Ok so I've been reading these boards (and others) for over a year now, and I'm about a month or two out from building my first refrence monitors for myself.
I've been reading up a lot lately on passive x-over design and I feel like I'm grasping it quite well. So on to a few questions I have for the moment......
1. I hear it is difficult/expensive to design a Zobel or similar network on a mid that will flatten the imp. curve in both directions... any takes on that? Will it be worth it in a 2-way design?
2. I have been lazy at looking at the numbers and don't feel like reverse-engineering a sample x-over to find out so I'll ask you guys
If you use two drivers with very different coil resistance (say 8ohm and 4ohm) do you just put a resistor in series with the 4ohm driver to have an effective 8ohm load? Is there a way to wire the two in the x-over to have like a 6ohm load at the speaker terminals?
.... I think that's all for now. Any help would be much appreciated!
I've been reading up a lot lately on passive x-over design and I feel like I'm grasping it quite well. So on to a few questions I have for the moment......
1. I hear it is difficult/expensive to design a Zobel or similar network on a mid that will flatten the imp. curve in both directions... any takes on that? Will it be worth it in a 2-way design?
2. I have been lazy at looking at the numbers and don't feel like reverse-engineering a sample x-over to find out so I'll ask you guys
If you use two drivers with very different coil resistance (say 8ohm and 4ohm) do you just put a resistor in series with the 4ohm driver to have an effective 8ohm load? Is there a way to wire the two in the x-over to have like a 6ohm load at the speaker terminals?
.... I think that's all for now. Any help would be much appreciated!
if this is your first project, you can forget the 3 way, its very difficult. Go for a two way. In a two way you can use a Zobel to flatten
the woofer's impedance rise, and to control the woofer's spl at the mid frequencies. Changing the resistor or the capacitor of the zobel a bit, you
can affect the spl as desired. Using a zobel on a tweeter is sometimes useful for the same reasons (controlling the high freqs this time). Special
care should be taken not to give a capacitive load at high frequencies when using a zobel for the tweeter, since its very bad for the amplfier. Use
a zobel on the tweeter unless its very necessary and you know what you re doing ( and having measuring equipment)
2) you CANNOT put a resistor in series with a 4ohm woofer. you ll destroy the amplfier's damping factor and loose much energy on heat. You can use
a resistor in series with a 4 ohm tweeter, but its not necessary, depends on the circumstances. A very large resistor in series with the tweeter
messes up its Q, so it should be done only when necessary ( to match the sensitivities). If you dont use a resistor before the tweeter,
all you have to do is calculate the high pass for a 4 ohm load (or better for the specific load it has on the crossover frequency). Then you can match
the sensitivities of both drivers with a resistor in series before the high pass filter. Crossover calculators
give inaccurate results but this is another story.
Another way to go, is to build a ready design (which is safer if the design is successful, but wont give you any additional knowledge, which you will gain by building your own design)
the woofer's impedance rise, and to control the woofer's spl at the mid frequencies. Changing the resistor or the capacitor of the zobel a bit, you
can affect the spl as desired. Using a zobel on a tweeter is sometimes useful for the same reasons (controlling the high freqs this time). Special
care should be taken not to give a capacitive load at high frequencies when using a zobel for the tweeter, since its very bad for the amplfier. Use
a zobel on the tweeter unless its very necessary and you know what you re doing ( and having measuring equipment)
2) you CANNOT put a resistor in series with a 4ohm woofer. you ll destroy the amplfier's damping factor and loose much energy on heat. You can use
a resistor in series with a 4 ohm tweeter, but its not necessary, depends on the circumstances. A very large resistor in series with the tweeter
messes up its Q, so it should be done only when necessary ( to match the sensitivities). If you dont use a resistor before the tweeter,
all you have to do is calculate the high pass for a 4 ohm load (or better for the specific load it has on the crossover frequency). Then you can match
the sensitivities of both drivers with a resistor in series before the high pass filter. Crossover calculators
give inaccurate results but this is another story.
Another way to go, is to build a ready design (which is safer if the design is successful, but wont give you any additional knowledge, which you will gain by building your own design)
Ah thank you for the help. The tweeter is the 4ohm component here and I am planning a 2-way as I have no room for even a small 3-way! The tweeter I have planned has no ferrofluid and thus a Zobel might help out with it's response.
Can you elaborate on that a little? I had not read anything about Zobel's creating a capacitive load.
...and as for going the kit rout.... that would be too simple! Why would I want to prevent the anguish and sorrow I will have when my first x-overs are wrong and sound like crap?
Can you elaborate on that a little? I had not read anything about Zobel's creating a capacitive load.
...and as for going the kit rout.... that would be too simple! Why would I want to prevent the anguish and sorrow I will have when my first x-overs are wrong and sound like crap?
if you decide to use a zobel for the tweeter, the capacitor should barely be large enough to compensate
the inductive rise of the tweeter's impedance. If for some reason* the capacitor is larger than that, the total impedance of the speaker will be capacitive at high frequencies, which may make the amplifier to oscillate (the amp experts could say more on this).
