Sometimes the progress is slow

Today has been incredibly busy, but incredibly frustrating. So far next to nothing has been accomplished towards the work I have to do, but I did help a few people out and we discovered some interesting issues with some software the lab uses. Okay not issues exactly, it just doesn’t work the way we thought it did. In any case, there has been a lot of running around and doing things, just nothing productive for myself. Some days are like that though, hopefully I will have better luck tomorrow or maybe even later tonight.

Well I’m literally trying not to count down the days until my projects are due, but it feels like, for the day anyway, that the world is against me. Today was our normal lab meeting, but that got cancelled, so I had a chance to help a few of my lab members figure out some other projects I help with. Mostly mechanical engineering stuff since that’s what I did for my BS and MS. Thankfully it was nothing super difficult, but it was time consuming.

We are ordering a few parts that need to be machined, but the person who created the parts didn’t account for the tolerances of the machine shop. For example, we needed a 16mm hole cut into a part and we said cut this hole to 16mm, so that our 16mm part fits into that hole. Easy mistake to make because there is a tolerance to machined parts, you’ll never get a part that is exactly 16mm and the tighter you make the tolerances the more money you need to spend to get it. We asked for a 16mm hole and they cut us a 16mm plus or minus 0.1mm (as an example), we got a hole that was 15.9mm, which was in line with what we ordered, but the part wouldn’t fit into the hole that was cut because it’s slightly smaller. A frustratingly amount smaller because you feel like it should go on and if you press hard enough it may actually go on, but we needed a slip fit not a pressed fit.

For those who aren’t engineering or machinists, a press fit is exactly what it sounds like two parts are pressed together and they don’t move when they are together, one part fits firmly into another. Normally you’ll freeze the part and it will pop into whatever you’re press fitting and when it warms up, it expands and suddenly you can’t take it apart without tools and possibly damage. A slip fit is also what it sounds like, the parts slip freely between one another and come apart just as easy. We could probably press fit these parts, but we needed a slip fit so I had to go into several CAD models to adjust for the tolerances.

This is also why I advocate for time in a machine shop for anyone in the engineering fields that may use machining services. This type of problem is so common I knew exactly what went wrong without seeing the files. Thankfully we can always remove material, but we can’t add, so the parts can be machined in house to fix the problem and I (hopefully) adjusted everything so it won’t happen on the next order. Expect an update in 2-4 weeks when we get the next batch of parts in to know for sure.

That would’ve been the extent of my work on other projects today if it weren’t for an email I got last night from another lab member who wanted my help with an analysis he was working on that I’ve done. He got some results, but the results where the exact same no matter which model he used, the standard model he was loading into the software we use (a MATLAB package) or the custom model we create from MRI scans. I didn’t have any scans (COVID) so I was using the defaults, but it was something important to figure out so I dug into the documentation. From what it said, it looked as if the MRI scans were only used for the plots that were made and not for the actual analysis. Basically we thought the software did more with the scans than that, but he realized this wasn’t the case.

I also took a dive into the code to make sure this was what was happening and sure enough it appears to be the case. We’ve found a solution using another set of functions in the software, but it also means that some of the other lab members may or may not have been doing an analysis wrong. We’re not actually sure about that last part, it doesn’t sound like a whole lot of people in the lab use the analysis that we were trying, so we may have found the issue because no one had ever actually used it. That was the good news, no one wants to have to submit a correction to something that was published. Now that we know it is an issue we can let others know before they try to do the analysis we were using. Neither of us have used the “correct” way yet, so I don’t know how hard it will be to do, but it will be nice to see how it’s done before I ever have to do it.

In short, a lot got done today, but nothing that specifically pushed the work I’m doing forward. I’m hoping tonight I can sit and make some forward progress, if only just a little. Even if I don’t tomorrow should be a better day for work, but who knows.

Enough procrastinating though, it’s time to get some work done! …maybe.