Man, my head really isn't in the game today.

Working on this long rectangular part with the pocket at one end, and I made a very dumb and obvious mistake.

Since working on the previous sets of plate parts, I found that measuring the actual tool diameter gave me much tighter tolerances and reduced that variable at play so instead of getting an average of +-0.002," I was getting +-0.0007," which was significantly better!

So, thinking that measuring the tools for each operation was the way to go, I decided to do it for this larger part on the 3-axis mill. I was using a 3/4" for the majority of the roughing and finishing of the side walls. Not thinking about it, I measured the diameter and found it was 0.721" and input that into the program. If I took a second to think about it, I would have realized that that's a far cry from the 0.75" the tool was advertised as and would have double-checked my measurement.

Unfortunately, I didn't, and the machine ran with an incorrect tool diameter set, and I got the part undersized by 0.012" when it should have been oversized by 0.02". This meant I had to scrap the part and start over. However, when going to cut the stock, once again, I made another blunder of a mistake and cut the stock too short. Normally this isn't much of a problem because there is plenty of leftover stock from previous projects; however, due to the length of the part... I was fortunate to find a single piece long enough to fit the size. 

The good news is we have the stock to remake the part; the bad news is, I wasted several hours of work and material cost over small, stupid mistakes. :/

Cutting the large block of aluminum today to get ready for the large bracket pieces!

The idea behind cutting this large cube of aluminum in half is to have two pieces in which I can get two parts out of the chunk of material instead of just one. It seems like a massive waste, even with cutting the material down, and it is with about 90% of the material being cut away and recycled. 

However, due to the specs of the parts, it's the only way I can make this part. Ideally, I would use a rolled corner bracket piece and just mill down the outside to get the right shape. However because this part will be used in nuclear testing, there have to be very specific requirements that each piece has to meet to reduce the number of variables that are encountered. I'm not sure about the exact use of the part itself, only a few ideas, but from what I understand it's meant for testing different materials under pressure. 

One of the parts that came in the other day is a large rectangular block of aluminum with an engraving on one side and an interesting pocket feature on the other.

This single pocket provides a fascinating challenge that you wouldn't think of by looking at it.

The problem with machining this pocket is milling out the radius corners within tolerance while still getting a decent surface finish. Because of the feature's depth and the tight fillets, you can't get too far down into the corners to get a decent finish. 

Initially, I programmed this part on the 5-axis with simultaneous movements to clean up those corners and side walls with a ball endmill all in one shot. However, after testing out a few of the 5-axis toolpaths, I couldn't come up with anything that would work very well, and most had multiple weird angles that the machine would have a hard time reaching. 

I got on with one of Fusion360's help team to help me figure out the solution to this problem and open my eyes to areas I may have missed. Unfortunately, the solution they gave was very similar to what I had come up with and wasn't ideal but would technically work. The problem is that it focuses more on getting the material removed than simplified axis movements, so it chooses the path of least resistance. Which, in this case, means that it will flip the part upsidedown, attempt to the machine through holes, and other sketchy movements that would crash the machine. 

I'm going to have to go back to the drawing board with this one and probably end up doing it all on the 3-axis and taking a ball endmill with small stepdowns along the sidewalls and bottom surface of the pocket. 

All of this to say, Monday.com is an amazing tool and very intuitive! It hooks up seamlessly with your current systems allowing you and your team to stay in your zone of genius without having to deal with the headache of chasing people down!

Working on a new set of parts today that pose a unique machining technique!

One of the parts is bracket shaped piece with two thin plates meeting at one point at the ends. The tricky part comes into play because of their size! With each side being approximately 8 inches long, the only way to make these bracket parts is out of a large block of aluminum and cutting away all the excess material. 

My boss bought a large square block of aluminum 10"x10"x6", and today I cut dovetails on both the top and bottom of it, then I will cut it diagonally from one corner to the other to get two triangular pieces that I can use to make the brackets. So instead of milling out a whole block of aluminum and getting one piece, I can get two by cutting the block in half! 

I'll be doing all the work on the 5-axis due to this piece's features and some of the tricky surface finish requirements along the side walls. To be totally honest, I am really not excited to work on this, a super big and sketchy part to make that won't be easy. But that provides the challenge, which will be very educational and exciting.

Running through the final parts of this bundle just in time for the weekend!

