Working on the first of many parts for this new client. Everything has to be extremely clean, and I spent the previous two days washing down the machine in preparation for these pieces. 

There aren't anything complicated about these parts, just big, fairly straightforward pieces. The tricky thing with these is you can't use any cleaning tools like files, sandpaper, or certain Scotch-Brite pads. Even the cleaning process for these requires no brushes or sponges to be used. 

The part I worked on today was a basic plate with a few countersunk holes. My boss wants to go the extra mile for this company, so I used a ball end mill to get the holes' large fillets. I didn't realize it until working on this project that the boring toolpath in Fusion360 works with tapered faces quite well, and I was able to quickly rough it out for the ball endmill. 

I am also diving in deeper into the speeds and feeds area of machining, reading through a few tooling manuals, and understanding more of the chip-load and how it relates to the SFM & RPMs of the machine. I'm quite surprised how far I can push these tools by using the correct speeds and feeds (who would have known!).

I got the Haas minimill coolant tank all cleaned up today and refilled it with the fresh!

Part of this process involved cleaning out the pump, which consisted of running several gallons of warm-hot water through the entire system to clear out any leftover debris or coolant. I accomplished this by taking an empty plastic bin and filling it with clean water, lowered the pump inside it, and let it run for a few minutes to flush everything out. 

Removing the coolant and chips from the machine's bed was the difficult part and took me a while to figure out how to do it. Since this is an older machine, it doesn't have as many fancy features, like an access panel or easily accessible hatches to perform coolant maintenance. So I was subjected to crawling on my knees with an outstretched am and a wide range of long objects to push as much of the grit toward me. 

I eventually cobbled together a method of using a long reach chip scooper to bring the bulk of the junk toward me, then using an (I am ashamed for this) t-square to get the chips from the very back out. I then took advantage of many old shop towels, which I used to suck up as much of the remaining liquid until it was fairly dry. Finally, I used a mop-like cleaner to pull the last remaining sludge from the back corners that the other tools couldn't reach. 

Once the underside was as clean as it could be without investing a ton more time, I poured the new coolant into the machine. Finding the right ratio of coolant to water was the difficult part, and I'm shocked the company doesn't have the ratio printed on the side of the bucket. But after doing a bit of research, I found the correct coolant-to-water ratio (4-10%) and eyeballed each bucket. When there were only 5 gallons left to go, I used a refractometer (a tool that measures the water content in liquids) and measured my new coolant against what was recommended, and I felt right inside the gate of 6.4-12.0%.

There aren't many great resources out there that explain how the coolant ratio work, so I had to jump around to several sites to get the gist of it. From what I understand, each coolant type has its own ratio, but many are very similar and don't need to be spot on. Plus, over time, the water evaporates, so the coolant becomes more concentrated. When you add more to the machine, you have to keep adding less and less coolant per bucket so you can compensate for the concentrate. 

Today I got to do the thing I've been dying to do since I got in the shop (mild sarcasm), which is cleaning out the coolant and chips from the Haas minimill!

We just got a couple of jobs to come in that require precise machining requirements due to the part's use. I believe it's got something to do with nuclear prototyping, and so many variables have to be removed when working in that environment. Because of this, the coolant being used on these parts cannot have touched any other metals except that which it is milling.

Since we've made copper and steel parts in the machine, it meant that I had to clear everything out and remove the old chips and coolant and replace it with new stuff. It was a little bit of a daunting task at first, just because there are so many nooks and crannies in the machine that would be super difficult to remove with the tools at hand. This machine is the only one that doesn't have an external hose hooked up to it, so I just extended the host on the machine and manually washed it out. 

It was a very disgusting job, and I got closer to coolant than I ever want to be again, but the inside is now looking beautiful and clean. I haven't yet completely emptied the internal tank, which will require a lot more work involving a wet vac, uncomfortable positions, and long arms. 

My boss recently made the wheels to go on the little truck models I've been working on, and they turned out amazing!

