Stainless Steel – Quick and dirty flex test

New Home Forum Mostly Printed CNC – MPCNC Hardware Development – MPCNC Stainless Steel – Quick and dirty flex test

This topic contains 39 replies, has 17 voices, and was last updated by Profile photo of Dui, ni shuo de dui Dui, ni shuo de dui 5 months ago.

Viewing 30 posts - 1 through 30 (of 40 total)
  • Author
    Posts
  • #14560
    Profile photo of Ryan
    Ryan
    Keymaster

    I bought some 1″ OD stainless Steel this week, and the flex difference is pretty obvious. I didn’t setup the dial indicator but took some quick pictures.

    I had 6′ lengths of tubing supported on both ends and hung a 10lb weight form the center point. I took a picture from as level as I could get to illustrate the flex.
    The standard EMT that we/I have been using was a bit more rigid than the .0185″ stainless curtain rod (too weak the bearings left a groove on the first pass).

    The 316l stainless did better what I tested what was sold as .049″ and .083″ wall thickness. The .049″ is much better than the EMT but really close to the .080″ and about $10 cheaper per 6′.

    I would recommend .050″ if you want to upgrade your machine and have some money to spend.

    #14635
    Profile photo of Jeffeb3
    Jeffeb3
    Participant

    Where do you find stuff like that?

    #14641
    Profile photo of Ryan
    Ryan
    Keymaster

    I found some at a hardware store (shower curtain rod), amazon, and a local supplier.

    #14642
    Profile photo of Benjamin Shaw
    Benjamin Shaw
    Participant

    I’m strongly considering ordering some of this unless I just go with regular emt.

    #16501
    Profile photo of Sergey Kovalyov
    Sergey Kovalyov
    Participant

    I’d like to point out that, for some reason, 0.065″ wall thickness 304 stainless steel tubes are about half the price of 0.050″ ones, at least at speedymetals.com (http://www.speedymetals.com/c-8276-round-tube.aspx?thickness=1). They seem to be the lowest price I could find – a quarter per inch or like $10 for 4′. 0.050″ on the other hand is $0.48 per inch, $19 for 4′.
    I ordered custom cuts from them as I’m building an odd size 27″ x 40″ (to match a space under my desk where my MPCNC would have to be stored between uses). A bit more expensive than buying whole-feet size but beats having to invest in a tube-cutting tool and cut yourself.
    Shipping was $20 though.

    #20386
    Profile photo of James
    James
    Participant

    1″ or 1.05″ OD SS Tubing Is 3/4 ID If i’m not mistaken… a quick search on amazon also confirmed that and really 1″ OD was 3/4 ID.. but this 316L stuff is $100 for a 3 footer……. that’s crazy expensive… any chance anyone’s used 304? I cannot get 316L locally and people want crazy money for it.

    #20389
    Profile photo of Barry
    Barry
    Participant

    Uhh, if it’s 1″ OD and 3/4″ ID, then it’s 1/4″ thick wall! It’s crazy expensive because it’s crazy thick!

    #20411
    Profile photo of Walter
    Walter
    Participant

    wuddn’t that be 1/8″ wall thickness barry?

    1/8″+1/8″ =1/4″
    1″- 1/4″ = 3/4″

    ur forgetting that its tubing so you would split the difference in half to find the wall thickness

    #22726
    Profile photo of John
    John
    Participant

    I just ordered some 304. When it gets here, ill do the a test and provide some pictures. I don’t have any 316l to compare it to, but maybe this info will help.

    Thanks,
    John

    #22730
    Profile photo of Barry
    Barry
    Participant

    @Walt, yea, you’re right! I ain’t got no good math, or english! šŸ˜‰

    #24712

    What are you thoughts on tweaking the files and running 3/4 Gas Pipe or Rigid Conduit, both are 1.050″ Diameter. Should be a lot stronger than EMT and still reasonably priced (about $10/10 ft stick at any Menards/home depot). Or does the extra weight start to become an issue?

    #24726
    Profile photo of Barry
    Barry
    Participant

    Yea, now you’re throwing around a lot of extra weight.

    #24727
    Profile photo of Walter
    Walter
    Participant

    Plus all that extra weight actually makes the pipe more prone to sagging not less. what you want is added stability not added weight. while stainless is slightly more heavy than zinc plated, dots extremely more rigid along its span. there’s always going to be a trade off.

    #24731
    Profile photo of autox3d
    autox3d
    Participant

    my local source 1.5mm wall 304 stainless is really cheap. around 30 bucks for 6 meters.

