High Performance/Low Cost Sails for
Small Sailboats
DIRECTIONAL STRETCH AND RESILIENCY IN POLYTARP
COMMONLY USED FOR SAILMAKING
Introduction
I have personally been using polytarp for a sailmaking material for over 14 years and still marvel at what I dont know about it. Unlike traditional sailmaking materials, polytarp seems to require very little shaping to make a functional sail. It also seems to be very strong, often breaking masts instead of failing itself. In fact, in 14 years of selling polytarp kits to others and building various sails from polytarp, I have only received feedback on a couple of instances of material failure. In one case the sail went through a tornado and received damage as it was whipped time and time again against the mast. In that case the customer made duct-tape repairs and continued to use the sail. In the other case in my first year of business, a customer mentioned that he had been out in a furious storm and had his sail fail although he never clearly identified whether it might have been a material failure or a tape failure or some other failure. He was clearly not disappointed in the material, though, blaming himself for the problem and refusing the offer of a replacement kit.
There
have been some fairly large sails made from polytarp
and the sail size record continues to grow. For a long time Chuck and Sandra Leinwebers Caprice carried
the largest spread of polytarp sails I knew of. They eventually replaced the
tarp sails with tanbark sails from the sailmaker they
use, but I recall seeing an email to one of their customers that said they could
not claim any improvements in performance for their new sails over
their PolySails. Ive
personally made a 280 sq. ft. lug sail for a customers 27 Roberts, but I would definitely have to
expand my workspace and improve my sewing machine to go larger. However, that
could happen. Just before writing this piece, I received an email inquiring
whether I thought a 554 sq. ft. Chinese lug could be built
from polytarp to power a 30 yacht. I answered yes because I think I now have
the data to support this answer.
I have suspected for some time that some types and weights of polytarp are far stronger and more resilient than most sailors realize. I also knew from experience that the material seemed to lend itself well to sailmaking, although I wasnt clear why. In 2006 polytarp and other sail material and architectural fabrics were tested for fabric tear strength with an Instron machine at the WL Gore factory (Goretex). The results showed that a 1 strip of 5.2 oz. polytarp compared favorably with Dacron and several other well-known sailmakers synthetics in initial strength. However, this test did not answer questions I had about the directional stretch and resiliency of the material.
Participation in the PDRacer forum at http://groups.yahoo.com/group/pdracer/ finally stimulated me to try further testing through an experimental design first suggested, I believe, by Australian designer and boatbuilder Michael Storer. He suggested that we might learn a great deal about polytarp by suspending strips of the material and stretching the strips with successively greater weights. Thanks to his suggestions and those of others in this very inquisitive group, I followed their lead and conducted the necessary tests to collect original data on directional stretch and resiliency in various weights of polytarp. My grandson Jackson Bell assisted me.
Procedure
I first cut 2 x 2 squares of polytarp from the corners of different pieces of polytarp. The three tarps chosen weighed 3.1 oz/sq. yd., 5.2 oz/sq. yd., and 5.5 oz/sq. yd., according to information listed by their manufacturers and/or suppliers. From the squares I cut equal strips measuring 1.5 x 20 from the long side, the wide side, and the 45 degree diagonal of each tarp square. I used strips of 1.5-wide tape to help cut the strips to a consistent width, then the tape strips were removed. Next, I made a 1 fold in the ends of each of the strips and placed a grommet through the end folds in the same place in each strip, thereby shortening each strip to 18 (plus or minus 1/32). Each 1 ½ x 18 strip measured 27 sq. in. overall.
After
carefully checking the dimensions of all test strips to see that they were
identical in length and width, I suspended each strip from a hook through the
upper grommet, then hooked an 8 oz. bucket through the
lower grommet and measured the strip length once again. This length served as
the base length as I checked the strips for stretch. Next, I filled the bucket with
quarts of water and measured the length of each strip again after each quart
(32 oz. or 2 lb.) after each quart was added. After
the fourth quart or 1 gallon had been added to the
bucket and the strip measured, the bucket was emptied, rehung,
and then the strip measured again after 30 seconds to check the strip for
resilience or recovery. 
