Hi everyone. This is my first post so please go easy on me.
I am building a pair of large transmission line speakers to replace my elderly TDLs. Rather than try to create a TL from scratch I decided to use a tried and tested design. I have heard the TL12 at Wilmslow's premises, playing many of my CD "warhorses" and I was very happy with the sound. I also recorded a sine sweep with Omnimic and the response curve is quite similar to my TDLs. So I am going to build their TL15 cabinet for the following drivers:
Volt RV3863 bass (8 ohm)
Volt VM752 mid (8 ohm)
Scanspeak D2904/710003 treble (4 ohm)
Drive units will be tri-wired to an external passive crossover, making component changes easier
Cabinets will come first including stuffing. Then mount the drivers for testing. I have Omnimic and DATS for this.
The plan is to measure each driver in turn using the Omnimic fast sine sweep to create FRD and ZMA files, then use x-sim to model the crossover.
When the crossover is built I would measure again to check the result.
I would welcome advice about the initial measurement as the Omnimic manual is vague on how exactly to set it up. I plan to measure at 1 metre on-axis for each driver with the same amplifier setting (using a tripod mount with a long vertical adjuster to keep the distance constant). Omnimic states I also need to measure the full range with all drivers but at 1 metre only one driver can be properly on-axis. I presume I position the microphone on the tweeter axis but why is this measurement necessary?
I know the tweeter and mid need a high-pass protection filter for the initial tests. Any suggestions? Single series capacitors? What values? Or steeper filters? If so, any suggestions?
Any other comments would be welcome. I have never done this before so I must plan every step carefully. As my woodwork teacher told me, "measure twice, cut once".
I am building a pair of large transmission line speakers to replace my elderly TDLs. Rather than try to create a TL from scratch I decided to use a tried and tested design. I have heard the TL12 at Wilmslow's premises, playing many of my CD "warhorses" and I was very happy with the sound. I also recorded a sine sweep with Omnimic and the response curve is quite similar to my TDLs. So I am going to build their TL15 cabinet for the following drivers:
Volt RV3863 bass (8 ohm)
Volt VM752 mid (8 ohm)
Scanspeak D2904/710003 treble (4 ohm)
Drive units will be tri-wired to an external passive crossover, making component changes easier
Cabinets will come first including stuffing. Then mount the drivers for testing. I have Omnimic and DATS for this.
The plan is to measure each driver in turn using the Omnimic fast sine sweep to create FRD and ZMA files, then use x-sim to model the crossover.
When the crossover is built I would measure again to check the result.
I would welcome advice about the initial measurement as the Omnimic manual is vague on how exactly to set it up. I plan to measure at 1 metre on-axis for each driver with the same amplifier setting (using a tripod mount with a long vertical adjuster to keep the distance constant). Omnimic states I also need to measure the full range with all drivers but at 1 metre only one driver can be properly on-axis. I presume I position the microphone on the tweeter axis but why is this measurement necessary?
I know the tweeter and mid need a high-pass protection filter for the initial tests. Any suggestions? Single series capacitors? What values? Or steeper filters? If so, any suggestions?
Any other comments would be welcome. I have never done this before so I must plan every step carefully. As my woodwork teacher told me, "measure twice, cut once".
Thanks. That helps a lot with the measurements.
Does anyone have an answer to my other question:
I know the tweeter and mid need a high-pass protection filter for the initial tests. Any suggestions? Single series capacitors? What values? Or steeper filters? If so, any suggestions?
When I run a sine sweep with the Omnimic CD I can select "bass removed" or "bass and mid removed" but then I need to run a sine sweep for the whole system, which will be full-range and could damage the midrange driver and tweeter. However, surely adding capacitance would change the response and introduce phase shift?
The Volt VM752 has an Fs of 400Hz and the Scanspeak D2904/710003 Fs is at 520Hz. Clearly it's important not to drive them as low as that.
I expect the final crossover points will be at 500Hz and 3,500Hz (as Wilmslow say their crossover is for the same drivers).
So what do I put in series with the mid and tweeter when I generate the FRD and ZMA files?
Does anyone have an answer to my other question:
I know the tweeter and mid need a high-pass protection filter for the initial tests. Any suggestions? Single series capacitors? What values? Or steeper filters? If so, any suggestions?
When I run a sine sweep with the Omnimic CD I can select "bass removed" or "bass and mid removed" but then I need to run a sine sweep for the whole system, which will be full-range and could damage the midrange driver and tweeter. However, surely adding capacitance would change the response and introduce phase shift?
The Volt VM752 has an Fs of 400Hz and the Scanspeak D2904/710003 Fs is at 520Hz. Clearly it's important not to drive them as low as that.
I expect the final crossover points will be at 500Hz and 3,500Hz (as Wilmslow say their crossover is for the same drivers).
So what do I put in series with the mid and tweeter when I generate the FRD and ZMA files?
Help! Crossover low impedance
I used Omnimic to create frd and zma files for all three drivers, plugged the files into x-sim and tweaked the compnent values to get the best response curve. The VM752 must be crossed over above its Fs of c.400Hz but that needed a large parallel capacitor in the HP circuit to the woofer. Now I have an imedance curve down to 2 Ohms around 300Hz and reducing it ruins the response from the Volt bass driver.
Any ideas where I'm going wrong?
PS I took a screenshot from x-sim but can't attach it.
I used Omnimic to create frd and zma files for all three drivers, plugged the files into x-sim and tweaked the compnent values to get the best response curve. The VM752 must be crossed over above its Fs of c.400Hz but that needed a large parallel capacitor in the HP circuit to the woofer. Now I have an imedance curve down to 2 Ohms around 300Hz and reducing it ruins the response from the Volt bass driver.
Any ideas where I'm going wrong?
PS I took a screenshot from x-sim but can't attach it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.