learn. 
Like Rich, I've built or supervised lots of speakers--- 50+ pairs.  It's begins to follow a pattern after a while.  You may find some designs using the same woofer.  If so, you might be able to use their crossover as a starting point. 
Good luck.   

[Folsom]

I have bundles of alligator clips dangling off of my audio rack and other places. 

[richidoo]

Wow, those look awesome!!   boiz got skilllllz.... 

The first try of a simulated crossover rarely works perfectly, especially if the simulation is not based on measurements of your actual speakers. After you measure your new sim will show what tweaks are necessary.

The waveguide time aligns the tweeter with the woofer so you're all set on that.

[Florent]

Revised the crossover, SPL looks better, only -2.5db in that 2.3kh dip. But Phase doesn't look gooed ! How can i fix that ?

[JLM]

DSP should only be used as "icing on the cake" or a bandaid, not a fix all solution. 

Good speakers start with good parts (drivers, crossover, cabinet).  There's no getting around a flawed design with DSP. 

Note that Buchardt is using DSP to tune out system minor resonances and diffractions of their new A500 monitor after taking 2700 measurements of each driver, but only after starting with properly paired drivers/cabinet/amps (in the lab, not by guess or by ear). 

[Florent]

Ok, so yesterday i soldered the power input to the amp module, made a cable with wall plug for the power supply, wired everything, took dozens of measurements with my multimeter and finaly hooked up two litle speakers i had in the atick. It works perfect.
Ordered a microhpone, the DAYTON AUDIO UMM-6 USB. I will have it on monday i guess, it comes from a french website.

[wushuliu]

Revised the crossover, SPL looks better, only -2.5db in that 2.3kh dip. But Phase doesn't look gooed ! How can i fix that ?

Revised the crossover based on what? Have you done any of the steps recommended earlier? Until you do those first, what you're simulating isn't going to be much help. Also, those images are too small to see in detail.

[wushuliu]

Below is a first timer thread using the same tweeter but a different 4ohm woofer. It's a good thread. Yes some will be over your head but you'll get an idea of what's involved to do it right and just how much Kirby left out. Notice his tweeter crossover (remember I said you'd want to cross more like 2.5-3k?):

http://techtalk.parts-express.com/forum/tech-talk-forum/1343815-the-super-bees-garage-two-way-build/page7

In my opinion, this tweeter is not a good choice for a first timer. Even Parts Express' kit that uses the tweeter has a lot of parts - and it's in their best interest to use the fewest parts possible.

So another lesson - A lot of cheap woofers and tweeters require more crossover work than the expensive ones. That's why they're cheap.

EDIT: Actually I take that back some, since really it's just a 3rd order electrical/4th order acoustic with an L-pad but still...

[Florent]

Below is a first timer thread using the same tweeter but a different 4ohm woofer. It's a good thread. Yes some will be over your head but you'll get an idea of what's involved to do it right and just how much Kirby left out. Notice his tweeter crossover (remember I said you'd want to cross more like 2.5-3k?):

http://techtalk.parts-express.com/forum/tech-talk-forum/1343815-the-super-bees-garage-two-way-build/page7

In my opinion, this tweeter is not a good choice for a first timer. Even Parts Express' kit that uses the tweeter has a lot of parts - and it's in their best interest to use the fewest parts possible.

So another lesson - A lot of cheap woofers and tweeters require more crossover work than the expensive ones. That's why they're cheap.

EDIT: Actually I take that back some, since really it's just a 3rd order electrical/4th order acoustic with an L-pad but still...

Thank you for your help, I do realise, more and more actualy, how much there is to learn, and i am exited about it !
I just fooled around with the measurements i got from parts express but you are right, there is no point going on with those measurements.
I have to make my own ones. I am currently waiting the delivery of my mic. I guess the next step for me is to wait until i have it, make FRD files and then start again on Vituixcad. Then only i will be able to know what parts to buy for the crossover. As you said, my first try will be to crossover at 2.5-3k.

You are right about the tweeter too, i should have picked another one... We shall call it a rooky mistake, and a good lesson.

