Making mistakes, again and again

Blog Picture: 

Last Tuesday at the Electrolab I set out to assemble together a large part of our test engine, including a concrete nozzle made out of fireplace sealant that Ryan had shipped to me by post from Finland.

As I have often experienced in the lab, the evening was marred with mistakes. I had already mentioned the 2mm drill I had broken last month when attempting to drill through an aluminium part. On Tuesday, the first mistake I made was during the process of melting paraffin to mold a new grain, after the previous one was partly used up in the first combustion we managed.

To melt the paraffin, rather than using an oven as the previous time, I used a heat gun which I aimed, vertically, in a can that contained the paraffin block to be melted. The method itself worked well enough, and the approximately 100g of paraffin melted to a fully liquid state in less than 10 minutes. However, by holding the heat gun vertically just above the can, it was subject in part to a return of the heat, with the result that a plastic part surrounding the heat gun nose began to melt. This did not harm the gun functionnalities, and its owner (who lends it to the lab users) assured me it was not a problem, I could not help but feel bad about my blunder steming from my inexperience.

The second mistake I made that evening was when drilling the steel tube sides, in order to prepare the holes to screw the nozzle retainers. I dumbly measured the space between the screw holes on the retainers, and reported that space on the outer surface of the tube, then drilled the first 3 holes. Obviously, as anyone who went through a middle school geometry class will know, the holes on the tube turned out not to match the position of the holes on the retainers.

The last mistake I made was when sawing the aluminium tube, to make it just a bit longer than the nozzle+grain couple. The previous time I had sawed it, I used a band saw, which did the job perfectly. So perfectly and quickly, in fact, that I felt almost foolish for using the band saw to cut through "tender" aluminium. So this time I used a manual hacksaw, and the result was that the cut was not vertical at all. Consequence: I will have to cut it with the band saw again, to obtain a clean part. Which is what I was told by a member of the lab after I began my sawing.


Why did I make these basic mistakes? Of course, there is my inexperience in DIY in general, and the fact that I did not go through a technical training (my background is in business). But I suspect it is more due to 2 key factors:

  1. my reluctance to ask others in the lab for advice / comments on what I am about to do, especially when it's the first time I do it. That is very probably because I already sollicit others so much.
  2. my focus on moving on by doing things, to the point that my actions may be too rash at times. If I took a bit more time to think my actions through before actually doing, it may help save time.

The important thing is to learn the right lessons from these mistakes. I see especially two steps, which I will start taking from this week:

  1. When I undertake a new action (using a new machine, testing parts in a new setup, etc.), I will systematically ask for advice / comments from someone knowledgeable about the area touched. Even if it means sometimes bothering someone, better that than being sorry after having broken something.
  2. I will plan the actions I will take during a given session at the lab beforehand, while not in the lab. This will allow me time to think things over while the session approaches, and probably help avoid mistakes and wasting time and resources.

This may seem obvious to anyone working in an engineering environment. I even apply similar rules in my daily work. Yet when out of your expertise area, with limited personal time to dedicate to a project, what was obvious in your comfort zone may become far more elusive. I guess there is no good reason it should be.

As Mickael Jordan said, "I fail over and over and over again, and that is why I succeed." Mistakes and failures are necessary parts of a product development process, and of the learning process I go through. But well, if we could reduce the number of mistakes we do in achieving a certain level toward our goal, that would not be bad ;o)


As a side note, I mentioned that Ryan had shipped through post the concrete nozzle: it should be noted that it was protected within a 3D printed PLA tube, which kept its role perfectly. We were not simply able to remove the nozzle from the tube once the screws were removed, we had to saw it on 2 opposite sides to remove it. The nozzle inside was in a very good shape!

As the picture at the beginning of this blog post shows, it takes some pushing to get the nozzle to enter the aluminium tube, but I believe that when we mount the final system it will go in all the way without large difficulties or risks of breaking it.


ryan.pulkrabek's picture

This may seem obvious to anyone working in an engineering environment.

No, not necessarily. Many engineers I have come to know aren't very familiar around the shop floor. Those that do become familiar have an upper hand, as design for manufacturing (DFM) is needed.

And don't be hard on yourself. We all make these mistakes. I am always breaking something because my curiosity gets the best of me. As one of my engineering professors once said, "Common sense always comes to us after something terrible has happened". Another professor once told me, "You learn the most on the shop floor. Once you've set your hair on fire, you learn not to do that again". Getting a degree in engineering only allowed me to understand the theory and calculations behind designs. Working on hobbies in the garage is how I learned to put the theory to use, which I value just as much as the theory.

As for drilling holes in the aluminium tube, I propose that I create a jig. Based on this jig, just drill where the holes are and all should match up perfectly.

Then, for inserting the nozzle into the tube. I recommend using a piece of wood. Place the wood on the nozzle, and gently hit the wood with a hammer. The wood is soft enough to absorb the impact yet still force the nozzle in.