5 ms·

I really think that every well rounded person/STEM academic should at one point in their life (hopefully sooner rather than later) expose themselves to some han

by s5300 5y ago

I really think that every well rounded person/STEM academic should at one point in their life (hopefully sooner rather than later) expose themselves to some hands on machining... like, even just a few hours on an old decrepit mill/lathe. I feel like the foundations of the industrial revolution (speaking within the US) are... very much forgotten/taken for granted given the fact the near entirety of our modern life and comforts exist because of it. There's fortunately a ton of great, readily available content on YouTube/elsewhere that's sprung up in the past 10 or so years you could just watch, but actually machining is such a cathartic experience... at least when you're in a position to do so that isn't in a grimy ass factory/warehouse with somebody yelling at you to push out more operations than is humanly possible, haha.

prpl 5y ago
Most working undergrads (RAs) in the physics department at my university had been trained in the Physics department machine shop at some point in time, so it was common for them. Probably the only program outside ME where that was true
- R0b0t1 5y ago
  Not sure this is normal. Have heard stories of physics department needing to contract to the engineer's union, and people getting in trouble for screwing together a wooden box.
  - jbay808 5y ago
    That's the scenario that I'd call abnormal! What a nightmare!
    - R0b0t1 5y ago
      Seems fairly normal from what I can tell. Issues like needing everything to be vetted my a mechanical/civil/electrical engineer, depending on what it is, and not being able to machine your own parts and needing to let the university machine shop bill time.
      - s5300 5y ago
        One of the funniest things I can remember about uni shop bill time was that some aerospace engi's needed some really special washers. They were quite expensive and the shipping time was going to cut a deadline extremely close. So they were able to send one of the grad students who watches over the after hours student machine shop a few hours away to a place that had the metal needed in stock and then have them CNC all the washers. They charged typically charged a good bit for physics/aerospace dept. to use said CNC, still ended up being much cheaper and much faster... for simple washers.
- azalemeth 5y ago
  Me too -- I had a 30 hour course in the workshop as a graduate student in physics. We covered both manual and CNC milling and turning, with Bridgeport mills and Colchester lathes. Completely opened my mind. Made me a better experimentalist. I spent about 99% of the time utterly terrified and wanting to check everything about ten times to make sure that I didn't wreck either the job or myself.
mmaunder 5y ago
CAM is also humbling and then exhilarating. And agreed, everyone should do it at least once.
ganzuul 5y ago
> expose themselves to some hands on machining... like, even just a few hours on an old decrepit mill/lathe. Excuse me but as a pro in this field I have to raise an objection. There are aptitude tests just for the education because some people who are otherwise brilliant are absolute terrors, a dangers to themselves and others when presented with machine tools. Trial and error works very differently when you focus enough pressure into carbide to cut steel like it was butter. You should watch the horriffic videos of people being turned into bloody messes on machine tools and THEN decide if this is something you want to try. Nobody knows in advance if they belong in the danger-category but school instructors are quick to spot them and talk them out of the field. If you are adamant on going at it on your own, stick to tools which don't output more than 1000W. In a panic you can produce that much power and you might only lose a finger instead of your life. ALWAYS pause and think about what you are going to do the very next moment and what you expect will happen. I'm sorry if it sounds like gatekeeping. Those videos are very convincing that the gate is there for a reason.
- quartesixte 5y ago
  I’m pretty sure a few hours on a Haas VF3 or ST20 running a simple part under supervision will be safe enough. I don’t think anyone in their right mind will be handing over the keys to a roomsized VTL any time soon...hopefully... Manual machines on the other hand—absolute death traps.
  - ganzuul 5y ago
    Getting your hands on a barely serviceable manual machine is easy though, which is why I started worrying. If the machine is out of production there might be something seriously wrong with it and if a novice starts compensating for the fault then the margin of error approaches zero. Compund this with youtubers not uploading when they made a dangerous or embarrassing mistake and things start to look a lot more doable han they actually are.
  - auxym 5y ago
    I manage a fablab used by grad students. We have both manual and CNC machines. Haas, too, but the cheaper stuff (TM and an older OM, VF stuf is $$$ for research budgets). The eternal compromise is this: it's way easier to dismember or kill yourself on a manual machine (https://www.nytimes.com/2011/04/14/nyregion/yale-student-dies-in-machine-shop-accident.html https://www.nytimes.com/2011/04/14/nyregion/yale-student-die...). However, it's even easier to seriously fuck up an expensive CNC mill with a programming mistake. Serisouly, these machine can and will destroy themselves if you tell them to, no questions asked or warnings. One single bug in your g-code and and there goes a 15K$ set of spindle bearings or worst. Of course, a few tens of thousands down the drain is still better than dismemberment. But that also means that sometimes our machines are down for months or years until we can get a new grant to fix them.

