Author Topic: The CNC experiment build.  (Read 50665 times)

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #75 on: February 25, 2015, 06:25:35 PM »
I need to pick your brains, and interest i guess.

So that 5.5kW motor on top of the head is weighing in at 45kg, the spindle is 35kg, the head is about 35kg, and the baseplate is around 20kg. So about 135kg.
The pitch on the lead screw is 5mm, its a 32mm screw and the head runs on 30mm heavy load linear bearings.. heres is the kicker..
The servo motor is a 1500W motor, that can hold 6Nm, and accelerate to about 18Nm, in my application the real acceleration is about 11Nm.

The question is.. would my servo motor be able to move that head without a counter weight. Or should i start thinking of redesigning the column in such a manner that counterweights could be added.

Im considering going with a 4kW motor instead and save 15kg, but - that would make me loose about 2Hp. I have decided i will go with a ABB motor tho - not just sure what size yet.


Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #76 on: February 27, 2015, 06:38:26 PM »
So i had some issues with the idea of 130kg milling head. But i got this formula off of a friend.
6Nm / (0.005/2/pi) * 0.8 = 6032N

That basicly means that in the lowend of the servo motors i can accelerate 600kgs with somewhat ease. the motors is 2600rpm, but lets say you have friction some strain and other issues, so it drops to around 2400rpm. Means if i work within the 6Nm holding power range of the motor, that 600kgs would be able to accelerate with 12m/s in theory. My motors can accelerate 18Nm, but with Amperage being an issue there im prob. limited to about 11Nm.

Still i dont think 130kg of milling head will be an issue. Not even sure i would need a counterweight actually. But i should prob. gear it down with some form of beltdrive to 2:1 ratio so i can actually utilize the theoretical acceleration.  *drawing new table mounts*
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline Brass_Machine

  • Administrator
  • Hero Member
  • *****
  • Posts: 5504
  • Country: us
Re: The CNC experiment build.
« Reply #77 on: February 28, 2015, 02:02:05 AM »
You could always put in your design an easy way to add a counter weight if needed. Even if it is only to allow a spot of cables to connect and an easy way to mount the weight. Better to design for it now than have to hodge podge it later.
Science is fun.

We're all mad here. I'm mad. You're mad.

Offline PekkaNF

  • Hero Member
  • *****
  • Posts: 2523
  • Country: fi
Re: The CNC experiment build.
« Reply #78 on: February 28, 2015, 02:27:38 AM »
I would design in gear or timing belt, least that gives you an option and there is always a possibilyt for a bigger servo.

But you definately need a brake that is spring operated and activates when power goes off. You might calculate how much the ball screw will hold when motor is off. My guess is that head weight would drive the motor and smash head down if fuse blows or something else would deactivate the servo motor.

If I understand it right couterweight is an option for slow machine. Counterweight adds mass, which you need to accelerate.

That might work out out, but if you need speed, you have a risk. If you have chains or cables AND you push the head down first (and shoot weight up) and then change direcion really fast you might hit it hard on rebound. Saw that happen once when masses were someting like 5t and and little smaller counterweight. Really big chains snapped and both parts crashed down.

Pekka

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #79 on: February 28, 2015, 06:38:07 AM »
Ah well i will prob. never go for a larger servo. The servos is 1.5kW in a 130mm frame. check my blog or earlier pictures i think i have them compared to a monster can of energy drink.
And im at the max for the servo drivers i bought.

But with that said, there is some deisng hinkys with putting a counter weight into this atm, for one i think i need to make the column wel mirrored, the outtake in the back atm, would be two outtakes on the sides instead and then i could prob. design some plate, gears, and chains.

The other way i think is to use wires and pulleys.. Where the weight hangs from a single pulley, and one end of that wire is stationary and the other end is connected to the faceplate for the head.
That could prob. be a hodge podge solution tho..

And as Pekka mentioned having 260kg accelerated will make for some funny behaviors when doing rapids. A friend of mine solve that by mounting the weight to a rail and using a brake on the slide, with no bearings, so when head is moving down it has some extra force needed for moving it down. guessing this could be achieved by making the weights between the head and counter different, but in reality you want to be able to move up faster than you wanna move down - at least in my head.

