An Inventor's Quest for the NHL Pt. 28

Controlling Deflections

This series follows my attempt to develop a product that I dream of getting into the elite levels of hockey. Previously on the Quest: Part 1, Part 2, Part 3, Part 4, Part 5, Part 6, Part 7, Part 8, Part 9, Part 10, Part 11, Part 12, Part 13, Part 14, Part 15, Part 16, Part 17, Part 18, Part 19, Part 20, Part 21, Concept Launch, Part 22, Part 23, Part 24, Part 25, Part 26, Part 27

---

It was strange to see how quickly the progress took on the nature of ordinary existence: the Surprise had not run off a thousand miles before the unvarying routine of the ship’s day, from the piping up of the hammocks to the drumbeat of ‘Heart of Oak’ for the gun-room dunner, thence to quarters and the incessantly-repeated exercising of the guns and so the setting of the watch, obliterated both the beginning of the voyage and its end, it obliterated even time, so that it seemed normal to all hands that they should travel endlessly over this infinite and wholly empty sea, watching the sun diminish and the moon increase. 

Patrick O'Brian, HMS Surprise

In most stories, grand adventures usually have a sense of only moving forward – every day, a new experience; at every turn, a twist. The very best stories, like the one above, hint at a different reality.

In reality, grand adventures (if my adventure can be considered grand) have a sense of moving in circles; repeating patterns that obscure the forward movement.

In my adventure, the cycle revolves around solving problems and it usually goes something like this: I notice a problem > become convinced that this problem will kill the project > realize that I think that every single time > break up the problem into smaller chunks > resolve one small chunk > get annoyed that the problem isn’t magically fixed already > resolve a few more chunks > realize anticlimactically that the problem is gone > eat a celebratory pizza > notice a problem.

In my current cycle, the problem is a familiar one: deflections. Ever since the first prototype, I’ve been trying to control the instability inherent in my design. Obviously, a facemask that deflected back into a goalie’s face wouldn’t be very useful.

But that’s what it looked like I had on my hands. The drop tests which showed the strength of the cage also showed its tendency to deflect a worrying amount.

One of the heavy drop tests. The cage didn’t break, but it did bend way more than it should have.

My writing apparently needs fresh suffering for vibrancy, so I can’t paint a convincing picture of my distress after those tests a few weeks ago. You’ll just have to trust me that my spirits were very low. 

They only revived when I realized that I’d gone through this cycle a few times before and that, at every turn, the project had survived and even progressed. The more trips around unscathed, the more I gain confidence in my abilities, and the shorter the melancholic part of the cycle gets. So, after only a day of moping, it was once more unto the breach.

First, I needed to be absolutely sure I actually had a problem. The drop test is just an approximation for a puck shot at a mask. The test can match the kinetic energy of a real shot or it can match the momentum, but not both at the same time. It’s possible that the deflections I was seeing were only a product of my testing.

Naturally then, I set up a test to test my testing. I took a shot at the cage, filmed it in slo-mo, and compared the deflection to drop tests at two different weights – one which matched the energy and another which matched the momentum.

The original shot

Momentum-matched drop test

Energy-matched drop test

It looked to me that – for deflections – matching momentum was the way to go. That was a relief, because it meant that those insane deflections I saw initially would only happen on a 600+ mph shot. I wasn’t completely out of the woods, though, because there was still too much deflection for my liking even at reasonable speeds.

What I noticed watching video after video of impacts was that a lot of the movement was coming from the crossbar. And sure enough, even though it looked like the crossbar was completely locked down, something in the design was allowing for it to move within its constraints.

I spent some time thinking through and sketching out a few theories but, really, there’s no substitute for trying things.

That kicked off a flood of testing. Decoupling the interlocking loops fixed the rotation of the crossbar. A really good step. Better orienting the rope to oppose the direction of the force didn’t make much of a difference. Crossing lines so that there wasn’t more than an inch and a half of unsupported rope killed local deflections. So on and so forth, but there still something missing.

After going through the videos again, I finally found it.

Look at the “V” formed by the bottom ropes and notice how the angle changes on impact. Clearly, the top two points of the “V” get closer together, making the the “V” taller and skinnier. This drops the crossbar down towards the mask and allows for slack. 

All that happens because the frame, which anchors the ropes, closes in on itself like a clamshell. I didn’t really think that would happen. I mean, I made this latest frame out of some really beefy, stiff steel. When I was bending it into shape, I had to put my whole bodyweight into the bender to get the metal to bend. But I guess it wasn’t beefy enough.

Before I got too involved in finding a perfect solution on how to stiffen the frame, I wanted confirmation that I was solving the right problem. So I jammed a piece of tube into the frame to keep it from closing in, and saw good improvement.

Next step is to find a way to make that adjustment more permanent. Hopefully, that will fully resolve my deflection problem. I don’t want to get ahead of myself, but if it does, I’ll enjoy a celebratory pizza, find a new problem, and start this cycle all over again.

Thanks as always for reading,

Surjan