Unfortunately, something neither of us picked up on until after the tires were on was that the back fender's angle that sticks out past the wheels is at a slightly different angle than the tire guard on the other side. It's slight but gives it a weird optical illusion that makes the wheel look slightly off-center. Thankfully it's not super noticeable when the windows and details are cut out after getting anodized, but I was allowed to change it before running the final blocks. 

Initially, I was a little afraid to do this, as making modifications to solid models in the past has been anything but a clean process. However, taking advantage of the surface body features in Fusion360, I was able to delete, modify, and remove faces, rather than solid 3D models. This made the change much easier, and there weren't any weird floating bits leftover that I would sometimes get. It also was able to show me very quickly where there was a disconnect on faces/features before taking it into the CAM environment. 

One of the steps of this unique machining method I learned from my boss was milling out the small pockets and features in the piece while still in a block. 

The thing I love most about this very different technique is that it's a one-shot go. There is no first and second operation; there are no soft jaws, there are no extra complicated steps. It's a simple push of the button, watch it run for 20 minutes, stick a block of metal on it, and hit the button again. The outcome is a finished part and requires no cleaning up! It absolutely isn't the most efficient way of making these parts, but it makes the most sense for just a couple of them and takes the least amount of time.

I'm really discovering that, not just in machining, but in life as a whole, you really can't cut corners. There is no fast track. You have to take it slow and steady; that is really the fastest way. It's painful at times, but the more patience shown will help you move faster toward that desired outcome. 

Top clients are the ones you love working with the most. They are most dedicated to your brand and buy from you religiously.

Unique Offering is what you offer that is different from similar businesses. Something that you do better or are the only one offering.

Systems is what you improve on once you identify your top clients and unique offerings. You build your systems to cater specifically to those first two circles, narrowing your area of expertise to do it better.

Hardlotion's Unique Offering is "Fixing dried and cracked skin with, few, all-natural ingredients that are good for you."

Looking through the current top customers and their previous orders, there seems to be about half that fall into the 'ideal customers' category. They buy the unique offering with few or no discounts and rarely buy the sideline products.

Once I've identified several top clients, I will then interview them to find their frustrations in the skincare industry and their 'wishlist' of things they want improved on. Finding out the frustrations they are running into will help me better identify how I can serve my ideal client and attract more like them. 

I would also ask for other vendors that serve them. Such as personal health companies. So that I can collaborate with them to bring better service to them.

Whew, am I frustrated, brace yourselves. 

I once again have run into a toolpath changing problem with Fusion360 and have no clue what is going on. My biggest problem is explaining the issue to other people as I can't replicate the issue, and there is no "issue" per se. 

Talking it out with my father, he helped me see a bit more of the problem and where it's occurring. In a very basic example, say you have a square shape, and you want your endmill to follow half the outline of the square to cut it out, so starting in one corner, you have it trace the outside and stops at the opposite corner. This works fine, and you get the machine to run it. 

However, say you change your mind and want the tool to run past the final corner and keep going half an inch longer. You make that change and nothing else and re-export the code. When you run this new program, it goes the extra half-inch, but now it cuts the corners leaving little chamfers, which you didn't input into the program. Technically, if you go back and look at the settings, you can see nothing wrong, and the program is doing exactly what you told it to. The machine didn't crash, the toolpath ran exactly like it was supposed to. But your final outcome was not, because now you have a square with rounded corners. 

My boss thinks there is some issue with the software where it's making changes to 'set' toolpaths, which technically I suppose it is because I am making a change to that toolpath. But I didn't tell the machine to round those corners off, but it did it anyway, and if I simulate it, I can clearly see that it rounds the corners off. 

My only idea as to the cause of this issue is over constraint.

Fusion360 likes to be "automatic" and will have preset values for you when you select toolpaths, which are great. However, those values read off of each other, which is where it gets complicated. Say you have a parameter in the setting of the above example, saying that it will make square corners as long as it is within one inch of travel distance, but anything above that will shave the corners off. Now your first toolpath works fine because it's within an inch, but if I wanted that extra half-inch, then it would push it outside the limit and turn on the corner-cutting setting.

