Need help reading blue prints for wood horns.

Hi All. I just bought a wood lathe and want to build a pair of 1Khz Yuichi horns. On the blue print there is a center line that is ticed off at some increment but I can't figure out what the increment means.

See the pic below. The increment starts at 4 and ends at 16. I've been reading blue prints for years and have never seen where a segment is not marked for length.
 

Attachments

  • Untitled.jpg
    Untitled.jpg
    94.3 KB · Views: 192

dumptruck

Member
2010-05-04 5:02 pm
MN
I thought I'd have a go at doing this drawing to scale, and it doesn't seem to work out at all? The 104mm line doesn't reach the outer radius if those are supposed to be cm up the middle. (edit: whoops, made a mistake, fixed, but same problem still)
 

Attachments

  • temp.gif
    temp.gif
    140.2 KB · Views: 166
Last edited:

dumptruck

Member
2010-05-04 5:02 pm
MN
Hmm, the distance from the focal point to the beginning of the flare is 34 mm and is [3] segments long, so seems it's 34/3 = 11.3333 mm between segments, but in the metric world this begs the question: 'why'?
Heh, well, I've still got this drawing open and that's easy to do... It doesn't seem to work out any less odd. I'm not sure where you're getting 3 segments for that, though, as no '0' mark is definite. Meanwhile, the 20mm deep section is clearly two segments, bringing it back to 1cm per segment.
 

Attachments

  • temp.gif
    temp.gif
    141.7 KB · Views: 142
Last edited:
Hmm, the distance from the focal point to the beginning of the flare is 34 mm and is [3] segments long, so seems it's 34/3 = 11.3333 mm between segments, but in the metric world this begs the question: 'why'?

GM

But then look at the segment between 4 and 6. Off to the left it indicates it is 20mm. So the segment between the crosshairs and 4 must be 14mm.

I just added to the CL with section 17, 18, and 19. Making the assumption that the section between the crosshairs and 4 is 14mm, and the sections are 10mm each, the total is 170mm just like the radius.
 
Hmm, the distance from the focal point to the beginning of the flare is 34 mm and is [3] segments long, so seems it's 34/3 = 11.3333 mm between segments, but in the metric world this begs the question: 'why'?

GM
I'm trying to read it (in my metric universe) and doesn't show the starting point for segment #4. It should be/read 40 (something) if 1x segment = 10 (mm or something). (10 mm = 1 cm)
It shows only starting/ending point for measurement 34 (mm).
This for some reason shows that the source/starting point for segments #1, #2, #3, #4... it's not the same as the measurement (or segment) of 34 mm. In my logic world 1x segment #4 = 40 mm, away from "segment 34". Don't ask me why... :(
 
The image attached in post #1 is not quite to scale. Except for one of the radial dimensions of the horn curve, the numbers seem consistant (the red segment shows where the curve is concave).

Dimensions are in mm except for the axis which is in cm. the centre of the arcs is at 2.6 cm

attachment.php



dave
 

Attachments

  • Yuichi-horn.gif
    Yuichi-horn.gif
    23.2 KB · Views: 180
Heh, well, I've still got this drawing open and that's easy to do... It doesn't seem to work out any less odd. I'm not sure where you're getting 3 segments for that, though, as no '0' mark is definite. Meanwhile, the 20mm deep section is clearly two segments, bringing it back to 1cm per segment.

Oh well........ From the drawing. I did say from the focal point, i.e. the radius arc and there's only one...... Yes, I noticed the discrepancy, but didn't know if the drawing was to scale, but figured you'd try it on your drawing, so nothing ventured, nothing gained. ;) Anyway, it wouldn't be the first drawing ever done in multiple scales for tooling purposes or drawn over a scaled grid that I'm not familiar with.

GM
 
Unfortunately, this is all they have for measurements. But here is the full print.

Why not just print out these plans full scale so that throat measures 25.5mm on the physical print. Then use a pair of calipers to carefully measure what the actual dimensions are every 1cm out. Seems like the best way to eliminate guess work, assuming the drawings are accurate of course.