As I said, the mic should arrive monday. Until then I guess that I'll just have to wait ! In the meantime, I'll keep educating myself !

About the phase, delay and Z axis, this is still a bit confusing for me. Pushing the tweeter backwards on Z axis changes it's delay ? Changing the delay makes drivers out/in phase ?
To illustrate my question..

In this picture, the drivers are out of phase, correct ? Z axis 0mm for both drivers.



If i push the tweeter on Z axis by 34mm, the delay is -100 and the graph looks like this. Are the drivers "in phase" ? What about that peak from the woofer ? Since that peak is above the low pass, do i have to worry about it ? Or am i completely missing the point ?

[Florent]

Or should it look like this ?

[wushuliu]

Or should it look like this ?

Yes, more like this. It sounds like you are talking about the z offset (acoustic center) values. You will just have to do an educated guess, but yes that is what in phase looks like. With an RS180, the offset is usually ~.025m, but since you have the waveguide tweeter which is also recessed that means it will be a bit less, maybe .020m? Anyway, don't worry about that so much for now.

[wushuliu]

Here is a useful step by step guide you can look at until the mic arrives. References different software but you get a grasp of all the steps involved. Maybe study what you can replicate in Vituicad:

http://audio.claub.net/software/DaveDalFarra/Simple%20Loudspeaker%20Design%20ver2.pdf

Here is a guide on doing measurements by the great Jeff Bagby:

http://audio.claub.net/software/FRD_Blender/White%20Paper%20-%20Accurate%20In-Room%20Frequency%20Response%20to%2010Hz.pdf

He uses different software so you can use his or study how to do the same with Vituicad. But same principles apply.

[Florent]

Here is a useful step by step guide you can look at until the mic arrives. References different software but you get a grasp of all the steps involved. Maybe study what you can replicate in Vituicad:

http://audio.claub.net/software/DaveDalFarra/Simple%20Loudspeaker%20Design%20ver2.pdf

Here is a guide on doing measurements by the great Jeff Bagby:

http://audio.claub.net/software/FRD_Blender/White%20Paper%20-%20Accurate%20In-Room%20Frequency%20Response%20to%2010Hz.pdf

He uses different software so you can use his or study how to do the same with Vituicad. But same principles apply.

This is fantastic ! It will take some time to digest tho !

Edit : Started back from scratch based on those pdf's. Using unibox, i am pleased to see that my original enclosure and port design are ok !

[Florent]

Hi !

Not much time lately, hard working.. Managed to take some measurements. Does this looks like ok ? Took those outside but not sure if my method was the best. The Jeff Bagby's way seems a little complicated to me. I Think outside is more simple/accurate, when the weather allows it...
I was surprised by the difference of dB of the two drivers, but i didn't change any parameter when measuring one or the other...

[wushuliu]

Hi !

Not much time lately, hard working.. Managed to take some measurements. Does this looks like ok ? Took those outside but not sure if my method was the best. The Jeff Bagby's way seems a little complicated to me. I Think outside is more simple/accurate, when the weather allows it...
I was surprised by the difference of dB of the two drivers, but i didn't change any parameter when measuring one or the other...

I have no idea what you did. You need to explain things in detail. Where was the mic positioned? How far away was it from the drivers? Why did you measure both at once? I'm very confused.

Honestly, just sign up at the Parts Express TechTalk speaker forum. There are a whole bunch of people who will walk you through what you need to do. It's the best place for beginners in learning diy speaker building, period.

http://techtalk.parts-express.com/

[Florent]

OK ! Thank you for your time anyway.

[Florent]

To answer your questions, the speaker was on a table outside, with no walls within 6 meters, the mic at the same level as the speaker, at 1m away from the speaker.
I did not measure them both at once. I hooked up the woofer only, with no filter at all, used REW and umm6 mic to measure. Then, did the same with the tweeter only. Didn't change volume on computer or volume on amp. Saved those two frequency responses as txt files, imported them in vituixcad and that is what you are looking at.