The next step was to take the bucket down and pour two gallons or 16 lbs. of water into the bucket. I then weighed he bucket with the water in it twice to make certain that the total weight of the water and bucket totaled 16.5 lb. before again suspending the bucket from the strip then measuring the stretch after 30 seconds. Often I could see the strip continuing to stretch slightly beyond the measured point beyond the 30 seconds. However, after one diagonal strip failed at the grommet while I was trying to photograph this stretching, I did not go beyond this time with other strip testing. Next, the bucket was emptied, rehung on the strip, and again remeasured for resiliency. Finally, at least three hours later, all strips were measured on the ground with no weight attached, to check once again for stretch and resiliency.
Strips
showing the result of stretch after the first two tests.
After the results of the first two tests, I concluded that the
composition of the film/scrim/film within each polytarp weight category might
be an important factor in helping to explain stretch and resiliency
characteristics of different weights of tarps. Consequently, I observed
representative tarps from which strips were taken
under a magnifying glass to establish the number of scrim strands per square
inch and whether the scrim appeared tightly or loosely woven. I also recorded
the country of origin for each tarp used in the samples. It might also have
been useful to verify the actual weights of the tarps that were
identified by the manufacturers/ suppliers. After observing well over
1000 tarps over the years, I believe there is considerable variation in the
actual weights of tarps of the same size as reported by manufacturers.
Sometimes these weights appear to vary even for a single lot of tarps from the
same manufacturer. I have also noticed that material quality varies
considerably by manufacturer and supplier. I suspect ,
for example, that the tarp identified as 5.5 oz. in this test might have been
identified as 6.0 oz. by another manufacturer. Similarly, the
5.2 oz. tarp might have been identified as a 4.5 oz. tarp by a more
reputable manufacturer. On the other hand, warranties provided by the
manufacturer of the lighter weight material lead me to believe the 3.1 oz.
quoted weight is accurate.
On the following pages are the recorded observations and results of the three tests. Note that I added strips of edge material to two of the tests just because I often use strips of this edge material for reinforcement material in corners, and I wanted to see how it compared to the other strips from the same tarp.
Photo shows the relative stretch of various test
strips of polytarp after being stretched by a 16.5 lb.
load.
Data Sheets
|
TEST 1 Test 1 was conducted on 1 ½ x 18 (finished length plus
or minus 1/16) strips from 5.2 oz.
polyethylene having a film/scrim/film composition made in |
||||
|
Measure |
Long |
Wide |
Diagonal |
Long Edge Strip |
|
Empty (8 oz.) |
18 |
18 |
18 |
NA |
|
1 qt. (2 lb.8 oz.) |
18 1/16 |
18 1/16 |
18 1/8 |
NA |
|
2 qt (4 lb. 8 oz) |
18 3/16 |
18 3/16 |
18 Ύ |
NA |
|
3 qt. (6 lb.8 oz) |
18 7/16 |
18 5/16 |
20 3/8 |
NA |
|
1 gal.(8 lb. 8 oz) |
18 9/16 |
18 ½ |
21 5/8 |
NA |
|
Emptied (8 oz.) |
18 5/16 |
18 Ό |
19 7/8 |
NA |
|
2 gal. (weighed, 16 lb. 8 oz.) |
19 |
19 risk of failure |
23 5/8 failure |
NA |
|
Emptied again (8 oz.) |
18 7/16 |
18 5/8 |
-- |
NA |
|
3 hr. later (0 oz.) |
18 ½ |
18 ½ |
20 5/8 |
NA |
|
|
||||
|
TEST 2 Test 2 was conducted on 1 ½ x 18 (finished length plus
or minus 1/16) strips from 3.1 oz.