Everything went suspiciously smoothly today, and I figured out how I messed up yesterday with the miss-aligned chamfer. In my process for milling these parts, I hold the stock with hard jaws and cut away the top portion and do the detail work until I have a hat-shaped piece, at which point I would flip over and set my G54 work zero off of one of the hole features. 

The mistake occurred when I was flipping the part over, and I accidentally turned it the wrong way around, so instead of the front-facing toward me, it faced the back of the vice. The hole looked to be centered in the part, so I didn't think to double-check it was in the right orientation. 

Trying something a bit different, sort of by accident, I set up what I thought was a 3/4" square endmill to finish the side of the larger parts, but when it was contouring the part, I noticed it wasn't cutting anything. Come to find out, instead of it being a 0.75" it was actually 0.02" undersized for some reason. I measured it and added the correct value into Fusion360 and got spot-on results, and the part came out to be within a few tenths. I'm tempted to measure each of my tool's actual diameters going forward to see if I can get tighter tolerances on my parts!

Making all these parts match in their appearance, I used red Scotchbrite to give it a streaked grain look, removing all the machine marks. This wasn't entirely necessary, but a nice touch to have them all match together!

Well, I spoke too soon in my last post, saying I was getting into a nice flow of mistake-free parts making, and I scrapped the first piece of the bundle today, which is too bad.

These parts are super simple 2-op jobs, and I'm doing them all using hard jaws, so flipping it over and finding my g54 work zero is the most time-consuming process. Unfortunately, I somehow forgot to reset the x-axis location point, and when the chamfer endmill was coming around to contour, the top of the part is cut in about a 1/6th into the top of the part, which puts it out of print. 

It's not too big of a deal and doesn't cost much to get the next piece of stock up and going. Going forward, I need to double-check my work zero settings and move the head back and forth to confirm it's in the right location.

Running through the rest of these plate parts fairly quickly since getting the machine set up yesterday! 

I'm starting to get into a new rhythm of programming and running operations that feels very different from my old processes! It may be that I am building more confidence that my CAM programs will run correctly without errors, but I've gone through several jobs now without any but cosmetic errors! 

I'm really enjoying playing around with Fusion360's setup settings and using that to create the stock around the part vs. my old process of manually drawing out the stock and placing the part within it! 

One of these sets of parts I got to work on today reminds me a lot of the Starwars currency 'credits'; they've got a nice weighty feel and similar shape to the ones in the movies! 

I am super happy today because I think I have a new record for turning out new parts from nothing to set the machine up, programming, and running the operation to inspection! 

The one tricky part with this that wasn't that much of a problem was the engraving. Each job that comes in from this client has part numbers on them that we need to engrave, and they have it cut into the 3D CAD itself to reference off of. I originally was using an engraving endmill to do these part numbers, but I had some issues where if I were slightly off in my z height, the engraving wouldn't be consistent with the rest of the parts. 

To fix that, my boss had me switch over to a 1/16" ball endmill and I used a pocket clearing toolpath to cut out each character. Fusion360 normally doesn't allow you to use ball endmills for clearing pockets, but I found if you lie to it and tell it you're using a 0.01" ball and leaving stock on the walls, you can get it to produce the CAM! 

To be fair, the parts themselves weren't complicated and had very basic features, but with a two operation run and setting all the tools up, my rough calculation comes to about an hour and a half! This is a single part that's bundled with other parts with similar features, so running through the remaining sets shouldn't take long with everything already setup!

Today I was able to finish up these plate parts by haphazardly drilling the difficult-to-reach holes!

Going into this project, making these threaded holes on the plates' side was going to be a bear to produce. Thankfully setting each of these pieces and getting them running wasn't as difficult as it was time-consuming, and it didn't take me long before I had them all finished!

Visuals are necessary to understand how I was able to drill holes on a 17" tall part with a machine that could only travel 14" in its Z height. I set the vice up against the X-Y axis table's back end so the plate would hang down below the bottom of the top surface. I then put a square in the vice and clamped each plate part onto that square using an indicator and light taps with a hammer to get it centered in place. 

From there, it was smooth sailing and only took a couple of minutes to program each part and get them running! I did have to run the speeds and feeds pretty slow for the drilling process to ensure the plate wouldn't get pushed down in the process!