The third option i have been thinking of is some kind of "air spring" by utilizing like a fire extinguisher tank pressure lined and some form of piston, one could in theory create an airspring, but how practical is that that is the question i guess. =)

I need to take me a ponder...  :doh:
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline PekkaNF

  • Hero Member
  • *****
  • Posts: 2523
  • Country: fi
Re: The CNC experiment build.
« Reply #80 on: February 28, 2015, 08:50:28 AM »
Holy crap 1,5kW servo. That is huge on on hobby size macine. It can some serious damage! Just gear it down, you will not need any couter weight. You need HW limit switched and one brake on lead screw to prevent the head from dropping when you power off.

Pekka

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #81 on: February 28, 2015, 08:53:30 AM »
Ah well its not a hobby sized machine it will be about 800 kg and half cubic meter of work area. And yes the driver takes hw limit switches and home switches Linux cnc take signal on the field bus as well so i will piggy back the hw signal to the field bus. :)
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #82 on: March 02, 2015, 04:42:32 PM »
So spent 3 days tinkering with the tramming possibility..

So basicly, 22mm holes, with a 16mm bolt, head is sitting on a 60mm dowel, and has a little "cam" that allows for tilting the head ever so slightly. That cam only has about 2.5mm radius, so it will not allow for alot of adjustment, but enough i think.
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #83 on: March 16, 2015, 04:12:54 PM »
So while waiting for the handymans to complete my new workshop floor, i continued making drawings..

Molds for base and pillar, so i know how much sheet will be used..
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

RobWilson

  • Guest
Re: The CNC experiment build.
« Reply #84 on: March 17, 2015, 04:22:11 PM »
Hi Neo


What do you reckon the column casting will weigh ?



Rob 

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #85 on: March 17, 2015, 06:10:15 PM »
A rough guestimation puts it in the 350kg range, and the base at around 400kg, table in and around 75kg and the head casting around 90kg..

The idea with the molds is to brace them with 45x90 and load straps for cargo.
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #86 on: March 25, 2015, 04:26:12 AM »
So the BT40 spindle has arrived. It is suprisingly good quality. Wit a rod mounted in the spindle that sticks 150mm out from the spindle the throw is less than 1.5/1000mm
I spins smoothly with some "noise" im guessing it need lubricant. With it i got a 64 tooth pulley for the motor and 3 ER32 holders.
Final price with shipping for one of these is approx 1480usd - and thats with DHL Express shipping from china to Sweden (arrived within 5 days).



Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

RobWilson

  • Guest
Re: The CNC experiment build.
« Reply #87 on: March 25, 2015, 05:41:06 PM »
 :drool: now that looks a real nice bit of kit  :thumbup: 

Were did you get the spindle from ?


Rob

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #88 on: March 25, 2015, 05:49:39 PM »
It's one of those Chinese manfactured OEM spindles you can find on alibabs.com . this one is a 130mm housing with 5 bearings. Which 4 them is angular contact bearings. Max rpm is 6000 and im guessing it needs some form of cooling. Need to ask the agent. :)
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline tom osselton

  • Hero Member
  • *****
  • Posts: 1256
  • Country: ca
Re: The CNC experiment build.
« Reply #89 on: March 25, 2015, 05:57:57 PM »

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #90 on: March 26, 2015, 03:01:53 AM »
Ah well yes i did look at the Tormach Spindle at first, BT30 is nice size as well but a bit small for the rest of what im designing.
And even tho this BT40 spindle cost twice as much as the Tormach, the shipping was 1/3 of what Tormach wanted. I paid less than 300 usd for shipping, and Tormach wanted close to 700 for the BT30 spindle in shipping costs only.

And then im no fan or friend of Tormach actually. Find them to be a BS company, slapping their brand on chinese parts and bragging about how they did the RnD when its obvious OEM parts rebranded for the most of it. Take their "new" pathpilot for example - which is mesa cards (standards) with linuxcnc in the bottom. And the RnD was done by an external contractor. That type of marketing may work in the states, but doesn't work around here. And well are even illegal in my country. =)
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline vtsteam

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6466
  • Country: us
  • Republic of Vermont
Re: The CNC experiment build.
« Reply #91 on: March 26, 2015, 08:33:32 AM »
$700 shipping!  :bugeye:  :bugeye: :bugeye:
I love it when a Plan B comes together!
Steve
https://www.youtube.com/watch?v=4sDubB0-REg

