I started having a look at box-planning the TH SPUD this weekend and it's not going to be as easy as it seemed at first. It's that S3-S4 section (see image). It's not a parabolic expansion, but an approximated expansion (conical?). It's possible to optimize the expansion, but it's going to take a bit more work. I've already got the panels programmed in the Excel spreadsheet - now I've got to work out how to work out the optimized angles and lengths for each panel highlighted in the image.
Attachments
The sections of the horn from there from S3 to S4 are actually several parabolic sections that will have to be approximated by different expansion, most likely conical. Unless we can get David to change Hornresp so it can model more than four sections....
Interesting that this TH SPUD has a parabolic model.
The Tangband 19Hz Tapped Horn – Volvotreter Homepage
The Tangband 19Hz Tapped Horn – Volvotreter Homepage
This is what the expansion looks like if you unfold the horn and ignore the extra volume in the corners. All sections are parabolic, but look at what's happening between S3 and S4. That can't be modeled in Hornresp as a one parabolic section, because it will end up being too large in volume (bright yellow section in the image). Modeling S3-S4 as a Conical section should reduce the error a bit.
Attachments
Hmm... that just might work out- just have the expansion angle added to each internal piece, e.g. if the first horizontal panel is at 2 degrees, the next horizontal angle should be at 2+2=4 degrees, the next one at 4+2=6 degrees, etc. That will keep the expansion angle through those sections constant, allowing it to be treated as one parabolic section.
I'll have a look at it later...
Well, my first instinct was right - this isn't a fold that can be modeled as a simple three or four section parabolic expansion. For that to happen, the yellow panels in the picture will have to be parallel to each other, and the red panel would have to be parallel to the top or bottom of the enclosure, and that's clearly not the case in the illustration.
I'm going to adjust the layout of the panels in the enclosure to be two parabolic expansions joined together at S3, and then once I've got that working I'll see if I can fiddle with the model a bit to see if I can achieve the panel layout of the actual build using an approximated conical expansion.
I'm going to adjust the layout of the panels in the enclosure to be two parabolic expansions joined together at S3, and then once I've got that working I'll see if I can fiddle with the model a bit to see if I can achieve the panel layout of the actual build using an approximated conical expansion.
Attachments