polyethylene having a film/scrim/film composition made in |
||||
|
Measure |
Long |
Wide |
Diagonal |
Long Edge |
|
Empty (8 oz.) |
18 |
18 |
18 1/32 |
18 1/16 |
|
1 qt. (2 lb.8 oz.) |
18 1/16 |
18 1/16 |
18 1/8 |
18 1/8 |
|
2 qt (4 lb. 8 oz) |
18 1/8 |
18 1/8 |
18 7/16 |
18 3/16 |
|
3 qt. (6 lb.8 oz) |
18Ό |
18 5/16 |
18 7/8 |
18 3/16 |
|
1 gal.(8 lb. 8 oz) |
18 3/8 |
19 |
19 3/8 |
18 3/16 |
|
Emptied (8 oz.) |
18 1/8 |
18 5/16 |
18 5/8 |
18 1/8 |
|
2 gal. (weighed, 16 lb. 8 oz.) |
19 |
21 risk of failure |
22 risk of failure |
18 Ό |
|
Emptied again (8 oz.) |
18 1/2 |
19 Ύ |
20 Ό |
18 1/8 |
|
3 hr. later (0 oz.) |
18 3/8 |
19 Ό |
19 Ύ |
18 1/8 |
|
|
||||
|
TEST 3 Test 3 was conducted on 1 ½ x 18 (finished length plus
or minus 1/16) strips from 5.5 oz.
polyethylene having a film/scrim/film composition made in |
||||
|
Measure |
Long |
Wide |
Diagonal |
Wide Edge |
|
Empty (8 oz.) |
18 1/32 |
18 |
18 1/16 |
18 1/16 |
|
1 qt. (2 lb.8 oz.) |
18 1/16 |
18 1/16 |
18 3/16 |
18 1/8 |
|
2 qt (4 lb. 8 oz) |
18 1/8 |
18 1/8 |
18 3/8 |
18 3/16 |
|
3 qt. (6 lb.8 oz) |
18 3/16 |
18 7/32 |
18 13/16 |
18 1/4 |
|
1 gal.(8 lb. 8 oz) |
18 1/4 |
18 9/32 |
19 3/8 |
18 9/32 |
|
Emptied (8 oz.) |
18 1/32 |
18 1/16 |
18 1/2 |
18 1/8 |
|
2 gal. (weighed, 16 lb. 8 oz.) |
18 7/16 |
18 5/8 |
22 11/16 |
18 3/8 |
|
Emptied again (8 oz.) |
18 1/8 |
18 3/16 |
19 15/16 |
18 3/16 |
|
3 hr. later (0 oz.) |
18 |
18 |
19 3/8 |
18 |
|
|
||||
Conclusions
1. Calculated Loads
With this data in hand, I was able to say that based upon the stretch and resiliency of the 5.5 oz. tarp that my email contact with the 554 sq. ft. Chinese lug sail could expect an 81 sq. ft panel of his sail to support a calculated load of nearly 7000 lbs. without failure so long as the load was equally distributed and the support grommets adequately reinforced. However, the person could expect increasing stretch, particularly on the diagonal, as loads got closer to the 88 lb./per sq. ft. limit that I calculated. Ultimately, this stretch could also lead to failure. How did I arrive at this conclusion? I took the data from the two-gallon test and extended it this way: 27 sq. in. will support 16.5 lbs. without failure. Extend this to 144 sq. in. or 1 sq. ft. by dividing 144 sq. in. by 27 sq. in. resulting in a multiple of 5.3333. Applying the 5.333 multiple to the 16.5 lb. load supported by the 24 sq. in. strip results in an 88 lb. load supported by a 1 sq. ft. piece of tarp. Multiplying 81 sq. ft. by 88 lb. equals a 7128 lb. calculated load for the 81 sq. ft. panel. Looking at it another way, I am saying that, if adequately reinforced at stress points, a 9 x 9 (81 sq. ft.) 5.5 oz. tarp with the load equally distributed over the surface could potentially support a load of this size. In reality there are many other forces that come into play when we are talking about tarp sails supporting wind loads of this size without failure, However, I think the principles are basically sound, and help explain why tarp sails are so strong for their weight and thickness.