Its a good practice to leave the impedance of the speaker slightly inductive at very high frequencies. A very long speaker cable will add some capacitance too so being exacty on the limits is not that wise.
* various reasons, for example you may find that a larger capacitor at the tw zobel will make the phases of both drivers to be perfectly matched at the crossover frequency, or that it will shape the tweeter's spl as desired.
At any case, an impedance measurement of the speaker should be taken when using a zobel at the tweeter.
the inductive rise of the tweeter's impedance. If for some reason* the capacitor is larger than that, the total impedance of the speaker will be capacitive at high frequencies, which may make the amplifier to oscillate (the amp experts could say more on this).
Its a good practice to leave the impedance of the speaker slightly inductive at very high frequencies. A very long speaker cable will add some capacitance too so being exacty on the limits is not that wise.
* various reasons, for example you may find that a larger capacitor at the tw zobel will make the phases of both drivers to be perfectly matched at the crossover frequency, or that it will shape the tweeter's spl as desired.
At any case, an impedance measurement of the speaker should be taken when using a zobel at the tweeter.
I agree with Jomor.
Tweeter zobel is almost always unnecessary. More concern is flattening it's impedance hump at resonance, especially if it's not ferrofluid cooled, but that depends on your crossover slopes.
Depending on your bass/mid, a zobel may also be unnecessary there too. High quality units with faraday shields have little rise in impedance so a zobel is then unnecessary. I strongly suggest you look for a quality basss-mid that has this feature (low inductance Le is always a sign). They produce less distortion and simplify the crossover.
It doesn't matter at all that the tweeter has a different impedance to the woofer.
Tweeter zobel is almost always unnecessary. More concern is flattening it's impedance hump at resonance, especially if it's not ferrofluid cooled, but that depends on your crossover slopes.
Depending on your bass/mid, a zobel may also be unnecessary there too. High quality units with faraday shields have little rise in impedance so a zobel is then unnecessary. I strongly suggest you look for a quality basss-mid that has this feature (low inductance Le is always a sign). They produce less distortion and simplify the crossover.
It doesn't matter at all that the tweeter has a different impedance to the woofer.
Try starting with your crossover as simple as possible and only add zobel or any correction circuits if absolutely necessary.
Just start with a few different xo designs and swap and change as it's the only way to learn.
One thing... make sure you use the impedance of the driver at the crossover frequency and not the nominal impedance.
Just start with a few different xo designs and swap and change as it's the only way to learn.
One thing... make sure you use the impedance of the driver at the crossover frequency and not the nominal impedance.
Yeah, I guess I might be jumping ahead a bit much..... trying to not have to re-do the x-overs a million times, but that sounds like it comes with the territory.
From what I've read thus far, I hear that the x-over point on a 2-way should be a good 2-2.5 octaves from the dirver's -3dB point... is that a bit much/not enough? Or should I expect it to be different for every combination.
If I do expect it to be different for my specific driver choices, are there any guidlines that will help me make a more educated guess?
I was leaning towards a 3rd order Bessel or something like that but the 90 degree phase issue could be bad.....so now I'm thinking 2nd order.
From what I've read thus far, I hear that the x-over point on a 2-way should be a good 2-2.5 octaves from the dirver's -3dB point... is that a bit much/not enough? Or should I expect it to be different for every combination.
If I do expect it to be different for my specific driver choices, are there any guidlines that will help me make a more educated guess?
I was leaning towards a 3rd order Bessel or something like that but the 90 degree phase issue could be bad.....so now I'm thinking 2nd order.
Your tweeter's minimum xo point should be about 2 octaves above tweeter's Fs. The mid woofer xo depends on it's behaviour, any major peaks or dips, how it rolls off. A good starting point is where the woofer is about 3dB down at 30° off axis as it's about to roll off.... then work backwards. You use the natural roll off on the woofer to your advantage, providing it's well behaved and doesn't have any major peaks, then you have to go lower.
Each set of drivers is completely different so there's no rules like 2x this or 3x that. What you are looking for is a smooth transition between the drivers, and make sure the driver is not working in a range which is not suitable for it and could be as simple as a 1st order electrical (combined with the driver 2nd order roll off) on the woofer with a 2nd order electrical (combined with say a 1st order driver acoustic roll off) on the tweeter that will give a 3rd order acoustic slope.
Remember it's the acoustic slope you need in the end which is electrical + driver acoustic roll off.
Some one around here once said go with the lowest order crossover that can get the job done and those are wise words indeed.
Something to think about.
Each set of drivers is completely different so there's no rules like 2x this or 3x that. What you are looking for is a smooth transition between the drivers, and make sure the driver is not working in a range which is not suitable for it and could be as simple as a 1st order electrical (combined with the driver 2nd order roll off) on the woofer with a 2nd order electrical (combined with say a 1st order driver acoustic roll off) on the tweeter that will give a 3rd order acoustic slope.
Remember it's the acoustic slope you need in the end which is electrical + driver acoustic roll off.
Some one around here once said go with the lowest order crossover that can get the job done and those are wise words indeed.
Something to think about.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.