    #25859

    On my machine, I use tubing initially designed for guide rails in the industry. Very heavy but it is impossible to bend. You might want to consider this option for a CNC if you have big loads.

    #25938
    Profile photo of Ryan
    Ryan
    Keymaster

    Heavy=no good

    #27394
    Profile photo of kyle
    kyle
    Participant

    Did 304 SS work out?
    Thank you in advance.
    Kn

    #27397

    Well, it doesn’t bend at all.
    I could compare with standard tubing using weights and a comparator if you want, I think I still have about 30 cm left of both type of tubes, But I’m pretty sure the big ones will win the comparison easily. You can feel the difference just trying to bend them by hand, the heavy ones don’t flex at all. They are made of hardened steel, I guess the difference comes mainly from there.

    Attachments:
    #27760
    Profile photo of Walter Rotter
    Walter Rotter
    Participant

    I presume that it has been asked and suggested many times over:

    What about square tubing?

    The theory:

    I’ve been using this calculator:
    https://www.easycalculation.com/engineering/mechanical/mechanical.php

    Round tubes:
    – steel tube
    – 1″ diameter – I will use international version 25mm
    – 0,08″ (2mm) wall thickness
    – 40″ (1m) long
    – 20 pounds (10kg) weight (it’s just a presumption that 84oz/in steppers could assert such a force on a beam)

    Deflection – 0.577″

    Square tubes:
    – steel tube
    – 1″ x 1″ – I will use international version 25mm x 25mm
    – 0,08″ (2mm) wall thickness
    – 40″ (1m) long
    – 20 pounds (10kg) weight (it’s just a presumption that 84oz/in steppers could assert such a force on a beam)

    Deflection – 0.010″

    Everything’s the same except the shape
    So, either the calculator is wrong (I’ve never used one so I couldn’t tell) or square tubes is far, far better solution.

    Or I’m missing something?

    I’ve checked local prices, square tubes are 20% more expensive.

    Since no one remixed this design for use of square tubes I’m really getting something wrong?

    Thanks!
    Regards,
    Walter

    #27761

    Hi Walter,

    Thanks a lot for this very interesting tool.

    If it is indeed accurate, it seems that wall thickness does actually matter quite a bit, which is in contradiction with what most people seemed to believe here.

    One thing seems strange to me: titanium appears to have more deflexion than steel. Isn’t titanium supposed to have better strenght characteristics than steel? (or maybe it is just far lighter for relatively similar characteristics?)

    Anyway, thanks again for this link

    #27764
    Profile photo of P3DCNC
    P3DCNC
    Participant

    Has anyone tried filling a rod with, I don’t know, concrete? Plaster of Paris? Epoxy? You get the idea.

    #27765
    Profile photo of Barry
    Barry
    Participant

    Filling a tube with something doesn’t make it stronger, just heavier.

    @Dui, ni shuo de dui
    Steel is stronger, but heavier.

    #27767
    Profile photo of P3DCNC
    P3DCNC
    Participant

    Um. You sure about not stronger? Heavier yes. But if you did this only for the frame rails and not the gantry ones…

    After all, those yellow steel tube barriers you see everywhere, they’re filled with concrete for some reason.

    It would be an interesting and simple test would it not? And as a solution, rather inexpensive and easy to implement (no mods to be made).

    #27770
    Profile photo of Barry
    Barry
    Participant

    They’re filled with concrete for mass. We’re also worried about weight. I misspoke about the strength. With a tube and a solid bar of the same material and diameter, the tube will be less strong. The problem is your weight goes up considerably. That’s why race car cages are tubes. Strength to weight is critical. If you’re going to start beefing up this CNC to use solid rods, you may as well build a conventional gantry style CNC. The MPCNC is supposed to be a low cost starter/hobbiest CNC.

    #27782
    Profile photo of Jeffeb3
    Jeffeb3
    Participant

    IIRC, Ryan’s answer to square tubes is that it takes 4 points to align them, instead of 3. Or really, 8 instead of 6. He answered it in the MPCNC 2.0 thread, I think.

    #27790
    Profile photo of Ryan
    Ryan
    Keymaster

    I built a machine that is just strong/rigid/heavy enough for milling aluminum slowly, that is also just light enough to 3D print slowly, and all the other things it can do. It is not perfect it does all kinds of things well but nothing perfect. Yes you can strap an extruder on a $250k mill but it has such large moving mass it will print worse than the MPCNC.

