Indeed, I had thought about setting the tweeter back a bit as well. It would have been helpful in one of the earlier versions of the xo that I was working with although I didn't think of it at the time but switching the electrical order up did a fairly good job of fixing that. It is worth contemplating once we are dealing with measured files that have accurate acoustic center delays.
The SB29RDC is a pretty robust tweeter and I'm going by another user's report of its successful use at 1600Hz. But that's why I'm also being fairly particular about a steep roll-off. And why I'm also considering what happens when you feed it 100W to be not very realistic. Even 40W might be too high for it. I'm not really sure.
I have used resonance compensation filters on both tweeters and woofers and like the results. Here, one on the tweeter might be helpful although with the LR4 xo at 1600Hz, the 600Hz resonance is already down about 32dB from the fundamental. Could still be a good precautionary measure though.
In this 2-way situation, I don't think there is any need to worry about the woofer resonance.
The SB29RDC is a pretty robust tweeter and I'm going by another user's report of its successful use at 1600Hz. But that's why I'm also being fairly particular about a steep roll-off. And why I'm also considering what happens when you feed it 100W to be not very realistic. Even 40W might be too high for it. I'm not really sure.
I have used resonance compensation filters on both tweeters and woofers and like the results. Here, one on the tweeter might be helpful although with the LR4 xo at 1600Hz, the 600Hz resonance is already down about 32dB from the fundamental. Could still be a good precautionary measure though.
In this 2-way situation, I don't think there is any need to worry about the woofer resonance.
Ok, I'm not at all familiar with the tweeter. Only general experience on crossing tweeters.
Using a waveguide is out of the question? You would probably need a large step with a 8" woofer.
With even order filters there are usually (maybe always, not aware of possible variations) a coil parallel to the tweeter that takes care of resonance peak pretty well.
My impedance correction test on woofers was actually a 2way, has a 1,5mH on 2x8ohm woofers with a bit high Fs at abt 70Hz (sealed). It stayed in place. Maybe partly the improvement was because I'm driving them with a low power class A amp (JLH69).
Using a waveguide is out of the question? You would probably need a large step with a 8" woofer.
With even order filters there are usually (maybe always, not aware of possible variations) a coil parallel to the tweeter that takes care of resonance peak pretty well.
My impedance correction test on woofers was actually a 2way, has a 1,5mH on 2x8ohm woofers with a bit high Fs at abt 70Hz (sealed). It stayed in place. Maybe partly the improvement was because I'm driving them with a low power class A amp (JLH69).
Thank you all for the responses, I am a little overwhelmed I must admit 
. I'll try to address the questions about my design.
I plan on using these as stand-alone floor-standing units, therefore vented. I was greatly inspired by Sean Casey's Zu Audio Druid VI speakers. I really like the look of them and I am thinking of having the woofer above the tweeter if possible. This leads into two points about the inspirations I got from the Druids, and Sean Casey's philosophy.
First, I am very intrigued by the idea of pursuing high efficiency, even if it means losing some flatness in frequency response. Now, don't take me wrong, I fully understand and respect the reasons and advantages of going for a flat frequency response, but my engineer mind finds efficiency ever so slightly more attractive.
Second, I've seen mentioned a couple of times the question of using a waveguide on the tweeter. I am open to the idea of using one, I was just initially put off the idea after reading an article on Troel Gravensen's website, where he was testing this driver, and experimenting with swapping the waveguide. If you'd like the link for this I can try to dig it up for you, but I am sure it is easy enough to find after a simple googling.
I got bored in the Christmas break and decided to make a model of the cabinet in fusion360. I made this with only the aesthetics in mind so keep that in mind.
I hope I am not committing any cardinal sins with my "requirements". Thanks for the help again.
PS: I am currently studying for exams, so I might not be too active in the coming week. I will definitely have a lot of questions once I can focus a little more on this project.
. I'll try to address the questions about my design.I plan on using these as stand-alone floor-standing units, therefore vented. I was greatly inspired by Sean Casey's Zu Audio Druid VI speakers. I really like the look of them and I am thinking of having the woofer above the tweeter if possible. This leads into two points about the inspirations I got from the Druids, and Sean Casey's philosophy.
First, I am very intrigued by the idea of pursuing high efficiency, even if it means losing some flatness in frequency response. Now, don't take me wrong, I fully understand and respect the reasons and advantages of going for a flat frequency response, but my engineer mind finds efficiency ever so slightly more attractive.
