Curving the front does more than curving the sides.I'm considering having a go at a curved cabinet as opposed to a standard box.
Curving the front does more than curving the sides.
I have a thing for Sonus Faber.
I'll probably be trying something like that - the lute shape, like the Olympica or their recent "low end" Veneres.
I know this might not make me new friends here, but I like speakers as much for the looks as for the sound ;-)
Thanks for your explanation 😉
For now it's pretty irrelevant though, because I'll first be testing / learning using a simple box. If, in a year or two, I find my speaker to be good enough, I'll start on the actual woodwork for a pretty enclosure - which means completely recalculating everything, but by then I should have picked up the necessary skills from working with the test box 🙂
For now I'll be using this box I just slapped together just so I can get going asap playing with DSP crossovers etc:

Last edited:
So I keep reading and taking advice - I should protect my tweeter from low frequencies.
I'll be using DSP for my crossovers, but I might make a mistake and the tweeter might receive the wrong input.
The Hivi RT1C-a needs to be crossed over higher than 4500Hz.
So, if I want the protection cutoff frequency to be (at least) two octaves lower, it would need to be <= 1125Hz.
Using this calculator : Car Audio - Speaker Crossover Chart and Capacitance vs. Frequency Calculator(High-pass)
I find that a 30uF capacitor would cut off anything below 1060Hz.
So if I'm not mistaken, a capacitor around 30uF in series with the tweeter should protect it from any unintended low frequencies, while still being low enough not to interfere with my DSP crossover that will be at >=4500Hz.
I'll be using DSP for my crossovers, but I might make a mistake and the tweeter might receive the wrong input.
The Hivi RT1C-a needs to be crossed over higher than 4500Hz.
So, if I want the protection cutoff frequency to be (at least) two octaves lower, it would need to be <= 1125Hz.
Using this calculator : Car Audio - Speaker Crossover Chart and Capacitance vs. Frequency Calculator(High-pass)
I find that a 30uF capacitor would cut off anything below 1060Hz.
So if I'm not mistaken, a capacitor around 30uF in series with the tweeter should protect it from any unintended low frequencies, while still being low enough not to interfere with my DSP crossover that will be at >=4500Hz.
I use a 15-20uF on that tweeter for that purpose, largest case. In practice, 10uF would likely be better.
Wolf
Wolf
The amp or the source might thump, the source might turn full range (eg. pc based processing, several ways this might happen), or you might kick out the cable.I might make a mistake and the tweeter might receive the wrong input.
Sure, use a smaller cap and make it part of the roll-off. It's easy to implement.In practice, 10uF would likely be better.
Wolf
I use a 15-20uF on that tweeter for that purpose, largest case. In practice, 10uF would likely be better.
Wolf
Eh?
As far as I've been able to figure out, inserting a capacitor introduces phase rotation within several octaves of its corner frequency, meaning that the capacitor will change the phase angle well into the tweeter's passband unless the value of the cap is very large.
Why do you think 10uF (which would put its corner frequency at 3180Hz, quite near the crossover frequency of the active filter) would be better?
Also, I read here: http://www.pispeakers.com/Speaker_Crossover.pdf on page 27 that a capacitor in series with voice coil inductance forms a resonant circuit that can cause peaking, which would be another reason to make the capacitor large?
Last edited:
Passive and active filters are no different in this regard. You'll get the phase change regardless of which way you do it.
Passive and active filters are no different in this regard. You'll get the phase change regardless of which way you do it.
Sure, but I think I prefer designing the DSP crossover without the influence of the tweeter protection cap. This way I can remove the cap later on once I think everything is finalized.
I recommend a cap for the tweeter if it is expensive. Phase roll goes away when you eq before xo. Net acoustic slope is what counts. I have burnt couple of tweeters ...
I have just received and started reading Vance Dickason's "Loudspeaker Cook Book". 🙂
While reading I've decided to buy the Dayton Dats V3 system - not because I *need* it to go on, but because I'm interested in the hobby and want to do some measurements.
I want actual data, and I want to figure out how to use that data properly 🙂
This blog continues..
While reading I've decided to buy the Dayton Dats V3 system - not because I *need* it to go on, but because I'm interested in the hobby and want to do some measurements.
I want actual data, and I want to figure out how to use that data properly 🙂
This blog continues..
A planar really has very little inductance. Heck, most teeters have less than 0.1mH of inductance.
I think 10uF is better because it will protect the tweeter better.
News flash- all xovers cause phase shift, unless you use FIR filters. Active or passive doesn't matter. You just have to know what to do with the resulting phase.
Later,
Wolf
I think 10uF is better because it will protect the tweeter better.
News flash- all xovers cause phase shift, unless you use FIR filters. Active or passive doesn't matter. You just have to know what to do with the resulting phase.
Later,
Wolf
working on my test-enclosure... I've doubled up the baffle thickness to 44mm just to experiment with different ways to make facets (is that a correct english term?).
I tried to do 2 tonight by screwing on metal strips to define the edges and then use a band sander manually to sand away everything until it hit the metal.
It worked, but it's not very precise.
Next time I'll try using a saw to saw along the edges of the metal strips, see how that works out.
and eventhough this is just a test-enclosure I still added a brace because why not 🙂
I tried to do 2 tonight by screwing on metal strips to define the edges and then use a band sander manually to sand away everything until it hit the metal.
It worked, but it's not very precise.
Next time I'll try using a saw to saw along the edges of the metal strips, see how that works out.

and eventhough this is just a test-enclosure I still added a brace because why not 🙂

Last edited:
By the way, it's very comforting to work on an enclosure you know will not be a "final" enclosure - it won't be veneered so no perfect edges needed, simply jigsawing driver holes / braces instead of trying to route them perfectly... (Still chamfering the backside though 🙂)
I can advise this to anyone suffering from crippling perfectionism - make something to play with knowing full well it won't be the final cabinet 🙂
I can advise this to anyone suffering from crippling perfectionism - make something to play with knowing full well it won't be the final cabinet 🙂
All right this might be a stupid question, but since we were discussing using a capacitor to protect the tweeter...
I just got my Dayton DATS V3 system.
The help tells me I can measure my tweeter just like I do my woofer.
Which means it will be sending a sweep from 5Hz to 20kHz to my tweeter.
Eh... wouldn't sending frequencies that low kill my tweeter?
Or is the sweep duration too short to cause damage?
I just got my Dayton DATS V3 system.
The help tells me I can measure my tweeter just like I do my woofer.
Which means it will be sending a sweep from 5Hz to 20kHz to my tweeter.
Eh... wouldn't sending frequencies that low kill my tweeter?
Or is the sweep duration too short to cause damage?
Last edited:
There should be an option to high pass the tweeter. Setting the highpass to say 200Hz is probably ok. I've swept my tweeters from 20 Hz (no higher than 2.8V) without issue, but they are high powered tweeters...
Note I've only used holm impulse, arta and rew but all have the ability to high pass the sweep.
Tony.
Note I've only used holm impulse, arta and rew but all have the ability to high pass the sweep.
Tony.
- Home
- Loudspeakers
- Multi-Way
- An idiot's blog to design and build speakers