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #92 on: March 26, 2015, 08:39:04 AM »
Well im guessing its an effective way of not selling overseas. Shipping a PCNC1100 would run me about 6000usd.. I exhaused those things before i decided to try this build.
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline philf

  • Hero Member
  • *****
  • Posts: 1115
  • Country: gb
Re: The CNC experiment build.
« Reply #93 on: March 26, 2015, 08:47:33 AM »
$700 shipping!  :bugeye:  :bugeye: :bugeye:

Steve,

I don't know why the cost shipping stuff from the US is so high but it seems out of all proportion. It puts me off buying anything.

I've sold microscope objectives to the US and it's only cost me a few Pounds Sterling to send from the UK. Buying the same thing from the US might attract shipping of $50 or more ( and then when it arrives add import duty and VAT) - it's just not viable.

Phil
Phil Fern
Location: Marple, Cheshire

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #94 on: March 26, 2015, 08:51:12 AM »
Yeah i bought a tube bender  :offtopic: it cost like 135 dollar - and the shipping was 48 dollar, and then came 13% customs on that and 25% VAT
In this case it even was a refurbished one, for bending stainless tubes..
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline vtsteam

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6466
  • Country: us
  • Republic of Vermont
Re: The CNC experiment build.
« Reply #95 on: March 26, 2015, 08:54:43 AM »
Yikes! For $6000 in shipping, one might as well have a vacation workshop abroad and ship there for $300. Land cost us $1200/acre, and there were trees to cut lumber to build. Of course I don't recommend the winters!  :lol:
I love it when a Plan B comes together!
Steve
https://www.youtube.com/watch?v=4sDubB0-REg

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #96 on: March 26, 2015, 09:01:42 AM »
Well im swedish.. winters doesnt really bother me, neither does cold or snow.. im used to it. =)
Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

Offline vtsteam

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6466
  • Country: us
  • Republic of Vermont
Re: The CNC experiment build.
« Reply #97 on: March 26, 2015, 09:12:50 AM »
It's the drive toward "premium" shipping that has created an insensitivity to shipping costs. There is often an inexpensive alternative, but manufacturers don't realize it (often not advertised well by the shiping companies) or are afraid of damage (not that there is actually any less liklihood with premium shipping).

This occurs domestically here as well. Now most Ebay sellers, for example send by a premium rated "Priority Mail" for small parts. This rate is often triple what "First Class" mail would cost for a small package under a pound in weight. The latter goes by airmail, and often arrives sooner than Priority Mail. Yet you are hard pressed to find First Class mail rates in signage at the post office, or online at the USPS website.

Try it on the USPS website - you'll see -- almost impossible to find first class mail for packages either domestic or overseas, yet it steers you into Priority Mail as quickly as possible. Check the difference in rates for sending something to Sweden, for instance between Priority Mail and First class.

Obviously you can't send a mill that way, but it's just an example of how many manufacturers don't actually get to see the difference in shipping rates available -- often it is office staff who unknowingly figure extreme shipping costs on an item, or even a web link to the shipping company's calculator application that does it.

For a high cost item, If you do your homework, find a good shipping company, and get a good rate, you can often contact a manufacturer by phone, and ask them if they will ship that way., and I bet 90% would agree to do that, if it's a no brainer for them.

Can't do anything about customs, or tariffs, but that's a local deal.
I love it when a Plan B comes together!
Steve
https://www.youtube.com/watch?v=4sDubB0-REg

Offline NeoTech

  • Hero Member
  • *****
  • Posts: 826
  • Country: se
    • Roughedge Hobbyworks
Re: The CNC experiment build.
« Reply #98 on: April 14, 2015, 02:10:50 PM »
So not been slacking off.. i needed some more space.. so well poured a concrete floor and made some walls..

Machinery: Optimum D320x920, Optimum BF20L, Aciera F3. -- I have not failed. I've just found 10,000 ways that won't work. http://www.roughedge.se/blogg/

RobWilson

  • Guest
Re: The CNC experiment build.
« Reply #99 on: April 14, 2015, 03:18:17 PM »
 :clap: :clap: :clap: I like the colour Neo  :)



Rob