Apr 6, 2012 at 7:15 am #1288379
I have a question that must have been asked hundreds of times before, but I can't find it. How much maximum deflection should a catenary curve have per given distance? I can find countless spreadsheets where I can enter the length of the ridge and the maximum deflection and it will spit out a nice curve, but how do you determine the maximum deflection? I'm using 1.1 oz silnylon and my seams are 164cm, 188cm and 155cm.Apr 6, 2012 at 7:49 am #1864324
Maybe 2.5 inches for an 8 foot diagnol length.
So, maybe 4 cm for 155 cm and 164 cm, 4.5 cm for 188 cm.
Are their 3 ridges in your tent?Apr 6, 2012 at 7:53 am #1864326
David DrakeBPL Member
@daviddrakeLocale: North Idaho
Here's what I did (after it was suggested on these forums sometime last year):
–Pinned up a long sheet of stiff paper on a wall (builder's kraft paper works well).
–Hung a length of 1/8" beaded chain corresponding to the length of the seam over the paper. Adjusted the tension of the chain until the deflection looked good. My longest seam was (IIRC) a little over 6' (roughly 200cm?), and deflection was roughly 1.5". Less deflection for shorter segments, obviously, but if you keep the tension more or less equal, that will happen automatically (ie, shorter span, identical tension=less deflection).
–Sprayed a line of spray paint over the chain, marking the paper with the chain's "shadow." Because beaded chain has little straight segments between the round balls, the line was quite accurate.
–Took down the paper, cut out the curve, used as template to cut fabric.
I like this method better than spreadsheet, CAD, etc, as I still would have had to plot out the curve on paper and cut out. I have access to a plotter, but printing costs were more than using the chain method. And plotting points by hand seemed like a much longer process than using the chain.Apr 6, 2012 at 8:00 am #1864329
I'm making a sidesleeper that sets up with 2 trekking poles. The actual ridge is 164 cm. The other seams are sloping sides.Apr 6, 2012 at 8:07 am #1864333
David, does it have to be a 1/8" chain or will any rope or cable do? I don't know if the weight of the cable is a factor here. I don't mind a bit of measuring, but your method may actually be the most accurate. On the other hand, won't the curve depend on how tight you pull the cable?Apr 6, 2012 at 8:58 am #1864352
John DonewarBPL Member
@newtonLocale: Southeastern Louisiana
Go to Lance Marshall's website…
…and click on the third item Catenary Curve Spreadsheet in the menu on the top left.
All you have to do is plug in your numbers and it calculates the deflection for you and plots the curve by giving you the deflection at multiple points along your ridge line.
You can increase or decrease the "Severity Factor" number to achieve the desired deflection. A smaller number gives you a steeper curve and a larger number gives you a flatter curve. The spreadsheet recommends starting at 400 which is what I believe I remember using when I made my tarp. I have noticed that the spreadsheet opens with 360 plugged into the Severity Factor box. I wonder if that is the number that Lance likes and uses. ;-?
It works great for me. Depending on your browser you may have to "enable editing" to be able to plug in your numbers.
NewtonApr 6, 2012 at 9:30 am #1864367
I have tried suspending a line and letting it droop, and also a spreadsheet, and I found the result using spreadsheet was better – smoother. But, either way works.
First, I lay out a straight line on the fabric. Then make a mark every 2 inches. Then carefully measure the deflection for each 2 inch mark and put a dot. Then just sew through the dots, or you draw straight lines between dots.
A template might be easier if you do many of the same curve, but I don't bother with this anymore.
Before you sew or cut, eyeball the curve from the end and you can see if you screwed up on any points.Apr 6, 2012 at 1:04 pm #1864439
David DrakeBPL Member
@daviddrakeLocale: North Idaho
For me, the weight and flex of the beaded chain was key, plus the fact that each bead is joined by a wire-thin link. I think cable would be too stiff, and string too light. Rope is too thick to get an accurate line. Consistent and perpendicular angle of the spray can is also important. I found the method accurate enough for my silnylon tarps (now that I think about it, the deflection was closer to 2" over 6-ish feet, similar to what Jerry recommends). But I've never tried the spreadsheet for comparison.
For equal spans (ie straight-line distance) the length of the chain controls the deflection. Making the chain longer or shorter is pretty much in how tight you pull it, although the chain is basically inelastic over those lengths. I used the beads as a kind of measure, ie, fasten one end, then fasten the free end one bead shorter, or one bead longer to adjust the curve. Easier to draw than describe in words…Apr 6, 2012 at 1:29 pm #1864455
Now that you mention it, I tried using a fairly light string, which didn't work so good. The curve wasn't totally smooth along the length.
If I had used a chain, maybe it would have worked better.Apr 6, 2012 at 5:37 pm #1864547
@redolearyLocale: Mid West
It may not be a perfect catenary, but you could just use a flexible batten (like a piece of 1/2" pvc). Take the length of the fabric, say 72", mark the center, then measure in 4-6 % of the lenght (so maybe 3 or 4 inches) and flex the batten until you hit the the mark, trace the curve and you're done. A 10' piece of 1/2" pvc is only a couple of dollars.Apr 7, 2012 at 5:39 am #1864661
Thanks for the advice everybody! I ended up using a spreadsheet and making cardboard cutouts. Even if it's only for one seam, I find that making a cutout is worth it. Measuring and drawing on silnylon is a PITA. Cutting of some of the panels has already begun!Apr 7, 2012 at 6:51 am #1864669
The only problem with a cutout is it can flex a little making the cat curve deflection more or less – if you're aware of this you can check that it isn't happening.
I think any smooth curve, like from a PVC pipe, would work fine.
You must be logged in to reply to this topic.