how to time align drivers (with minidsp) ?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

How long is a piece of string once you get into the details ?

Clearly simple time delay will align the drivers acoustic centres, after
that its not clear what are you talking about related to the acoustic
crossover functions employed, digital phase correction for complex
x/o functions causes more trouble than it fixes usually, YMMV.

rgds, sreten.
 
i´m just talking about acoustic centers , i thought there would be some software with test tones or similar to achieve this , i tryed to time align my tweeters and woofers and did not like the results . Seemed as i delayed the tweeter it caused the tweeter volume to go down , maybe this is too complicated to do in a 3 way plus sub :(

anyway after 20 years of listening to these speakers without time alignment i can live without it :)
 
Last edited:
@ lduarte,

John Reekie did a very comprehensive tutorial on how to time align drivers with a miniDSP.
It's located in the Press Reviews (along with some of his other great articles)

Product Reviews | miniDSP
In the "miniDSP 2x4 for 4way Open Baffle project", he gives a clear explanation but please take the time to read that whole article (part1&2) together to really understand the steps.

DevTeam
 
It will be very hard to do such a thing by ear. With symmetric crossover slopes you can calculate the time-delay needed to account for the distance between the center of the coils for both drivers.
I would however strongly encourage using measuring equipment.

If you want to correct the direction of a mid-and-tweeter lobe take a series of measurements on the vertical axis. I will use -30,0,+30 degrees for this example but the more you get the better. Some measurements will show flat response at the crossover frequency and other will show a notch. Adding time-delay to the tweeter will rotate the lobe(flat FR) upwards. Assuming a symmetric slopes XO you can adjust the time-delay and remeasure until the 0 is flat and the -30 and +30 notches become symmetrical.
If the filter slopes are not symmetric the notches will not be symmetric but you can still reposition the lobe to your liking.

---
You may also use time-delay to adjust the phase of a crossover similarly to adjusting the slope of the crossover filter. This would be more useful in lower frequencies like between a mid and woofer. Ideally you want the phase of both drivers to track linearly through the crossover frequency. If they are misaligned the phase plot will not track smoothly and adjusting the time-delay affect that.
 
@ lduarte,

John Reekie did a very comprehensive tutorial on how to time align drivers with a miniDSP.
It's located in the Press Reviews (along with some of his other great articles)

Product Reviews | miniDSP
In the "miniDSP 2x4 for 4way Open Baffle project", he gives a clear explanation but please take the time to read that whole article (part1&2) together to really understand the steps.

DevTeam


The site with the tutorial seams to be down. Like to find out how to time-align drivers in a speaker.
I did read the tutorial some times ago but need it to refresh my memeory.
 
Time Alignment

Doing time alignment with MiniDSP.

1. Using the same crossover frequency and slope for the lowpass and highpass driver...

2. Place your microphone at the desired level for listening and aim it between the centers of two drivers at about 3' distance.... ( you could measure closer, but if you measure further, there will be increasing reflections that will swamp the phase "picture")

3. Reverse/Invert the tweeter phase.

4. Take a measurement

5. Delay the tweeter phase by 1 step

6. Repeat steps 3 & 4 until you see a dropout / cancellation precisely upon the crossover frequency. You will take MANY measurements. And, you will see the frequency response dropout "valley" gradually increase as the ideal phase alignment approaches.

7. Remove the phase inversion for the tweeter.

Bam! You are in phase a the crossover point.:D
 
Same as above for me, but I only done it between the mains and sub. I applied a time delay by ear to get roughly in the correct region, then started measuring. I took about 30 measurements, in 0.1ms increments.

Before and after response attached. Red is sub. Orange is Mains, Green is total.

Before graph shows almost totally out of phase, lots of destructive summation

After response shows my best attempt down to the closest 0.1ms at constructive summation. Sounds much better now this is corrected.

Thanks to this link - DIY Speakers and Subs - AVS Forum | Home Theater Discussions And Reviews
 

Attachments

  • Destructive Summaition.jpg
    Destructive Summaition.jpg
    51.9 KB · Views: 315
  • Constructive Summation-1.jpg
    Constructive Summation-1.jpg
    52.1 KB · Views: 317
Doing time alignment with MiniDSP.
1. Using the same crossover frequency and slope for the lowpass and highpass driver...
2. Place your microphone at the desired level for listening and aim it between the centers of two drivers at about 3' distance.... ( you could measure closer, but if you measure further, there will be increasing reflections that will swamp the phase "picture")
3. Reverse/Invert the tweeter phase.
4. Take a measurement
5. Delay the tweeter phase by 1 step
6. Repeat steps 3 & 4 until you see a dropout / cancellation precisely upon the crossover frequency. You will take MANY measurements. And, you will see the frequency response dropout "valley" gradually increase as the ideal phase alignment approaches.
7. Remove the phase inversion for the tweeter.
Bam! You are in phase a the crossover point.:D
I do not recommand this method. I've investigated it with many crossover configurations, results are unsure, for example :
https://www.diyaudio.com/forums/multi-way/346156-crossover-null.html#post5996314

Searching the emission plane of each driver refered to its support front plane using the minimal phase method is the way to go. It was used as soon as 1977 :
AES E-Library >> The Application of Digital Techniques to the Measurement of Loudspeakers
Using steps responses with order 1 or order 2 Linkwitz-Riley crossovers and direct-reverse connections are also very instructive when well interpreted.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.