Tuesday 21 May 2013

Prototyping 2 - Day 10


Day 10 was a mad rush, plain and simple. 3 days left including today on the clock. Considering I had only planned on spending 2 weeks working on the project (and to be fair only 2 weeks in the UK), I was way into over time. There couldn't be any more delays, my flights had been put off but were finally booked so this was it. I needed to work hard and fast to get the lenses completed, otherwise this really would have been all a massive waste of time.

My primary focus became completing the 90.2 lens. With the new tool working a treat, it was time to tackle the central lenses, including a double concave element. The very centre elements are made of polycarbonate to balance chromatic aberrations. The polycarb I had was 70mm in diameter, and I needed it to be 51mm for the machine to process. So into the manual lathe it went and over about half an hour I turned it down layer by layer.




I then parted it off in the centre and turned down steps so that the small chuck of the CNC could hold them. Slow and steady wins the race with parting off like this; the tool was working a longgggg way down the billet and was at risk of choking. Hence the reason I mounded it in the new vari-speed lathe, so I could take it right down to 50rpm and work my way up to the best speed.


From there it was straight into the CNC. I had spent the previous night coding all the remaining faces so that they were all ready to go. The profile cut as expected, although this was the first large polycarb element I had cut and I was a bit frustrated with the cutting lines still being visible in the new surface. Polycarbonate is a nightmare to polish, so this didn't impress me. Having said that, 2 days ago I didn't even know if it was possible to cut this element, so a difficult to polish face is better than nothing.



Above you can see the glory of my incredible coding abilities! All those steps were mathamatically worked out in SLDWRX and coded straight in to hollow out as much material as possible before cutting the final profile.


What an amazing sight. Okay maybe i'm getting a little sentimental about an inanimate object, but for someone who's been staring at these profiles as lines on a screen for months seeing it cut in real life against the odds of everything going wrong, it's pretty satisfying. Although.... can you see in the centre? Cutting marks, eurgh I hate polycarb.


Parted off, reversed, tapped into the 50mm collet, and straight on with the next code file. No problems there, easy cut and the collet held as it should. Element complete! It's amazing how fast things can go when you have all your problems already ironed out. Next was a very similar looking element, but on the opposite side of the formula, with a smaller diameter.



However, with one side cut and then parted off, I couldn't just dump it in the machine. This element has a diameter of 47mm, designed to fit into a removable barrel inside the inner barrel. Cutting the outside diameter on the CNC is easy, but I don't have a 47mm collet to hold it in reverse. Without any aluminium of the right dimensions, I turned to my left over lengths of 50mm diam Nylon to make a collet to hold the slightly smaller optics.


Have I mentioned how much I dislike cutting Nylon? Hot, stringy, generally poor surface quality. Taking the centre out with a drill bit caused a fair bit of heat. Yes that is the drill steaming from the lubricant boiling on the tip.


Parting off is even worse. The material just glues itself to the tools, as you can see below. It might not look to bad in the photo, but if you remember a couple of days before this exact thing caused my acrylic billet to smash and come flying out of the lathe.



It's always a good rule of thumb to check your tools. I show lots of pictures of the me and the machines working away, but I spend just as much time setting up tools and machines to do what I want. I swap tools between lathes all the time as I hope from lathe to lathe cutting things that have different requirements. Just as an example, here is how I set up the boring tool to cut on the biggest machine we have, a Harrison metal lathe. The only reason I took photos of this is because I was about to use the tool when I noticed it was way off. This is the easiest way to either cause an accident or bugger your part up. Always check! Don't trust the person before you, even if it was yourself!


To get the boring tool set up I was taking a tool from a smaller lathe which had been ground down to give a long shank and a small cutting tip, ideal for my purpose. Having removed it from the holder and mounted it into one of the Harrisons tool blocks, I locked it into the tool post and swung it round backwards. I then jammed a revolving centre into the tailstock and slid it upto the tool. The tip of the revolving centre should be exactly, surprise surprise, in the centre. The boring tools tool block can then be adjusted vertically to meet point to point, ensuring it's at exactly the right height. It might not seem like a big difference, but it could cause real problems if I were to run the boring tool lower than it should be, it could easily throw the part out.



With the tool ready I cut the step as accurately as possible. The step was exactly 47mm diam, meaning the elements will push tightly in and hopefully be sturdy enough to go back into the machine in reverse and not be ripped out.



No problem! I went ahead and cut the reverse face. One interesting thing about using nylon instead of aluminium as a collet is that you can pretty much use it as softjaws, freely cutting into them along with the part. In this scenario, the cut that was required went beyond the limits of the collet, and so the front face was trimmed off along the same profile as the element. The materials are so similar that it didn't have any negative effect.


Look it's me!


I take a lot of photos of machines spinning and parts looking shiney. But in reality this is what I spent most of my time looking at. Polishing. P1000 6000 12000 and then paste, then buffing wheel. And then again once you realised you missed a scratch. It takes foreeeevvvvveeerrrr and is the most boring process in the world, hence there not being many pictures of it.


Next was the difficult concave-concave lens. This one is scary because although it looks thick from where I am parting it off in the image below, the gap between the surfaces at the very centre is going to be just over 1mm. Not a lot of room to go wrong then...


The Nylon collet I made was tight, very tight. Rather than widen it a little and risk messing it up, I decided it couldn't hurt to be tight and secure, so I gave it a couple of convincing taps with a spanner.


The element cut fine, and it was impressive to see how thin it actually was in the centre. Again I was dissapointed by the cutting lines being evident. It was only happening on the polycarb parts. In the future when I have more time I think I'll test out some feed speeds to find a better optimum.

3 elements in a day, all the lenses for the 90.2 lens cut, good going! Now isn't it time something went wrong? Last job of the day, make the removable barrel which wil hold half the elements can be pulled out of the back of the inner barrel. Oh I know! Why not mangle it in the lathe!


Honestly, the easiest job just go up the sodding wall sometimes for no reason. The tubing was 1.5 mm wall thickness. To hold that in the chuck you had to be realllll careful. The chuck could easily deform it and turn it into a triangle with any sort of pressure. So I tightened it just enough as I thought I could get away with it holding but not distorting. Haha I was wrong. It didn't hold in the chuck, it caught on the tool and thought hmm, lets buckle and mangle myself. Seriously annoying at the end of a day. Infact it was so annoying (it was the final part needing to be made before final polishing and assembly could start on the 90.2 lens) that I ended up staying behind and finishing it.


In the end I got a part cut (although I had to turn it down on a lathe which was gears to run soooooooooooo slowly that the belt actually just stopped at one point, and I have no idea how to re-gear it). I had to sand it down a tiny tiny bit for it to push fit into the inner barrel without falling out. Advice to anyone ever, don't sand aluminium. I only did it because I had no time and no patience left. Aluminium dust embeds it's self in your skin, turns everything a deep dark grey, and dear God don't breathe it in! Anyway, at the end of the day I had completed a huge amount and was finally ready to go into final assembly mode for the 90.2 lens.