I'm a 17-year-old machinist working for a small prototyping shop in northern California. I document my journey through my learning experiences and everyday shop life!
Taking things a bit slower today and spending a bit more time checking and reviewing each step before proceeding.
Thankfully, after yesterday's crash, it looks like there isn't any major lasting damage, and after dialing everything back in, there appears to be repeatability, though only time will tell.
I got the wonderful opportunity to hone my deburring skills on a very complex part that had many small "fun-to-remove" burrs. I say this sarcastically because anyone in the machining industry knows how time-consuming and painful it can sometimes be with more complicated pieces.
I've got a possible deburring tool idea in my mind that may speed the process up, though I'm not sure how viable it is to make. The basic idea is a small steel tube, cut at an angle, so you've got a pointed tip at one end you would then pinch to get a corner shape at the end. My idea is that you could use this to scrape the tight corners where the edges meet up to another face and clear both faces it comes in contact with, rather than having to lightly go back and forth on either side with an Exacto knife.
Well, little did I think this day would come so sudden, but it did, and I'm honestly not sure what to think about it.
I had my biggest, possibly major crash on the VF2ss trunnion today, which did not sit well with my nerves.
A little backstory;
After getting the jobs finished up for new clients, my boss put me back on the truck marketing pieces to get those finished up at long last. I quickly ran through the finishing toolpaths (cutting the windows, engravings, and highlights) pretty quickly without incident. However, when I was happy with all the changes and was about to run the final cycle and walk away, there came the machine crash.
What happened was the tool somehow thought it would be a good idea to mill out the base of the trunnion instead of the part and subsequently had a nasty crash. The strange thing is, I didn't make a single change to the toolpath that caused that crash; I only made changes to the following ones that were not linked in any way.
My boss had me walk away from it and let the machine rest overnight before trying to fix or run anything, just to let it settle, and probably, let me settle as well before any more catastrophic events.
Thankfully (it could have been worse), it appears that the crash wasn't quite as bad as I was initially thinking and maybe a flesh wound. The endmill broke, and the tool holder is shot, but other than that, it seems that the 5-axis trunnion doesn't have too much lasting damage. I will find out for sure tomorrow, though.
In his sermon today, something the pastor said gave me an idea for an interesting analogy related to benevolent works.
There seems to be a large divide or sometimes imagined divide between churches that profess salvation through works vs. salvation through grace. The thing is, I'm not sure that there really are many churches that profess salvation through works as there used to be. So, setting aside for a moment that your works cannot save you, the question then comes up, what about the good works we are to perform? Where does that come into play? And why are so many people uncomfortable talking about it?
You were one of the few who went to the good front when called. Warfare is not easy and not always pleasant, but it is for a good cause. Your commanding officer pulls you all together and is in tune with your needs, and fulfills them.
He does not push you beyond what you can withstand. He tests you with trials to strengthen you against the enemy. All who are under him do well and serve well, but not all seek special dangerous missions. Most are content to fight only when commanded. Those that choose to take risk and go after those difficult operations win internal battles and honor after the war is over for their actions.
Though times are hard, their commanding officer looks after their needs and rewards those who take courage and look for all those opportunities to serve their good side. The actions of those valiant men benefit not only their fellow officers but the entire battle and war.
The actions of these mighty men did not get them to the battle, but rather, after they got to the front they performed these good works and were given strength and opportunities to do so. They were drafted and went like all the rest, but their actions were benevolent ones and not required.
They were performed because they were so blessed as to be able to fight for the right side.
I finished up these parts for the weekend, and boy, am I glad to be done!
This one feature that I'd been dreading to do ever since I saw it I was able to make with a 0.0002" (two tenths) tolerance!
I briefly mentioned it in my previous post, when the L-shaped part is put with the elbow stick pointing up, there's a small pocket relief at the very bottom. Normally this wouldn't be a problem, and you could take several stepdowns while finishing that side vertical face. However, the problem is that the vertical wall isn't actually vertical and comes in like a dovetail. The angle is very slight, only 0.88deg, so it isn't easy to see by the naked eye, but still there.
I was able to mill this feature with a custom ground back 3/8" endmill turned into a slot mill. Using this tool, I took several light passes to clean up the bottom of the pocket; then, with the tool's reduced neck, I was able to get under the angled piece and clear out the pocket.
