Da Towa of Testin'!

Another post to break up the mechanics posts. This time, about some of the test equipment I've been setting up. In this post, we'll take a look at my on-going construction of a drop testing tower, mainly for testing garment samples based on the tests outlined in CEN EN13567 (The standard that FIE homologation is based on. See these posts for more details).

Da Towa pities the fool. Note the manly tool bag.

I really have to blame one friend in particular for this tower, since it came up in the course of a discussion about testing the penetration of textile armor. The object was simple, at first: use a drop tower arrangement to see how much kinetic energy was required to penetrate a sample. The idea went like this in my head:

1. Build a tower with a penetrator that was kind of like a sword point.
2. ???
3. Prof... I mean get results.

Well, I sat down to figure out how I could actually get useful information out of something I could build in my backyard without access to any of the fancy stuff I can use at work (like a budget that isn't me and my wife's bank account). During the course of my search, I decided I'd see what I could find out about how the FIE tested fencing clothing, which led to several earlier posts. Well, as it ended up, they described a configuration I could actually use to develop my own: a drop tower with a falling sample holder of a known weight, and a penetrator with an attached force transducer to measure the forces during the impact. This was basically the big (and more scientific) brother of a simple drop tester described for testing SCA rapier garments.

Da Towa is born

The operating principle of a drop tower is exactly what it says on the tin: you drop stuff, onto other stuff. Simple, versatile and remarkably effective.

My idea became to build a PVC tube to act as a guide for a falling mass, into which I could put a fabric sample. At the bottom of the tube would be a chisel or a punch that would act as the penetrator. Eventually, I'd be able to attach a force transducer and a data acquisition system to the penetrator and get some Really Useful Data. At the top, I could put a pulley and cable system so I could raise and drop the falling mass from different heights, as measured by a tape measure attached to the side.

How hard could it be?

Making a sample holder

The first thing I needed to figure out was if I could construct a sample holder that could reliably hold a garment sample. Because if I couldn't actually hold a sample, there would be no point in building anything else. The sample holder described in CEN EN13567 would need to be custom made, since it involves a particular locking ring arrangement meant to keep fabric from slipping when it hits the penetrator at over 6 m/s and 90 J. While I have the knowledge to make the thing, I lack access to the equipment and didn't want to pay for a custom machined part.

So I headed down to my local Home Depot and became that dude hunched over in an aisle that the clerks left alone, because they wanted no part of what he was up to... whatever it was. It certainly wasn't plumbing. As an aside, this is really one of the best parts of mechanical engineering: I've had to do it for school, work and now for my own pet project.


I was finally able to settle on an arrangement of pipe and conduit fittings I liked, that were about the right size compared to the EN13567 sample holder. For reference, mine holds a 58.4mm diameter sample with an exposed area 34.5 mm in diameter, the EN13567 holder is a 68 mm diameter sample with a 35.7 mm diameter exposed area.

Sample holder, first concept.

What you can't see in the above image is the arrangement of steel conduit lock washers and conduit reducing bushings that would lock down on the fabric sample when the pipe bushings (the large dark colored bits on either side of the shiny coupler) were tightened:

Lock washers and reducing washers as used in first concept.

Since WMAW had passed, I decided that my frankengambeson had lived a good life and it was time for it to donate its body to science. Using samples cut from it, I convinced myself that this washer arrangement would hold a sample in the gorilla approved manner:

Yep, I smacked it with a really heavy hammer. Said beastly hammer is seen to the right.

But as it ends up, later testing would show that the lock washers had a nasty tendancy to shift or cut into the fabric sample, possibly affecting the failure mechanism by letting the fabric slip. So I ended up just making use of more of the reducing washers, whose handy little lip appeared to be able to clamp down well enough to not let even 2 layers of 10 oz canvas slip:

Handy little lip on the reducing washers.

Another thing I learned in my initial testing was that the way I had mounted the eye hook wasn't well enough centered and caused some wobble in the falling mass, so I had to revisit that as well. I ended up cobbling something together out of PVC pipe fittings and some glue that would ensure the eye bolt stayed closer to center. I also added a section that I could fill with copper BB's or other stuff to add mass (the EN13567 falling mass had to weigh 5 kg-f, but I wanted mine variable). The result was my second sample holder concept:

Sample holder, second concept. Note duct tape: this one has to work better.

With this method, I have been able to get the sample holder up to a total weight of 3.65 kg-f, while still being able to have a drop distance of over 1.9 m (from the point of the chisel to the bottom surface of the fabric sample). This height is sufficient to reach 6.1 m/s, and an energy of 67.8 J. Standard EN13567 requires a speed at impact of 6-8 m/s and an energy of at least 90 J, so I can get the right speed but am still short on energy due to the lighter mass of my sample holder.

The second concept parts are as follows, for a cost of roughly 30 USD (not including the BBs). Note that the sizes quoted are not the actual dimensions, but the sizes the parts are sold as (a 2" conduit coupling is larger than 2" in diameter).