2. Stretch Direction
One conclusion that I think is clear from the data from the heavier weight tarps is that the stretch of the long and wide strips are very close to the same. Whereas for the lighter tarp, there was a significant difference between the long and wide strips. This data would suggest that one could lay out the most-stressed edge of the sail either along the wide or long edge when constructing a sail from heavier polytarp. When constructing a sail from 3.1 oz. tarp, the data suggest a home builder should lay out the leech along the long side.
3. Sail Layout
The significant stretch of the diagonal strip over the long
side and wide side strips in all tests might explain why tarp sails many times
require no darts or broadseaming to make decent
sails. If the diagonal can be oriented so that it stretches out near the
forward edge of the sail, one could expect the polytarp sail to
virtually shape itself to the airflow. In making a lug sail, for
example, its a good idea to orient the leech along the long side, put plenty
of rounding in the head to both help with sail shape and allow for diagonal
stretch, and very slight rounding in the luff to account for the stretch that
will lie somewhere between the diagonal 
and
long side. Placing the weld or seam that occurs every 5 to 6 in tarps
toward the leech and parallel with it will help prevent that area from
stretching as much as the area nearer the luff.
With a triangular sail, operating on the same principle, it might be better to orient the sail so that the foot is along the wide side and the luff is more oriented toward the long side so that the high stretch diagonal radiates at close to a 45 ° angle from the tack. Slanting the luff somewhat inward toward the head when constructing a sail like the leg o mutton might make even more sense. That way the leech would be less along a diagonal and less likely to stretch out of shape. Of course the stretch on any side can be somewhat controlled by rounding, reinforcing line and overlap at the edges, as well as by placing the edge near a weld. However, as some have learned, if there is too much tension in the line in a leech that has the stretchy diagonal material perpendicular to it, there is a danger of creating a hooking leech.
4. Matching Polytarp Selection to Sail Needs
Comparisons
of the 5.2 oz. stretch and resiliency with the 3.1 oz. figures show some
unexpected results. The diagonal stretch on the lighter weight 3.1 oz. strip
was nearly 1 ½ less than the heavier
5.2 oz. strip that also broke near the grommet. As I pointed out earlier, part
of the reason for that break might have been because the grommet was not set all
the way down on the material, allowing extra stretch around the grommet hole. However,
I attribute the difference more to the composition of the scrim in the two
tarps. In the 5.2 oz. tarp strips there were lots of
gaps in the scrim visible under the magnifying glass whereas with the 3.1 oz.
material and the 5.5 oz. material, there were no noticeable gaps in the weave of
the scrim. That means there were many places in the 5.2 oz. strips where there
was only a film to film contact instead of a
film/scrim/film contact. Consequently, there might have been specific areas in
the 5.2 oz. strips that were weaker and more susceptible to failure than the
strip as a whole. In the end, I think it was the relatively
poor quality of the 5.2 oz. samples that made their results more like the 3.1
oz. results than the 5.5 oz. results. Does all this mean that a 5.2 oz.
tarp wont make a good sail? The answer is no. Because
of the tight weave in a heavier tarp, it might need darting to achieve the same
shape that a lighter tarp can achieve without darting even in relatively small
sails. On the other hand, a 5.2 oz. tarp might stretch far too much for a
larger sail over 100 sq. ft.
5. Resiliency of Polytarp
The final conclusion I think that can be drawn from these results is that polytarp, on the whole, is a very resilient material. With the lighter, more realistic loads that the 2.8 oz. loads represent (1 gallon of water in the bucket), nearly all the strips are likely to return to their original lengths when no longer loaded. It also means that polytarp sails can probably stretch well briefly to handle the higher loads that gusts bring and still retain their forms. All in all, I think it is this resiliency that makes better quality polytarp the best value among sail materials.
PolySail International
Email polysail@polysail.com or call Dave Gray at 317 385-3444
PolySailsSold on the Web since 1996. Customers in all 50 states and around the globe.
This page updated on 2/8/2010