It wasn't as scary to run as I initially thought. My expectations were that the undercut would have a very rough finish, and the corner wouldn't blend very well with the angled wall. However, I was pleasantly surprised by the outcome. I took several tenth staircase steps along that wall to blend the tool's square shape in better with the remaining surface. The tool didn't squeal, and everything ran very smoothly!
I am thrilled to see these parts finished, a lot of headaches and mistakes, but all learning, that's what I can console myself with. The overall tolerance on these pieces is within 0.0008," which I'm extremely pleased with!
What is an Effective Checklist, and why should I make one?
When looking to improve our workflow, we often start with a checklist going through each detail. The problem is that how many of those fancy checklists ever stick around?
Through numerous tests, I've put together a very simple checklist that condenses the information down to what actually matters and helps us stay focused without getting lost in the weeds.
The basic idea behind this process is finding what information you actually need to follow through with the task without error. In a perfect world, we would put all the information in this checklist, but too many details will quickly overwhelm the user.
What software do you use?
I use Google Calendar for my checklist creation as it's got a clean and visual interface that many people are familiar with.
Using Google Calendars also has the added benefit of being sharable, so anyone within the same custom calendar can make changes that update the rest of the team automatically.
Additionally, you can hide/show the tasks separate from your normal calendar information.
How do I know what to put in my checklist?
Finding what information to input into the simplified checklist is easier than you might think and doesn't take long. Perform your task as you would normally, and as you're going through the steps, take notes of the most important things to be aware of. Keep in mind you don't need to know the reasons behind certain actions, only that they are made.
For example, in a blog uploading workflow, you may need to change the author, it's easy, but very important. Condense that information down into as few words as possible while still making sense. No need to be grammatically correct. A condensed version would look like "Change author"
You can also use your new simple checklist as a hub for finding links to information relating to the task.
As an example, I can find the new blog content and website admin embedded into the words of the checklist, making it easy to follow through.
What if someone makes a change?
If you're working with a team and some new important change needs to be communicated to everyone, you can use bold, italics, and underline text to highlight specific details. This clarifies that something new about the process needs special attention, and the master list would then be referenced.
I also highlight details in the simple checklist to make sure I don't make the same mistake twice. For example, I uploaded the first image to a blog as an imbeded image when it should have been set as a featured image. In my workflow checklist I highlighted the area I made a mistake on.
What's the most important thing I should be aware of?
K.I.S.S. Keep it simple stupid.
We so easily get caught up in making beautiful systems that only work in theory, but theory doesn't mean it's effective. We think and see things visually, so keep words to a minimum and just go with the flow of how you do the task right now. It's about reminding yourself of the important details, not every detail.
Nearing completion of these L-shaped blocks with only a couple of features to go!
Neither my boss nor I thought about a specific indented feature on this part that could nearly have been impossible to create.
One wall of the L-shape is at an angle inward; you can't access a small pocket feature down at the bottom that needs to be square with the bottom face. The L shape's long segment is similar to a dovetail where you have the two angled walls coming in toward the middle.
The only way to get an undercut pocket would be to use a long-reach dovetail, or in this case, a slot mill with a reduced neck. Unfortunately, because of this, the surface finish will be difficult to make look pretty, so we're not going to get points there. It's one of those really unique features that you seldom see but can really bite you in the bud if you're not careful when quoting parts like this.
Really liking the new cleaning and deburring station in the shop with organized spots for each tool! It makes things much easier when you're working with another person; there's a spot for each tool, so you always know where to find it without wasting time looking.
Finished up the first operation of the first set of parts.
Unfortunately, I made a stupid mistake yesterday when trying to use the tool wear compensation in Fusion360. I accidentally had the toolpath generate on the centerline of the contour rather than having the cutter's edge follow the wall.
I got to the shop this morning to check on an overnight run, only to find the Haas minimill paused at a single line of code with the spindle on and coolant running. I have this happen several times in the past, and I keep forgetting to save the code that caused it.
I'm not sure what I'm missing, but I hypothesize that the machine is getting overwhelmed by the information and pauses. When I came in this morning, it looked like someone paused the machine rather with a feed hold, but nothing happens when trying to hit cycle start.
It ran through the first 3/4" of the program just fine, and only when the machine raised the tool in the z height did this happen. It seems pretty consistent with the previous times it's happened.