by s5300 5y ago

prpl 5y ago

Most working undergrads (RAs) in the physics department at my university had been trained in the Physics department machine shop at some point in time, so it was common for them. Probably the only program outside ME where that was true

R0b0t1 5y ago
Not sure this is normal. Have heard stories of physics department needing to contract to the engineer's union, and people getting in trouble for screwing together a wooden box.
- jbay808 5y ago
  That's the scenario that I'd call abnormal! What a nightmare!
  - R0b0t1 5y ago
    Seems fairly normal from what I can tell. Issues like needing everything to be vetted my a mechanical/civil/electrical engineer, depending on what it is, and not being able to machine your own parts and needing to let the university machine shop bill time.
    - s5300 5y ago
      One of the funniest things I can remember about uni shop bill time was that some aerospace engi's needed some really special washers. They were quite expensive and the shipping time was going to cut a deadline extremely close. So they were able to send one of the grad students who watches over the after hours student machine shop a few hours away to a place that had the metal needed in stock and then have them CNC all the washers. They charged typically charged a good bit for physics/aerospace dept. to use said CNC, still ended up being much cheaper and much faster... for simple washers.
azalemeth 5y ago
Me too -- I had a 30 hour course in the workshop as a graduate student in physics. We covered both manual and CNC milling and turning, with Bridgeport mills and Colchester lathes. Completely opened my mind. Made me a better experimentalist. I spent about 99% of the time utterly terrified and wanting to check everything about ten times to make sure that I didn't wreck either the job or myself.

mmaunder 5y ago

CAM is also humbling and then exhilarating. And agreed, everyone should do it at least once.

ganzuul 5y ago

> expose themselves to some hands on machining... like, even just a few hours on an old decrepit mill/lathe. Excuse me but as a pro in this field I have to raise an objection. There are aptitude tests just for the education because some people who are otherwise brilliant are absolute terrors, a dangers to themselves and others when presented with machine tools. Trial and error works very differently when you focus enough pressure into carbide to cut steel like it was butter. You should watch the horriffic videos of people being turned into bloody messes on machine tools and THEN decide if this is something you want to try. Nobody knows in advance if they belong in the danger-category but school instructors are quick to spot them and talk them out of the field. If you are adamant on going at it on your own, stick to tools which don't output more than 1000W. In a panic you can produce that much power and you might only lose a finger instead of your life. ALWAYS pause and think about what you are going to do the very next moment and what you expect will happen. I'm sorry if it sounds like gatekeeping. Those videos are very convincing that the gate is there for a reason.

quartesixte 5y ago
I’m pretty sure a few hours on a Haas VF3 or ST20 running a simple part under supervision will be safe enough. I don’t think anyone in their right mind will be handing over the keys to a roomsized VTL any time soon...hopefully... Manual machines on the other hand—absolute death traps.
- ganzuul 5y ago
  Getting your hands on a barely serviceable manual machine is easy though, which is why I started worrying. If the machine is out of production there might be something seriously wrong with it and if a novice starts compensating for the fault then the margin of error approaches zero. Compund this with youtubers not uploading when they made a dangerous or embarrassing mistake and things start to look a lot more doable han they actually are.
- auxym 5y ago
  I manage a fablab used by grad students. We have both manual and CNC machines. Haas, too, but the cheaper stuff (TM and an older OM, VF stuf is $$$ for research budgets). The eternal compromise is this: it's way easier to dismember or kill yourself on a manual machine (https://www.nytimes.com/2011/04/14/nyregion/yale-student-dies-in-machine-shop-accident.html https://www.nytimes.com/2011/04/14/nyregion/yale-student-die...). However, it's even easier to seriously fuck up an expensive CNC mill with a programming mistake. Serisouly, these machine can and will destroy themselves if you tell them to, no questions asked or warnings. One single bug in your g-code and and there goes a 15K$ set of spindle bearings or worst. Of course, a few tens of thousands down the drain is still better than dismemberment. But that also means that sometimes our machines are down for months or years until we can get a new grant to fix them.