I’m going to build 2.5-way speakers. My 1.25” tweeters can go as low as 1.5kHz. And my 10” woofers can reach that frequency.
At first, I plan to wire the woofers as diagram below, the traditional 2.5-ways configuration:
Unfortunately, I just found the woofers are stamped 4 Ohms on their magnets, and, the measured Re showed results of 3.6 Ohms. So, they are rated 4-Ohm drivers.
Accordingly, is the diagram above still usable? Because the diagram shows two woofers connected in parallel to each other, although there is a L3 in series with the second woofer. If not, please suggest me the more suitable configuration.
At first, I plan to wire the woofers as diagram below, the traditional 2.5-ways configuration:
Unfortunately, I just found the woofers are stamped 4 Ohms on their magnets, and, the measured Re showed results of 3.6 Ohms. So, they are rated 4-Ohm drivers.
Accordingly, is the diagram above still usable? Because the diagram shows two woofers connected in parallel to each other, although there is a L3 in series with the second woofer. If not, please suggest me the more suitable configuration.
You may be stuck with a 2 way and baffle step correction.
I can imagine a .5 series crossover segment but you would need to model it and see if it is even reasonable. I suspect that the gain you expect by driving both woofers in the lower frequency range would be taken away by the loss you get from the higher impedance seen by the amp in that same range.
I can imagine a .5 series crossover segment but you would need to model it and see if it is even reasonable. I suspect that the gain you expect by driving both woofers in the lower frequency range would be taken away by the loss you get from the higher impedance seen by the amp in that same range.
Yes, that's exactly what you have at low frequencies. Series connection is possible. The 0.5 circuit is no less tricky, you might want to simulate it.
Could you explain in more detail what to do with the series-connected woofers?
When you connect two identical woofers in series, shunt one with an inductor or capacitor and apply a frequency-independent voltage across the whole network, the on-axis response just stays the same as with a single woofer, only the directivity changes. When the impedance of the inductor or capacitor is large, each woofer gets half the voltage, produces half the sound pressure and on axis with equal distances, those halved sound pressures add up to the full sound pressure again. Or am I missing something?
When you connect two identical woofers in series, shunt one with an inductor or capacitor and apply a frequency-independent voltage across the whole network, the on-axis response just stays the same as with a single woofer, only the directivity changes. When the impedance of the inductor or capacitor is large, each woofer gets half the voltage, produces half the sound pressure and on axis with equal distances, those halved sound pressures add up to the full sound pressure again. Or am I missing something?
You're right about that. One way is to use two parts to the filter.
@presscot Do you have the model numbers of you speakers? Do you know how wide the baffle of your speaker box will be?
It seems to me that this is a situation that many may find themselves in so if we go a little further then we can get more knowledge about this solution from the the experienced on the site.
Jeremy
It seems to me that this is a situation that many may find themselves in so if we go a little further then we can get more knowledge about this solution from the the experienced on the site.
Jeremy
The woofers used are model ADS S10 car subwoofer. They’re 4-Ohm with 2” single voice coil.
The baffle width is 29 cm.
I had a similar situation and placed the 0.5 woofer in it's own enclosure and bi-amped. So 2.5 became 2.2 channels with an active filter and gain control for the 0.5 woofer. Crossed over around half the baffle step frequency as full BSC was too much. Also have used the same method with a passive filter for the 0.5 woofer.
Assuming that the woofers are ideal 4 ohm resistors and that everything else is ideal as well, I calculated these poles and zeros for the transfer from the input voltage to the sum of the voltages across the woofers:
Pole at -2160.962915 rad/s, corner frequency 343.9279298 Hz
Zero at -4000 rad/s, corner frequency 636.6197724 Hz
Pole at -10283.48153 rad/s, corner frequency 1636.666918 Hz
Zero at -16666.6666... rad/s, corner frequency 2652.582385 Hz
At the high-frequency asymptote, the gain is 1/3, so there is a factor of 3 rather than 2 of baffle step correction.
