Hello there!
I need some help in designing a workable crossover for this driver, because the original is crap, doesn't have any correction for the upper range peak.
I am not really an expert in metal cones, also the (twin) peak
seems to be pretty wide, so if anyone out there could model a useful circuit for this, I would appreciate it very much.
(The Speaker in question is Avalanche Audiostone 6.5)
The planned X-over point is at 2.5k -2nd order- to a 1" Seas alu dome with a waveguide
I need some help in designing a workable crossover for this driver, because the original is crap, doesn't have any correction for the upper range peak.
I am not really an expert in metal cones, also the (twin) peak
seems to be pretty wide, so if anyone out there could model a useful circuit for this, I would appreciate it very much.(The Speaker in question is Avalanche Audiostone 6.5)
The planned X-over point is at 2.5k -2nd order- to a 1" Seas alu dome with a waveguide
Attachments
Last edited:
Personally like to use this toolkit: OmniMic + DATS
However you may use any other tool and mic combination. You'll need both electrical (impedance vs. frequency) and acoustical (frequency response vs. db amplitude) measurements.
Supplies:
A bipolar (or better yet, film) 20uF capacitor to protect the tweeter. Don't spend more than $5. You can forego the cap if you use test signals without bass and you feel lucky. The cap will change your numbers, but you won't fry it by accident. Just depends on how luck you feel. Most tweeters can take an accident at 1 W (2.838 V) though. Accuracy vs. cost of a new tweeter.... your gamble.
Test leads with alligator clips.
AC Voltage Meter / VMM with at least 2 decimal places of accuracy.
Steps:
For all measurements measure IN the cabinet, and in place. Put microphone 3' away on tweeter access unless speaker is very close to room boundaries.
1 - Measure speaker acoustically and electrically using OmniMic + DATS. This will be your baseline.
If your speaker is ported, use the port to provide test leads directly to each driver. Otherwise, you'll need to carefully disassemble and connect the drivers to the inputs without moving the cabinet, even 1 mm is too much. Another good solution is if your speakers are already setup for bi-wiring. Just hook each driver directly. Lastly, if your speaker really is sealed, try cheating by slipping the test leads to the tweeter through the back plate where the speaker connectors are and connecting the woofer directly.
- With the speakers all disconnected, calibrate test amplifier output to 2.838 Volts, or as close as possible. In OmniMIc this is the 50Hz signal. Any good digital Voltage meter will do. Do not use this signal when the speaker is hooked up. This is just for your meter. Relative accuracy is more important for you than achieving exactly 2.828 V. If the closest you can get is 2.9, or 2.7, that's fine, just don't move any volume knobs after you start. If you can get 2.828 exactly this makes it easier to directly compare your sensitivity to other speakers, but, meh! Just don't change the volume in the middle of testing and your crossover will work out fine.
- Measure tweeter impedance. The impedance of the tweeter can be measured in or out of the cabinet if you find it more convenient.
- Attach woofer only directly to inputs. Measure impedance and frequency response.
- Attach tweeter only via test capacitor to inputs. Measure acoustics.
- Place tweeter and woofer in parallel. Measure acoustics. Save this as your "interferometry" baseline. You'll use it for speaker distances later.
Crossover Simulation
- Grab XSim, from Bill Waslo. Drop down a tweeter and woofer, and import the individual impedance and frequency response (FR) data.
Interferometry
We must calculate the correct tweeter to woofer distance to finish calibrating the simulation.
Draw a schematic for the circuit using the tweeter and woofer. Import the baseline to the FR chart. Notice the FR won't match the "interferometry" baseline. Increase woofer delay until it matches as much as possible. Probably going to be an inch or so.
Once all that matches, you are ready for crossover simulation.
Best,
Erik
However you may use any other tool and mic combination. You'll need both electrical (impedance vs. frequency) and acoustical (frequency response vs. db amplitude) measurements.
Supplies:
A bipolar (or better yet, film) 20uF capacitor to protect the tweeter. Don't spend more than $5. You can forego the cap if you use test signals without bass and you feel lucky. The cap will change your numbers, but you won't fry it by accident. Just depends on how luck you feel. Most tweeters can take an accident at 1 W (2.838 V) though. Accuracy vs. cost of a new tweeter.... your gamble.
Test leads with alligator clips.
AC Voltage Meter / VMM with at least 2 decimal places of accuracy.
Steps:
For all measurements measure IN the cabinet, and in place. Put microphone 3' away on tweeter access unless speaker is very close to room boundaries.
1 - Measure speaker acoustically and electrically using OmniMic + DATS. This will be your baseline.
If your speaker is ported, use the port to provide test leads directly to each driver. Otherwise, you'll need to carefully disassemble and connect the drivers to the inputs without moving the cabinet, even 1 mm is too much. Another good solution is if your speakers are already setup for bi-wiring. Just hook each driver directly. Lastly, if your speaker really is sealed, try cheating by slipping the test leads to the tweeter through the back plate where the speaker connectors are and connecting the woofer directly.
- With the speakers all disconnected, calibrate test amplifier output to 2.838 Volts, or as close as possible. In OmniMIc this is the 50Hz signal. Any good digital Voltage meter will do. Do not use this signal when the speaker is hooked up. This is just for your meter. Relative accuracy is more important for you than achieving exactly 2.828 V. If the closest you can get is 2.9, or 2.7, that's fine, just don't move any volume knobs after you start. If you can get 2.828 exactly this makes it easier to directly compare your sensitivity to other speakers, but, meh! Just don't change the volume in the middle of testing and your crossover will work out fine.
- Measure tweeter impedance. The impedance of the tweeter can be measured in or out of the cabinet if you find it more convenient.
- Attach woofer only directly to inputs. Measure impedance and frequency response.
- Attach tweeter only via test capacitor to inputs. Measure acoustics.
- Place tweeter and woofer in parallel. Measure acoustics. Save this as your "interferometry" baseline. You'll use it for speaker distances later.
Crossover Simulation
- Grab XSim, from Bill Waslo. Drop down a tweeter and woofer, and import the individual impedance and frequency response (FR) data.
Interferometry
We must calculate the correct tweeter to woofer distance to finish calibrating the simulation.
Draw a schematic for the circuit using the tweeter and woofer. Import the baseline to the FR chart. Notice the FR won't match the "interferometry" baseline. Increase woofer delay until it matches as much as possible. Probably going to be an inch or so.
Once all that matches, you are ready for crossover simulation.
Best,
Erik
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.