CNC metal part wanted

Discussion in 'Hobby Engineering' started by akashra, Nov 16, 2019.

  1. akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    I'm looking to get a part created as a prototype which is extremely simple, but has to be extremely precise.

    The part I need has to be a perfect 30mm hemisphere to fit in to a cavity this size, and have some kind of small handle to be able to hold/remove it once inserted. This can be welded on if needed. It needs to be extremely well polished and smooth. It would need to be made of something non-corrosive like stainless steel.

    I do not know the tolerance of the 30mm in this case or how exactly close to 30mm it is (eg, 30.10, 29.80)- which is what I need to find out - so for this it might be ideal to have multiple parts made in very close sizes.

    Do we have anyone on OCAU that could help with this, or is this something I would need to contact an industrial company to get this made? If so, what do we think this might cost - both to prototype and as a manufactured part once I know the exact tolerances required?
     
  2. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,114
    If you know exactly what you want - in measured terms - then you might be able to put together a drawing and get quotes/purchase from CNC machinist workshops.

    But if your OP is more or less the only details you have to work off of, you either need to take it to the above step and take a risk or engage a mechanical eng consultant to get it to the above step.

    May be some people here that are able to offer their help in both cases.

    For example, perfect? What is perfect? 0.01 mm? 0.005 mm ? 0.000000... ? Does this include the surface finish? At what range of temperatures?
     
  3. OP
    OP
    akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    I know exactly what I want in terms of object shape, I just don't know how precise the current figures I have are or the tolerances. In fact since writing this up I've come up with an even easier model that uses less materials.

    The measured size needs to be after the object is polished. The range of temperatures it would be used at is 18C-31C - typically 18C-21C for the surface, but in contact with a single drop of liquid at 26-34C which is intended to form a pattern/effect when it spreads between the stamp and the mold.
     
  4. OP
    OP
    akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    Does this diagram help?
    [​IMG]
     
  5. Hater

    Hater Member

    Joined:
    Nov 19, 2012
    Messages:
    4,521
    Location:
    Canberra
    would a 30mm ball bearing with a neodymium magnet stuck to it for lifting purposes suffice?
     
  6. OP
    OP
    akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    It might... but it'd have to be a perfect fit for the mould.
     
  7. aXis

    aXis Member

    Joined:
    Jun 27, 2001
    Messages:
    5,447
    Location:
    Kalgoorlie, WA
    Can you instead pour something into the mould to make an exact negative?

    Also if the mould can be modified slightly, I would suggest using a fine abrasive between the stamp and the mould and then moving or spinning the stamp so that they each polish each other.
     
    Last edited: Nov 20, 2019
  8. OP
    OP
    akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    That is currently how we do it - pour chocolate in it, let it temper, and remove it with a toothpick or cable tie that'd been left in it before it tempers. But it's not re-usable - you can't clean it without affecting the surface etc. But I'm not going to pour something in to it that would damage it, alter it, or make it no longer safe to use. You absolutely don't want to use anything abrasive in the mould.
     
  9. underskore

    underskore Member

    Joined:
    Nov 5, 2002
    Messages:
    4,236
    Location:
    3198
    this is really a job for a very time generous backyard machinist with serious toolmaking skills or something for any reasonable CNC shop out there.

    am I reading your diagram correctly in that you want a ~5mm wall thickness?
    you probably don't need anywhere near the accuracy you think you might

    I can point you to a couple of smaller shops that might not punish you on the price for this kind of job (both out Hallam way)
     
  10. Butcher9_9

    Butcher9_9 Member

    Joined:
    Aug 5, 2006
    Messages:
    2,182
    Location:
    Perth , East Vic Park
    Cant say for sure but I think this is what the OP is after or something similar. Part hold down would be pretty tricky.