Is the idea behind the circuit to make the correction a bit smoother than with just an LR parallel network in series with one woofer?
No.
In practice a driver is far from ideal when it comes to applying a theoretical filter. The baffle step is far from ideal as well, often having misshapen lobes.
The circuit gives a method for creating the shape of a kink in the response. It also demonstrates usable deviation and separation of the two woofers.
In practice a driver is far from ideal when it comes to applying a theoretical filter. The baffle step is far from ideal as well, often having misshapen lobes.
The circuit gives a method for creating the shape of a kink in the response. It also demonstrates usable deviation and separation of the two woofers.
This might give more background:http://www.planet10-hifi.com/downloads/Dual-Driver-Wiring.pdf
AllenB showed my preferred wiring. In a real room parallel 2.5 gives too much bass in a real room.
dave
AllenB showed my preferred wiring. In a real room parallel 2.5 gives too much bass in a real room.
dave
What is a facor of 3?
A parallel 2.5 way has 6 dB more output below the bottom XO point. For flat response they would have to be way out into a large room or outside.
Real rooms typically need 0-3 dB extra gain. 6dB is a facor of 4, 3dB and change in impedance are a factor of 2.
dave
A parallel 2.5 way has 6 dB more output below the bottom XO point. For flat response they would have to be way out into a large room or outside.
Real rooms typically need 0-3 dB extra gain. 6dB is a facor of 4, 3dB and change in impedance are a factor of 2.
dave
Last edited:
A lot of your options regarding crossover details WILL depend on how 'GUTSY' your amplifier is. ( can it take 2 ohms for example )
You can incorporate inductors that have higher DCR than usual > especially the second coil in your first circuit >
However, you may need to use active EQ to get the 'nice' bottom end you desire.
PS.
Inductors with higher DCR actually cost less.
You can incorporate inductors that have higher DCR than usual > especially the second coil in your first circuit >
However, you may need to use active EQ to get the 'nice' bottom end you desire.
PS.
Inductors with higher DCR actually cost less.
@planet10 In post #6, https://www.diyaudio.com/community/threads/2-5-way-with-two-4o-woofers.421228/post-7872919 , there are two woofers. As this whole thread is about 4 ohm woofers, I assume they are 4 ohm.
At high frequencies, one of them is shunted by a capacitor, so the series connection is then 4 ohm. It is driven from an 8 ohm resistor with an inductor in parallel, so that's 8 ohm.
An 8 ohm-4 ohm voltage divider divides by 3. It divides the voltage by 3, so that's about -9.54 dB.
Of course I assume now that the woofers are impedance-corrected so they can be treated as resistors, maybe the woofer's inductances are supposed to limit the correction. Besides, even when the woofers are supposed to be pure 4 ohm, the highest corner frequency is 2652.582385 Hz. Maybe the woofer is already supposed to be crossed over there. I don't know.
At high frequencies, one of them is shunted by a capacitor, so the series connection is then 4 ohm. It is driven from an 8 ohm resistor with an inductor in parallel, so that's 8 ohm.
An 8 ohm-4 ohm voltage divider divides by 3. It divides the voltage by 3, so that's about -9.54 dB.
Of course I assume now that the woofers are impedance-corrected so they can be treated as resistors, maybe the woofer's inductances are supposed to limit the correction. Besides, even when the woofers are supposed to be pure 4 ohm, the highest corner frequency is 2652.582385 Hz. Maybe the woofer is already supposed to be crossed over there. I don't know.
Last edited:
4 to 8 Ω is a factor of 2, You are stretching things with to bring in a voltage divider.
dave
How high a woofer can go is dependent on the woofer.
More of a trend is crossing over lower to get better integration with the XO (ie closer to quarert-wave spacing of the drivers relative to the XO frequency.
I cross the woofers — that go up to 10k on their own — at 250 Hz.
dave