Physical T-M time alignment includes the crossovers. If you mount your T-M on a flat baffle with your early crossover some square wave tests would show the delay on an O'scope. You can also send a square wave to each driver separately and measure the time delays to subtract out the separation delay. A small total alignment delay might be covered by a commercial 0.75" - 1" dome tweeter with an integrated waveguide like the Tymphany BC25SC08-04 or Tang Band 25-302SH or larger Tymphany H26TG45-06(previous picture). A quarter-round router bit creates a spherical curve, which is a good shape for a dome exit.
I know this thread is now some 4years old but was hoping someone could shed some light on the 3W discovery crossover component values. I see that system7 has tried to replicate them, but it looks like the images of the crossover design are now gone? Does anyone know crossover value? Thanks in advance, Brett
No, No, No. I wouldn't publish even my notions of the values in the ScanSpeak-3W-Discovery.
That would be robbing him of his livelihood. All Troels' children starving. Unable to afford medical bills for his momma...
What I was saying is that armed with the Steen Duelund filter theory, you find most three ways start to look the same.
Minor complications include bafflestep and tweeter time alignment. But Duelund filters line up phase and amplitude very well indeed. This was a similar project in Visaton Boxsim:
What it's saying is you really just have to replicate the slopes, adjust for loudness, impedance and frequency response and maybe allow for time-alignment, and they are really all the same. Below are the missing parts of the theory which imageshack let me down on. Hope it helps.
That would be robbing him of his livelihood. All Troels' children starving. Unable to afford medical bills for his momma...
What I was saying is that armed with the Steen Duelund filter theory, you find most three ways start to look the same.
Minor complications include bafflestep and tweeter time alignment. But Duelund filters line up phase and amplitude very well indeed. This was a similar project in Visaton Boxsim:
What it's saying is you really just have to replicate the slopes, adjust for loudness, impedance and frequency response and maybe allow for time-alignment, and they are really all the same. Below are the missing parts of the theory which imageshack let me down on. Hope it helps.
Attachments
Hi Steve, thanks for that. That has really helped me a lot. I've read all the articles, all makes sense really! I noted that troels' has included 2 capacitor values on the build page, so I think I'm on the money with all the values now, I started with the two known values and worked backwards from there, then tried to match his posted response. Boxsim really does make life easy, doesn't it next is to build it and measure the real world results
next is to build it and measure the real world results- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.