The problem arises when I set up my initial operation, and I inevitably need to change something, so I go back and change it for the next run. Then I run through the new program, thinking it will only make that one change. However, that change may cause another effect, in essence turning on the 'corner-cutting setting.' It's not the fault of the software, but rather how it was setup. If I were forced to input every single parameter and setting into each toolpath, I would not have a problem, but that would take huge amounts of time when you get into more hairy 5 axis projects when you have 50+ toolpaths. So you have to have some preset information to make the workload possible. 

If there is a way to turn off all presets in Fusion360 and leaving them blank so that I would be forced to fill them in before generating the toolpath, I feel that it would solve this problem. Unfortunately, I'm not sure if this is possible or not and will have to do more research. The most frustrating thing about this is that I can't seem to find anyone who really understands this problem or has faced it. 

Working on these simple 2op parts on the 4th axis today to familiarize myself with the setup when working on the more complicated pieces.

My boss has an exciting way of making small intricate parts that require multiple angles using the 4th axis rotary. On the face of it, it would seem perfectly natural to do this, but it gets a little more complicated when to do it all in a single setup. For example, these two small rectangle pieces that I made today were all in one setup and required no cleanup after pulling them off the machine. 

To make these pieces, I used 1" round stock, and taking a large endmill; I was able to quickly remove the majority of the material, leaving a small shoulder on the remaining leaving a small rectangle box at its final dimensions. Once you complete all your detail work, the problem arises, and are left with the part hanging on the stock. Cutting it off without any skin or material that has to be removed later is difficult, and many people clean the pieces up by hand. However, there is an alternative. 

Cleaning the top surface of the part and the shoulder you recessed, you can then double-stick tape a small metal block on the underside of your piece, attaching it to the stock. Then, taking a small endmill, you can carefully step your way down the material, roughing, then finishing the side face until you part it off. Using this method, I finished the part on the machine and only had to wash it off before bagging it with no cleanup required. 

Everything stayed in its place, and after measuring it, I found I was within 0.0005" on it all. From start to finish, the pair of these little rectangle parts took approximately 3h 32m. Programming them took 37m, and setting up the 4th axis and getting everything dialed in took the majority of the time. One of the best parts of this trick using round stock and tape is that you can keep pulling the material out, and you don't have to reset any work offsets. 

Man, I am pleased with this piece; the complete turnaround time was about 3.5h from start to finish. That includes programming, setting up the machine, running first op, soft jaws, and second op. Definitely, my fastest part yet, and everything is within 0.0005."

I'm taking a slightly different method when programming this and the following pieces where I am not relying on any automatic toolpath generation. Previously I would use the 3D adaptive and pocket clearing toolpaths, which take your initial 3D model setup and automatically put together the program for you, which I would run into issues when needing to make small changes to previous toolpaths.

I am taking advantage of this new method of creating custom sketch geometry and manually drawing out boundaries I can select. The downside is that it's not connected to the 3D model itself, and I can run into issues if I move the model geometry. It does take longer to program the parts, but I'm getting that insurance of doing everything manually, so I know right away when something doesn't add up, rather than finding out later. 

The part is very basic and only required two operations, the second being held in soft jaws. I put these pieces side by side in a piece of bar stock and bored a hole in the center I could use as my work origin when flipping them over. When I cut the soft jaws, I could technically use the same setup origin for the second operation, but when I am using any clamping force, the geometry could move a tad, which would throw things off. Utilizing the hole, I can get an accurate location point for the part once it's in the vice and clamped down. 

I've got 5 more pairs of parts to make, which will be done on the 4th axis and get really hairy when it comes to small detail and hard-to-reach places which I'm really excited for. I love making smaller parts vs. big ones; something about it is so much more satisfying.