I was recently playing with an active 2-way and the woofer had a phase plug. In order to get a reasonable null in the frequency response at the crossover point with a symmetrical crossover and with the phase reversed on one driver, I needed to add delay to the woofer which I would think suggests that the acoustical center of the woofer was out in front of the tweeter.
The design used a flat baffle. My measurements were taken at 1m, and nearly at tweeter height (The height I used is what lined up with my ears at listening distance - if that explanation makes sense. So that I would be time aligning the drivers as closely as possible for my listening position. )
Is this a reasonable and simple way for me to get a decent time alignment? Is it the phase plug that caused me to get a different than expected result? Or is this just the differences in relative phase of both drivers at the crossover point and something that can happen quite often? Am I missing something about time alignment here?
The design used a flat baffle. My measurements were taken at 1m, and nearly at tweeter height (The height I used is what lined up with my ears at listening distance - if that explanation makes sense. So that I would be time aligning the drivers as closely as possible for my listening position. )
Is this a reasonable and simple way for me to get a decent time alignment? Is it the phase plug that caused me to get a different than expected result? Or is this just the differences in relative phase of both drivers at the crossover point and something that can happen quite often? Am I missing something about time alignment here?
Perhaps. Once time alignment was a thing, by itself. Nowadays the idea isn't considered to be as important. With one exception...
You mention nulling the output when reversed, so you must have been reading the tutorial for where it wasn't possible to measure phase. Normally it is preferred to measure phase instead, and use it in your simulations.
I had a similar issues. The type of filter you use has an effect on the time domain of your system. So you can't use the same delay time for different filter types and slopes. For example a 48dB LR filter delays a driver more than a 12dB Bessel filter. As a start you can check the impulse response. There you can see if there is major difference between mid and high driver. But getting the phase aligned perfectly may be a bit more work imo.
In my case I had a perfect FR but the midbass was too late compared to the high frequencies.
Phase plugs on larger cone mids tend to push the acoustic center a bit forward, but it isn't always the case, depending on how far you push the driver past its pistonic range. VC inductance and cone mass are playing a huge roll here.
On a 2 way system, I always verify pulse response to get relative phase established. Otherwise there's no way of knowing for sure how to integrate the drivers electrically and acoustically. Sometimes you just need mixed electrical slopes.
On a 2 way system, I always verify pulse response to get relative phase established. Otherwise there's no way of knowing for sure how to integrate the drivers electrically and acoustically. Sometimes you just need mixed electrical slopes.
Ah OK. These points help to make sense I think. My crossover point was 2.2khz, the woofer was a poly cone (6.5") which does not seem all that stiff and it does also have a fairly high inductance relatively speaking.
I only slapped these drivers together to see how they would sound together on my active system before trying my hand at my first crossover design. To keep costs down using parts I had on hand from earlier projects which limited me just a little. In the end, I did piece together something that I would call quite acceptable. I just used a symmetrical 2nd order crossover and ignored the time alignment issue figuring I could tilt the cabinets if necessary. I didn't find it necessary though.
Not bad for using parts that were just collecting dust, I must say. Total cost, $3. 🙂
I only slapped these drivers together to see how they would sound together on my active system before trying my hand at my first crossover design. To keep costs down using parts I had on hand from earlier projects which limited me just a little. In the end, I did piece together something that I would call quite acceptable. I just used a symmetrical 2nd order crossover and ignored the time alignment issue figuring I could tilt the cabinets if necessary. I didn't find it necessary though.
Not bad for using parts that were just collecting dust, I must say. Total cost, $3. 🙂
Last edited:
When i modified my funktion ones phase plug i got 60deg difference 5.7k. Better 0deg but slight worse 30-45deg