Home Theater Forum and Systems banner

1 - 14 of 14 Posts

·
Registered
Joined
·
11 Posts
Discussion Starter #1 (Edited)
Hi, I'm new to speaker building and design. I was wanting other opinions on my design for my first build. I have not built these yet, and any improvements or suggestions will be appreciated. So, it will have two HIVI B3N 3" and two Peerless 830880 passive radiators. I really want to use passive radiators, I think they're cool, and think they would be fun to use. I attached a screencap of the bassbox model for this design, if you want more info/pics please ask.
 

·
Super Moderator
Joined
·
1,921 Posts
Re: Wanting opinions/improvements for fist build

Welcome to HTS! You need five posts before you can share links and pics, please visit the Post Padding Thread.

Using two B3N and passive radiators is a cool idea and should work well for bass performance.

You should, however, consider a "1.5-way" crossover design. This will add a series inductor in front of the lower woofer which will roll-off the highs. This will greatly help the off-axis high frequency response of the speaker. Generally you only want one source for high frequencies, unless you really know what you're doing, or there will be acoustic phase issues.
 

·
HTS Moderator , Reviewer
Joined
·
5,742 Posts
Re: Wanting opinions/improvements for fist build

Thank you. Do you have any suggestions for what frequency I should use for the crossover?
for a good crossover you're going to need a mic and record responses from the individual drivers to get a good idea and work backword from there. not saying it's impossible but a crossover is probably the most crucial and difficult part of designing a speaker using untested components
 

·
Registered
Joined
·
3 Posts
Re: Wanting opinions/improvements for fist build

I like to think that I am not overly stupid, but for the life of me I can not see where I make posts. Can you advise please?
 

·
Registered
Joined
·
29 Posts
Re: Wanting opinions/improvements for fist build

for a good crossover you're going to need a mic and record responses from the individual drivers to get a good idea and work backword from there. not saying it's impossible but a crossover is probably the most crucial and difficult part of designing a speaker using untested components
I heartily agree. Having worked for several speaker manufacturers I can tell you that there can be a pretty great tolerance between the speaker that you have in hand and the published spec. If you can make good measurements you can better dial in what you are trying to achieve. And I would suggest making most of your measurements outside and away from the influences of boundaries.
 

·
Super Moderator
Joined
·
1,921 Posts
Re: Wanting opinions/improvements for fist build

I heartily agree. Having worked for several speaker manufacturers I can tell you that there can be a pretty great tolerance between the speaker that you have in hand and the published spec. If you can make good measurements you can better dial in what you are trying to achieve. And I would suggest making most of your measurements outside and away from the influences of boundaries.
for a good crossover you're going to need a mic and record responses from the individual drivers to get a good idea and work backword from there. not saying it's impossible but a crossover is probably the most crucial and difficult part of designing a speaker using untested components
Kinda.

For a 1.5-way you can easily ballpark the crossover because of physics, which doesn't need the driver's individual response. The limit for upper frequency response of two drivers in a speaker is due to the center-to-center spacing of the drivers, and the corresponding 1/4-wavelength.

For example the closest you can possibly mount two B3Ns (3.5" diameter) would be 3.5". The rule of thumb to avoid lobing in the speaker's polar response would be to keep the crossover under 13500/3.5=3.86kHz. You would then simply need the speaker's true complex impedance to size a single inductor to place in series with the lower woofer to pull off the 1.5-way. If you can't measure a true impedance sweep, you could use a textbook 1st-order LPF and add a zobel filter to the woofer.

I like to think that I am not overly stupid, but for the life of me I can not see where I make posts. Can you advise please?
To create a new thread:
 

Attachments

·
Registered
Joined
·
11 Posts
Discussion Starter #8
Can you give me any links to any articles so I can further read up on this subject. I'm also confused where 13500 came from in 13500/3.5=3.86khz.
 

·
Super Moderator
Joined
·
1,921 Posts
The number 13500 is the speed of sound in inches per second, I believe. So if you realistically had closer to 4", you'd want to stay below around 3.3kHz, making something like a 3kHz crossover to shoot for. Without doing any measurements and the B3N specs, a textbook filter (using an online calculator) would be about a 0.4mH inductor in series with the lower of the two woofers. Add a textbook zobel filter to compensate for the woofer's rising impedance and it would be fairly accurate, again without measurements.
 

·
Registered
Joined
·
11 Posts
Discussion Starter #11
So after some research I believe a proper zobel filter for this would be a series 15uf capacitor and 6.4Ω resistor in parallel with the lower driver. I'm I correct?
 

·
Super Moderator
Joined
·
1,921 Posts
That's about right. I get 8.1Ω and 16.4uF based upon the specsheet impedance at 1kHz and an Re of 6.5Ω, but as Zobels go either would probably work fine. To go the extra mile you could verify this by importing the impedance data provided by Parts Express, strip the text and leave the columns of data then rename to .ZMA, and import it into Passive Crossover Designer (PCD). You'd want the R and C values such that the rising impedance is as ruler-flat as possible through your crossover region in order for your textbook filter to work properly. Ideally, you would start to use PCD rather than rely on online filter calculators for speaker designs.

To finish off your novice design, you could also add an RL shelving filter to account for baffle step. Usually 2-4dB is a good range depending on the speaker use and your taste. If you want them to sound bassy and tubby, try something in the 4-6dB range. This will compensate the high frequencies down, since low frequencies suffer an efficiency loss as they become omni-directional. The penalty is up to -6dB on a wide open environment. Another pesky limitation of physics. So, my recommended crossover includes 1) Zobel on each woofer to accommodate textbook filters 2) 1st-order LPF for the second woofer 3) Parallel RL shelf filter in series before everything else to serve as baffle step compensation. Let me know and I can sketch a schematic.

Oh and one more thing, after looking at the B3N specsheet I would say lower your 1st-order LPF to something like 1kHz. I'd say so because of the strong rise in the response and it won't hurt to attenuate the woofer breakup in the lower woofer as much as possible.

I think the simple 1.5-way with the B3Ns is a great first project for a beginner, you can always make mods later if you want. There is nothing worng with textbook filters and designing speakers without measurement equipment. There is a steep learning curve in speaker design and it's good to start simple. This should be a nice sounding set of speakers and will be light years ahead of just dropping woofers into a box and calling them a speaker. Good luck :T
 

·
Super Moderator
Joined
·
1,921 Posts
Yep, that's the Zobel and 1.5-way. Then for a BSC circuit you'd add the parallel RL in series right after the speaker terminal. It depends on the dimensions of the face of your finished speaker (the baffle), and your chosen attenuation, but that R and L would maybe be in the neighborhood of 0.3-0.5mH and 1.5-2.2Ω.

Once again, with this many components in the mix it would be best to model it in PCD to make sure everything will interact properly. That being said I don't see an issue with going for it.
 

Attachments

1 - 14 of 14 Posts
Top