Today I got to finish up the rest of the parts on the first operation. There are two sizes of these parts and did all the first operation in one go before cutting the soft jaws since the location was a known one and easy to set the stock in place. 

My boss also solved the issue I had yesterday, where there would be a small cap on top of the part that should have been cut off when facing the material but wasn't. Wayne thinks it might be due to the released tension in the stainless stock, and rerunning the facing operation, that seemed to solve it. Each part I loaded ended up with the same small cap on top and had to rerun the first toolpath on each one. 

After making enough of the first operation parts, we moved over to the soft jaws. For that, Wayne had me put a 1/32" parallel to separate the two jaws, to give it some clamping force when the parts are in. I then faced the aluminum and located the center for the soft jaw milling operation. It came out a bit rough, and after consulting my boss, he reran the program after deburring the edges. From there, the parts fit in pretty nicely, though there was a slight radius in the crevises of the pockets which I didn't see on the used endmill I loaded. Thankfully it doesn't look like it will be an issue as the fillet on the part matches the radius on the inside pocket. From there, Wayne put a 1/16" pin in the locating hole, and I used a dial indicator to find the center to set the G54. I ran out of time for the day and will finish the final operation tomorrow!

In between runs, I started programming the other parts, which will be made in copper. I unfortunately, I won't be able to share any images of the elements themselves as they are proprietary, but I'll do what I can to share pieces of it. It's got a lot of nifty corners and crevices, which were a bit of a pain to figure out how to do because of interior radiuses done in metric.

Today I got to run the milling operation on the Haas minimill that I programmed last week!

I loaded the new soft jaws into the vice and faced it. I then used a couple of parallels to space the two apart so I could cut a groove down the center to fit the raw 1/4" stock in the middle. From there, I loaded all the tools and found the center to set as my G54 (work origin). While loading the endmills (each bit has to be manually measured in using a height gauge block), I accidentally sent the z-axis crashing down on the material, breaking the 1/16" ball endmill and leaving a small fragment in the stainless stock. Thankfully there was no lasting damage, but it made me jump when it happened and left me a bit on edge for a while after. 

After loading and setting the tool heights, I started the operation. Everything went according to plan and got through all the toolpaths without a hitch. I did set the peck-drilling process to take 0.0004" by accident, and so that took quite a bit longer than intended. Once the first operation was complete, I loosened the vice only to see the part practically spring out at me due to the released tension in the material. Right away you could see it didn't look the best, it has significant facets which the contouring toolpath should have cleared up, and you can see the top face somehow has a little cap on top of it from what I am guessing is a tool height issue. 

I edited the toolpaths to correct these issues by changing the tolerance and added smoothing (allows the toolpath to roll around instead of point to point), which I posted the code again and set the machine running. I also re-measured the tool height offsets on each of the tools to ensure everything was correctly aligned; this time, it ran much better and had an excellent surface finish (I forgot to take a photo). However, there was still a small cap on top, not as tall as it was before, but even noticeable, and I'll have to do some digging to figure that out. 

My boss gave me the day off for independence day so I thought I'd write about my new Vice Fixture from @pocket_nc!

After talking a bit with a few different people about my old clamp system, I designed a couple of months ago; they pointed out a few flaws. The main flaw with it is the rigidity of the design, the only point of contact with the stock is the lower half maybe 1/2" in total, and the rest is free-floating; this causes vibrations to go through the stock and leaves imperfect surface finish quality. 

I talked with my boss about it and mentioned @pocket_nc had a circular stock fixture to hold round rods of material with an ER-40 collet. It's got sufficient clamping force to hold it securely in the vice and provides much more rigidity. I was convinced that it was worth it very quickly, especially since I didn't have to do as much work calculating where the stock was in the machine. My boss also mentioned that you could take 3" round stock and put it on a lathe to cut one end down to 1" (the largest diameter the clamp can hold). With a piece of circular stock cut down to 1" on one end means you can pretty much mill out whatever size part you like on the machine with relatively stable work-holding.

It came in on Tuesday, and I was dismayed to find the tool to tighten the fixture, not in the package. I contacted Pocket NC, and they responded within an hour, apologizing for the mistake, and shipped a new one right away. The next day I received it, not even 24 hours after I got the package. I love the people at @pocket_nc; they are always super friendly and have fantastic customer support!

Nothing too special today, much the same as yesterday programming another similar part and double-checking my speeds and tool orders. 

I also figured out the issue I was having with post-processing to Haas machines. The error I was getting was related to the program name; it requires that you only have numbers, and a minimum amount (in this case 4) and no letters. However, you can change the name after generating the G-code when you save the file. 

I also started programming one of the parts that were to be made in Copper but didn't make too much progress due to some complicated features in the part I wasn't sure how to mill out. It had an interior pocket with the edges filleted to a metric value. I attempted just to use a metric ball endmill to get that corner, but I kept getting weird finishes in the simulation. 

Today I was focused on finishing the CAM for the part I worked on yesterday. 

After talking with my boss, I decided to scrap what I did and start from scratch. I did it pretty messy and wasn't very easy to make quick changes. I also had oriented the part the wrong way and had the first operation on the underside of the piece instead of the top. The reason for doing the top of the piece first was purely for aesthetics, and if things didn't look quite right, it's better to have that on the underside where it wouldn't be as important. I figured that since this is such a small part and for the client, they would want accuracy over aesthetics he said it should be dead on regardless, and so work the side that would have the best finish. It would also make clamping these in soft jaws much easier as I wouldn't have to mill out a groove for the pin on the bottom.

Since this is a two operation part, I decided to make a solid model for the stock rather than try and create a box around it in the CAM setup section. Having a pre-designed object would help with locating the part after flipping it over by drilling a hole through the center, then probing it from the other side. 

I'm pretty pleased with what I came up with and need to double-check some of the toolpaths, but I think it's just about ready! I also did have some issues with the post-processor, giving me an error when I posted it to Haas. It wasn't very clear as to what this issue was, and so I'll have to play around with it a bit more to find out why. 

The new work laptop arrived yesterday, so I spent this morning setting it all up! The computer is a Dell laptop, not sure what make or model, but it runs well with Fusion360 thus far and is about twice as fast generating g-code as my Lenovo ThinkPad.

I also started programming one of the parts I mentioned yesterday. It proved to be a bit more complicated than I initially thought, as it requires two operations and flipping the piece over. I've never made or used soft jaws before, so I'm not entirely sure if I programmed it correctly (will be going over the CAM tomorrow with my boss) and had to do some research on how to make them. Huge thank you to @saundersmachineworks (NYC CNC) for the many videos on this subject!

The material is Stainless Steel, and after doing a bit of looking around online, it's relatively difficult to cut. Thankfully Fusion360 has some suggested speeds and feeds for them already, which are very handy. I'm not sure if its a recent update or not, but I don't recall seeing this library of preset S&F in the tool library before. I was able to get this whole part down to only three tools, a 1/2" 5fl sq endmill for facing, then a 1/8" 5fl sq endmill for the adaptive clearing, and finally, a 1/16" ball endmill for the fillets on the top face. 

The programing itself took much longer than I would have liked, but there was a slight learning curve to figure out how to locate the part, and what tools to use. 

Today I was mainly focused on cleaning the Haas VF1, inside and out to be ready for contract work. Pretty much the same process as the previous machines, spraying the inside down with the coolant hose, then removing the chips and wiping everything down. 

Wayne also got a couple of contract parts for me to work on using Fusion360, which I am incredibly excited for! The pieces are somewhat simple in design, so they shouldn't be too tricky to mill out. One will be in 301 stainless steel, small oval-shaped parts with very little geometry; need to make 16 of these in 1/4" square stock with each one side by side. I can't go into too many of the details as I am not sure how propriety the parts are and will leave it at that for the time being. The other job is a two-piece part, one lower and upper half to with somewhat complex shapes a box apes and requiring a two side operation using soft jaws. 

The computer arrived shortly after I left work today, so I'll be setting it up and programming the parts tomorrow, which I am very stoked for!

I'm honestly so blessed with this fantastic job and wonderful boss. Wayne bought me a machinist toolbox, edge finder, and my pair of Mitutoyo digital calipers! He's been a remarkable mentor thus far and wants to see me succeed all the most as much as I do, it seems! He's shown me a lot of useful tips around the shop, how to machine delicate parts with ease, tool identification, machine operation, and has given me a lot of freedom to figure things out on my own as well with setting up the Haas edge and tool probes. I don't think I could have asked for a better boss than I have, and I'm looking forward to the months and years ahead!

Today (Friday) was a lot more focused on getting the endmills sorted and so there isn't much to write on.

Wayne ran his test operation with the finishing passes (see the last post), and they came out nicely, which means the issues he's been having with the VF 2ss is something related to the 5axis table and not the spindle. He had some surface finish issues with one of his 5axis parts and was trying to figure out where the problem originated.

Today I got to finish up the calibration system and learn how to square the vice and setup the workpiece.

To locate your workpiece to match the program, you use what's called G54. For instance, let's say you have a 4" square block of aluminum, and you want to find it in the jaws; you would use the probe to touch off each of the four faces to divide then and find the exact center (work origin). It sounds more complicated than it is, and there are preset programs to do it all for you after setting in some necessary information. 

I spent a frustrating 20 minutes trying to figure out why the probing would keep failing on the Y-axis direction until I found out that you had to be pretty close to the approximate center for it to work correctly (approx 0.25"). 

After I got the stock probed, Wayne set up a quick program to test how I did for the whole calibration system. The tool touch off probe wasn't working like it should and wouldn't bring the tools down low enough for the endmill to touch off. It turns out I just forgot to add the height of the ring gauge when setting it up yesterday. 

Once everything I got everything dialed in correctly, Wayne ran the operation he set up, and off it went! He did the blanket like form to test if there was an issue with finishing toolpaths in the Z direction (unrelated to what I did) which is why he sent off the 5-axis table.