I asked this question on a public forum for Fusion360 and got one person responding with a very probable solution. He said that he had similar past issues where the power would fluctuate and cause the machine to stall for a second, just enough to pause the machine where it is without moving forward or corrupting the file. My thought is that since it's such a large file, the machine has to read through it all at such a pace that it would require less of a power surge the larger the code.
As it happens, this file had several thousand lines with about a 3 hour run time.
I get to work an hour before my boss each morning to warm up the machines and clean the shop.
Today I had a little extra time on my hands this morning before my boss arrived, so I decided to re-organize the cleaning & deburring station in the shop using some leftover pick and pull foam for shipping the rocket kits. I'm really pleased with the outcome, and it makes it much easier to keep the worktop clean!
I found that some of the smaller tools that aren't as thick as the foam are more difficult to grab in a pinch. So I only removed half the foam so that the tool would be propped up toward you, presenting itself to be readily available.
I believe that if your workspace or tools are clean, you automatically take better care of it. This is why some automotive companies require you to sign a document saying you will keep the vehicle clean while it's in your ownership. If you drive a beat-up old car, most probably won't maintain it as much as they would a nicer-looking vehicle.
I absolutely love cleaning and organizing, and I'm in the process of slowly organizing the shop in between contracts and janitorial duties :)
Working on these somewhat complicated parts, I got to program yesterday.
There are three L-shaped pieces, and I'm doing them all together in a large piece of bar-stock. Unfortunately, I had to run through a somewhat tedious process of removing the material around it before I could finish the sidewalls and get the small details because of their shape.
Because these parts have such high walls and tight corners (see my previous post), I had difficulty getting good surface finishes. I'm actually a little surprised how well the 1/4" long flute endmill did cleaning the sides at multiple steps. We don't have a tool long enough to get the entire depth of the wall in one shot, so my boss reduced the tool's neck so I could take it at several stepdowns. I would previously have issues where there would be a witness mark showing where the end of the flutes ended and the endmill's neck started; this time, there wasn't!
The difficult feature of this part is the corner where the two inner lines meet up. They have a hole in that corner to ensure the block that gets set into it is perfectly square to the two side walls. Unfortunately, this means that cleaning the inner walls has to be done with a small endmill which chatters more the longer the tool sticks out. I was able to clean up the majority of the wall surface with a 1/4" SQ, but as it got close to the corners, there was a very slight hook where the tool couldn't quite make it. Because of this, I had to take a very long reach 1/8" sq and take very slow cuts (low rpm and feedrate) to make sure that back corner was square.
The problem with this is that because you are using two different endmills to clean the same surface, it's nearly impossible to get a seamless transition due to the actual diameter of the tools you're using. Because of this, I had to guestimate the actual diameter of the 1/8" and compensate for the offset by leaving negative stock to leave on the side walls. It's not perfect, but I was able to slightly undersize the sidewall by 0.0003" (3 tenths) which is acceptable for this part so long as it's not positive). You can see in the image above the faint difference between 1/4" and 1/8" sq endmills.
Working on a new set of parts that pose a very subtle but interesting challenge.
One of these pieces' requirements is that you can't use any sandpaper or Scotch Brite to finish up the parts' surface due to their final use. This wouldn't normally be an issue; however, many steep walls lead into tight corners on this job. They are all L-shaped, and the parts' outer contour is no big deal, and I can use large endmills for it. However, since they have an inside corner that needs to be pretty sharp, I have to use a smaller endmill.
In one section, the wall spans nearly 1.5," and I have to somehow get an 1/8" endmill to finish the walls without any burnish marks from the shank. The solution to this problem is using a combination of a 1/4" endmill to finish the majority of the surface, then come in with a reduced shank cutter and get the remaining spots the 1/4" couldn't reach.
One problem may arise, which is the diameter of the tools themselves; when endmills are ground down, they are typically undersized vary slightly from their written size. To fix this problem, I will be using tool cut diameter compensation (CDC) inside the machine and program to sneak up on the correct size.
For example, say you cut a square block with a 0.5" endmill, and you want the block to be 1" square. When you cut and measure it, you find it's actually 1.008". You would then input into the diameter offset -0.004" (0.008/2), and the toolpath would adjust to compensate for the offset. Most of the time, this isn't necessary as the difference is so slight it falls into most tolerances. But in this case, where the wall's surface needs to be clean and accurate, I will have to make sure it's as close as possible.