• 2" rigid electrical conduit coupling (1x)
• 1-1/4" to 1" rigid electrical conduit reducer washer (8x)
• 2" to 1-1/2" pipe hex reducing bushing (2x)
• 1-1/2" to 1-1/4" pipe hex reducing bushing (1x)
• 1-1/4" x 6" (and 10") pipe nipple (1 ea)
• 1-1/4" PVC female threaded pipe adapter (1x)
• 1-1/4" to 1/2" PVC reducing bushing  (1x)
• eye bolt and some washers

The tower body

This part was actually the easiest bit. From the minimum speed in EN13567 of 6/ms, and setting the kinetic energy at impact equal to the potential energy at the maximum height, I was able to calculate a minimum tower height (1.84 m). My garage height is closer to 2.6 m, so I decided to go a bit longer to allow for the sample holder and a higher possible speed.

It was then back to Home Depot with my sample holder to figure out what size pipe would work. As it ended up, 3" schedule 40 PVC pipe was about the right inner diameter. So I grabbed a 10' section of it.

The body I cut to 8 feet in length. Because I knew I'd want some sort of door to make changing the sample and penetrator easier, I also cut out a section that I used to make a hatch I could open. The resulting body looked like:

Imagine like 8 feet of this.

To construct the ends, I grabbed two 3" to 1-1/2" PVC reducing couplers and two 1-1/2" to 1/2" PVC reducing bushings. At the top, one of these reducing couplers with a reducing bushing jammed into it would act as the guide for the thin steel cable that I decided to use to raise and lower the sample holder. At the bottom, the other would help form part of the penetrator holder.

The penetrator and its holder

Since I'd found that 3" schedule 40 PVC was what I was going to build the body out of, and I found the smallest reducing couplers I could at the local store, I had to figure out how to hold the penetrator.  But first I had to figure out what exactly that penetrator would be. I knew one thing though: it would be fairly small (the sample impact area is 34.5 mm in diameter), like 1/2" or less.

I couldn't easily reproduce the penetrator in EN13567. The standard describes something I'd have to custom machine and harden... not happening in my basement/garage. Instead, I opted for cold chisels and pin punches. The size choice was basically 'find what I can that's close to either a sword point or the 3mm penetrator in EN13567'. My first choice was a 1/2" cold chisel, whose outer diameter at the body was about 3/8". This was a bit too small to nestle into the center of a 1/2" piece of PVC, and I didn't actually want a press-fit because the chisel would need to move a bit to actuate a force transducer. The solution was some rubber O-rings, and the final result looked like this:

The penetrator holder.

The advantage to this arrangement is that it would also allow me to easily swap out penetrators, assuming their body was 3/8" in diameter, or I could find suitable spacers for the tube.

The whole shebang

After attaching the top and the bottom, I found a suitable place in my garage and mounted the tower to my wall. I decided I'd locate it near a power outlet (currently not functional), since eventually I'd want to have electronics hooked up to it.

The tower, mounted to my wall. Again note the manly tool bag.

To the rafters above the tower, I attached a 2x4 with an eye bolt mounted to it. The eye bolt holds the pulley that guides the steel cable used to raise and lower the sample holder. To the outward facing surface of the tube, I attached a 60" tape measure that I could use to determine the height of the sample relative to the top of the penetrator. I decided to use a u-bolt to hold the door closed during testing.

Then I started dropping stuff to see if everything worked. Because that's what you do. We engineers have a fancy word for it: verification. And if you are planning on changing stuff until it works the way you want: calibration. OK, not quite. But I did need to verify that things worked the way I intended them to work. I headed over to Joann Fabrics and bought the cheapest canvas remnant I could find. It ended up being about 10 oz cotton canvas. I used 2 layers for each sample.

Some of the first tests.

Aside from the lesson mentioned earlier about the lock washers ripping the fabric (you can see that on some of the samples above), it quickly became apparent how difficult it would be to get information out of this setup without instrumentation. I dropped a sample with a 3.65 kg-f weight from about 3" and the penetrator flew right through it.

So I put on my thinking cap, and yanked out the penetrator. For those who don't know, cold chisels are meant for cutting and breaking things. Like metal and concrete. The smaller ones come fairly sharp. My first attempt was to just round the edge off a bit with a file. It came out OK and seemed to help things a bit.

Rounded cold chisel penetrator.

 As I did more tests, I began noticing that the failures at the penetrator seemed to be occuring at one corner, then ripping the fibers open. The penetrator came out, and the thinking cap went back on.

I decided I'd try to knock the included angle down to the 120 degrees that the EN13567 penetrator uses by filing it by hand and ensuring that no edge of the chisel was higher than the center. Using a machinists protractor, a vice and some patience I was able to come pretty close. But I haven't had a chance to test out the new edge yet.