    Tons of reasons for using tubes brought up many times.

    What you are not looking at is mass, round is the strongest for it’s mass. Round is also the most true, square is structural not meant for rails (neither is round but it always comes out nicer). Round is also equally strong in all directions, square is not. 33%more, hardware, more expensive, harder to build, heavier, harder to true without introducing expensive things like eccentric hardware. Square is not readily available in standard sizes. Lots of square tubing machines out there, I didn’t want to build another one. I actually went to dozens of shops of a few months time and put my hands on all the material I could, EMT was far and away the winner in all categories, most importantly price.

    Roll cages are not made of square stock for a reason…Mass.

    Wall thickness, of course it has an effect, that effect diminishes very quickly as you move towards the center. The center actually does nothing. I spent too many semesters learning this stuff. Choices have to be made guys, do you want it ultimately rigid and molasses slow due to moving mass and inertia, or do you want a machined tuned to all of its parts. If you have a perfectly rigid rail system then the plastic is the weak point then you use metal parts and on and on it goes. I feel these tubes match the strength of printed parts wonderfully.

    Filling tubes. Yes filling a tube will make it better, but not very much, largest effect is vibration damning and increasing the weight. This has been asked so many times, I think it is kinda funny no one has sacrifice $2 worth of conduit and just tried it.

    This is to be built with no power tools other than a 3d printer, or buy them if you don’t have one. This is a fun project to use and learn and grow. If you need more spend the money the next options are about $5k. I built a machine you can do all kinds of things with.

    If you change something to make one function better (eg milling) you will sacrifice some other function (laser speed), or more money. Designing a machine has thousands of decisions to make, these are the decisions I have made.

    There are a lot of machines that have been reinforced for milling, but they do nothing else with them. Perfect! If that is what you need go for it, that is kinda the point of me releasing all the info that I do. If I just sold a kit you would be stuck with my choices, you are free to make changes.

    I love and appreciate all of you guys, I know my replies kinda make me sound like a dick but I promise I do not intend it that way. There is just so much information to get out and I have said it in many other places each time it just gets condensed down. Please do not let this discourage you from trying to help make this a better machine. Trust me I have another Idea in my head that One day soon I hope to get started on, lots of things can be changed and I love to think I learn from all of our experiences with the machine.

    Thanks you all again, keep the conversations flowing and I will jump in with my opinions.

    Just a quick story to try and bring this all full circle (still under NDA so I will generalize). At my previous job I made things for people for mass production. They usually brought in a prototype and we took it from there. These people had seemingly random stipulations. One of the last projects had a time constraint meaning it had to move in 5 seconds, really for no good reason, I even felt it was too fast. If they would have said sure two more seconds is fine they would have had a much better product, but choices had to be made. If your garage door opener took 4 more seconds to close but was absolutely silent and the drive was half it’s size would it be worth it? That is designing a machine in a nutshell.

    #27792

    @Barry

    He was talking about filling the frame tubes, not the gantry ones. There is no harm in doing that, it will be a bit stiffer. Actually, even the ones in the gantry shouldn’t be a big deal, we are not running our machines are insane speeds so the inertia is quite limited anyways.

    Either way, I assume that the difference will be hard to notice… If you plan on making your frame stiffer you better just replace your bars with thicker ones directly. It doesn’t cost much and in both case you’ll have to tear apart your machine anyway…

    I don’t know if this is the biggest rigidity issue on your machines anyway. On mine, the weak point will be the middle carriage long before the tubes.

    #27796
    Profile photo of Ryan
    Ryan
    Keymaster

    The weakest part should be the middle x and Y tubes, as there are only one of them, supported by two outer tubes. Making the outer tubes stronger will/should have minimal effect as they are already effectively two times stronger than the tube they are supporting.

    At the same time every little bit helps.

    #27797
    Profile photo of kyle
    kyle
    Participant

    Vicious1 thank you for your designs, and creating this tool. I’m a product developer as well so I completely understand your design process and choices. So I wanted to say thank you for your work and open source approach to this project.
    Kn

    1 user thanked author for this post.
    #27800
    Profile photo of Ryan
    Ryan
    Keymaster

    Thanks! Sometimes the choices can be overwhelming and slow iterations, That is the one place a boss comes in handy, forcing a choice.

    Not completely open source but I have my reasons…

Viewing 30 posts - 1 through 30 (of 40 total)

You must be logged in to reply to this topic.