Second, I've seen mentioned a couple of times the question of using a waveguide on the tweeter. I am open to the idea of using one, I was just initially put off the idea after reading an article on Troel Gravensen's website, where he was testing this driver, and experimenting with swapping the waveguide. If you'd like the link for this I can try to dig it up for you, but I am sure it is easy enough to find after a simple googling
. I got bored in the Christmas break and decided to make a model of the cabinet in fusion360. I made this with only the aesthetics in mind so keep that in mind.
I hope I am not committing any cardinal sins with my "requirements"
. Thanks for the help again.PS: I am currently studying for exams, so I might not be too active in the coming week. I will definitely have a lot of questions once I can focus a little more on this project.
Attachments
Last edited:
I agree it's not clear cut but when it comes to choosing a target for this exercise I'd do the same, it makes sense at this stage.
In XSim you can get the driver phase by a small checkbox on the dialog when opening the curves in the SPL window. XSim doesn't store window configurations so we have to open them manually each time.
I'm prone to agree.
Really??? I would have never noticed
That's because I had to revert to 3rd order to get rid of that ringing in the shunt inductor. You may be able to clean it up and still get your curve.
Okay, makes sense.
Think of it as "sibilance control"... a minor inefficiency for a better purpose.
Nice!
I'm not sold on the woofer on top thing though.
Along with the acoustic and electrical we should also consider the physical... wouldn't that speaker be somewhat top-heavy and thus something of a tip-over risk around kids and pets?
So I have been reading this with interest and wondering. If someone were to try designing speaker for the first time would it not be an advantage to biamp the speaker and use an active crossover. It seems like not having to consider a passive crossover might be an advantage. Also does anyone know of a design with a 8" woofer with one of the newer "ribbon" style tweeters you see so much of these days. I would really like to try that with an active approach.
This aspect could also materially affect Z offset, therefore putting your design off course. Unless of course, you knew the baffle geometry beforehand and basis for the ZMA and FRD files provided
I was planning on casting the base out of concrete, I assumed that would add enough mass to balance things out.
I'm not sure that it would be an advantage but it may produce a useable result a little quicker. In the end the considerations for producing a good speaker are the same either way, so there's no free lunch.
It has been done.
Here's my attempt. I targeted a net 87dB sensitivity. Although the woofer is stated at 92dB - I have assumed a narrow baffle and being high off the floor, you wont get much boundary re-enforcement.
I have redone the FRD/ZMA files - based on SB Acoustics measurements plus applied baffle step and diffraction ripple on a 962(h) x 274 (w) baffle, placing the woofer 825mm from the floor and tweeter below it at 667mm from the floor. The assumed listening axis is the woofer (being on top) and Z offsets for crossover modeling are 4.15mm for tweeter and 39mm for woofer, with an assumed 3 metre distance.
A net volume of 42L (excluding port), tuned to 33Hz (FB) will give you an F3 of 37Hz.
PS: of the 3 crossovers (excluding mine), Douglas' modeled the best but didnt' have enough BSC in my opinion and tweeter needed a little more padding.
I have redone the FRD/ZMA files - based on SB Acoustics measurements plus applied baffle step and diffraction ripple on a 962(h) x 274 (w) baffle, placing the woofer 825mm from the floor and tweeter below it at 667mm from the floor. The assumed listening axis is the woofer (being on top) and Z offsets for crossover modeling are 4.15mm for tweeter and 39mm for woofer, with an assumed 3 metre distance.
A net volume of 42L (excluding port), tuned to 33Hz (FB) will give you an F3 of 37Hz.
PS: of the 3 crossovers (excluding mine), Douglas' modeled the best but didnt' have enough BSC in my opinion and tweeter needed a little more padding.
Attachments
Well technically it is stated at 92dB for 2.83V/1m, and as it is a 4ohm driver that's 2 watts, it's a little misleading, the woofer sensitivity is more around 89dB. So the 87dB target sounds pretty good to me.
I've noticed no one has designed a crossover with a BBC dip at the crossover point. Now, I don't know if it is an antiquated philosophy at this point or if people just don't bother with it. I'd love to hear your thoughts on the concept.
I have another question, this might be a little stupid but oh well. When is was modeling the box for these drivers I noticed that you cannot achieve a smooth roll-off and good power handling, ie. keeping the cone excursion within Xmax. My question is, do manufacturers add a highpass filter bellow the -3dB point of the woofer as low-frequency protection of sorts.
If you decide to read this thread you'll get the lowdown. There are a number of potential reasons covered if you read far enough. BBC Dip
In VituixCAD you enter the Z offset in the driver tab not in the crossover. That's why all drivers have Z=0. Since I had already accounted for rolloff in the baffle diffraction response, I didn't enter an additional Y offset either.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.