    Fusion files and STL - https://www.thingiverse.com/thing:3990905
    [​IMG]
     
  11. mick41zxr

    mick41zxr Member

    Joined:
    May 27, 2008
    Messages:
    10
    This job could be done quickly & easily on a lathe with a ball cutting attachment or CNC lathe.
    If its for food use 304 or 316 stainless should suffice. You need to determine your tolerance but most shops should be able to hit 30.00mm +/- 0.05.
    Regards,
    M
     
  12. OP
    OP
    akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    Wow. Yes, that's pretty much exactly what I'm after. Notwithstanding whether the suggestion of just attaching a neodymium magnet to a pre-fabbed 29.80 ball bearing doesn't do the job.

    5mm was just an example. It deosn't really matter so long as a persons fingers can somehow grip it.
     
  13. underskore

    underskore Member

    Joined:
    Nov 5, 2002
    Messages:
    4,236
    Location:
    3198
    this is getting a bit off track now but are the thermal characteristics of this steel "ball" going to be an issue for the process?
     
  14. Cererus

    Cererus Member

    Joined:
    Jan 14, 2002
    Messages:
    1,556
    Location:
    ≥∞
    Ball bearing 'balls' are hardened steel and will corrode, you can get 316/304 stainless loose balls from various places e.g.

    https://www.alibaba.com/product-det...m=a2700.7724857.normalList.186.3b137821tIxEg0

    Also ring round local industrial bearing shops who might be able to source one.

    You can then find a machinist to drill a hole into it and glue or braze a rod as a handle, you can drill into ball bearings with a carbide drill using a lathe (i'm an ex fitter/machinist)

    I'm still a vague as to why it has to be so polished you're making hollow chocolate hemispheres which get filed and then joined together, no one will see the inside surface ?
     
  15. OP
    OP
    akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    Doubt it. You're typically working between 15 and 31 degrees, so unless there's significant expansion between those temperatures over a period of a few seconds, no. It will literally be used to insert in to a mould for a second or two max. Heat transfer would only be from cocoa butter tempered and dropped at 29 degrees celsius.

    Firstly, I'm not joining them, but that's kinda irrelevant to this, nor am I interested in the inside of the shell. The shell still gets filled with ganache, caramel, praline, gianduja - whatever - and then capped.
    Tempered chocolate contracts - that's how it is able to release from the mould. It's also part of what gives it that perfect shine when it's correctly tempered.
    One particular artistic effect people like to use with coloured cocoa butter uses that tiny amount of space between the mould and sphere surface to create a feathered effect. Too much space and you just get a smooth, even flow, and no effect.

    Usually, to do this effect, you pour chocolate in to the mould, let it temper (so it contracts and releases), drop in a single little drop of tempered cocoa butter, let it force itself in to tiny, river-like lines where it can fit, then pull that tempered hemisphere out again. What you're left with is a pattern of coloured cooca butter on the shell. You then spray with another colour or, if the colour you used has enough reflectiveness in front of the colour you're filling with, just fill directly behind that pattern. But you can only use each tempered piece once - you can't clean it, you don't get the same effect if you try to use the same piece for more than one cavity.

    The steel object is to be used to replace that chocolate piece you're using as a stamp.
     
  16. underskore

    underskore Member

    Joined:
    Nov 5, 2002
    Messages:
    4,236
    Location:
    3198
    I'm not concerned about expansion, moreso how the heatsoak could change the tempering process steel tends to suck the heat out of anything.

    I'd buy any old steel ball off eBay and a magnet to do some proof of concept testing, and keep in mind that there's a good chance a stainless ball won't be magnetic.
     
  17. OP
    OP
    akashra

    akashra Member

    Joined:
    Apr 25, 2003
    Messages:
    3,799
    Location:
    Melbourne, AU
    Heat loss is okay - it's tempered by heating to 40-45 degrees, cooling to 27.0 degrees with movement, then heating back to 28.5C (28.0-29.0) with movement which it's kept at as a working temperature. So long as it's put in the mould at 28.5C and then only cools from there without movement, heat-loss is fine because at that point you're now bringing it back to 18C.
     
  18. psychobunny

    psychobunny Member

    Joined:
    Mar 30, 2006
    Messages:
    2,383
    Location:
    Melbournish 3750
    totally agree, turn half a ball on the end of a rod, then cut a post in behind it. if i still had my lathe i'd do it
     

Share This Page

Advertisement: