A new cast iron 3 axis challenger weighing in at over 600lbs! - Page 14
Close
Login to Your Account
Page 14 of 15 FirstFirst ... 412131415 LastLast
Results 261 to 280 of 294
  1. #261
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,495
    Post Thanks / Like
    Likes (Given)
    3090
    Likes (Received)
    672

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    #258 cameraman
    hmm we were also thinking about have 2 stands 1 sheet metal style that has areas that allow you fill it with sand and epxoy. or the cast iron stand which would add about 500 to 600 lbs to machine alone haha. do you think it would be a good idea to release a DIY machine stand with the machine to the people who dont want to buy one ? We have one right now that took my friend about 2 days to weld together and is holding Gen 1 right now.
    Greg Jackson Chief engineer and founder of Tormach, (very early days of Tormach) did not have a proper base for the machine, but had a very elaborate plywood structure... REALLY elaborate piece of engineering in wood. [I bought probably one of the first Tormach's sold in the USA. Mainly as I was enthusiastic about Greg's idea/ concept.]. But one had to build this plywood monstrosity yourself from plans and would take just waaaay too much time. So I bought a rigid table from McMaster Carr (sp) that was rated for about 6000 lbs and was impervious to oil etc. Andf just bolted through that. The machine weighed 1100 lbs. Then Greg started recommending the table I picked to others as a quick fix to tide people over. Later they did ship a more elaborate steel table and tray and all that + pump units etc. Made in China, but comparatively large in space but not that super rigid or anything.


    I've done a fair amount with various large scale filled epoxy systems and I would not recommend that for "Home use" … If you have seen "Peeps" on YouTube trying to build their own granite machines (with some success) it seems like such a PITA and creates such a mess and requires some skills there. Also not 100% sure everyone is that skilled at mixing accelerator with resin and other materials + variations in temperature and humidity from Canada to New Mexico at different times of year in people's garages etc.


    Some Epoxies , the good stuff is really expensive, especially in volume, (that's why I was wondering if for a bench top horizontal) you could ship granite slab from an architectural masonry / stone cutters yard. Basically a thin grave stone / thick kitchen granite slab , doesn't need to be metrology grade.


    Sand ? Yeah why not... Probably most folks could handle quick setting concrete for a base if they had too (accelerated Portland cement ).

    You can sell the machine without a stand / base but you do have to have a disclaimer that you are not responsible for injury or mishap.

    Greg Jackson had the argument that his machine was like a knee mill in rigidity but he just severed the the "Knee" and base so the force loop is much smaller / more rigid he would claim. I.e. saying that all that weight in the base is a waste/ pointless. For me one of the great things about a good knee mill or a bed mill is that it DOES have a decent base even for a light machine (compared to some).

    ______________________________

    Got some hay to hit , Later / Ta.

  2. #262
    Join Date
    Jan 2016
    Country
    UNITED STATES
    State/Province
    New Jersey
    Posts
    191
    Post Thanks / Like
    Likes (Given)
    62
    Likes (Received)
    80

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    #253 Milland
    Yes I been following their progress and it seems the trump taxes hit him very hard and he made some mistakes along the way. We are tying to avoid some of those mistakes by only doing the final assembly and testing on the machines. The machining has been out sourced for kickstater batches to reduce possible hold ups in production. We know it will take about 3 to 4 months to cast ,machine, heat treat,paint and ship all the parts needed for the mill. We would need 2 months to test and assemble the 5 pre production units we plan to make to make sure the machining is done correctly and test our assembly procedure. In addition to that we would need about 2 months to DFM the castings and to design and test the options. so add about 2 months of slack and accounting for things that could happen we are looking at a delivery window of 10 months at worst 9 months at best. We also limited the amount of low cost units to make sure its not suicide to do this.Hmm Ill take note of that when we finalize the mill. They above renders allow us to reuse the base and table ( to reduce cost air freight is expensive)
    I'm on your email list and saw your press release this morning. A few thoughts from someone who has owned a few hobbyist machines and designs machines for a living:

    I think you're backing yourselves into the same corner Swissmak did here, although maybe not as badly. Realize that your supply chain is much, much longer than his since you're using castings, and that if those castings fail you are gonna be screwed if you let them get all the way through machining and delivered to you. Are you planning on being at your machining vendor (who hopefully is also doing the casting) to inspect the castings pre-machining? If not, your worst case window is at least one more round of casting at +4 months, plus all the lost money from bad castings. And am I understanding correctly that you plan to ship these by air? I would seriously reconsider that, for sure. Even a correctly packed, partially filled container will be much, much cheaper (although it does add to your timeline). IMO it seems like, as someone aptly put it in the SwissMak thread, you're 'seeing how everything can go right'. The way to succeed is to ensure that when everything goes wrong, you don't have to close shop.

    From a technical standpoint:

    I just don't think a 5HP spindle is realistic on this machine. Have you done the FEA (or even some hand calcs) to estimate the deflection you see when using all that power? Compare to the "7.5HP" of a Haas mini mill, which weighs four times as much and is still considered a pretty non-rigid machine.

    I would also go back and do some serious cost analysis on the ISO20 spindle. What's the total cost to your customers look like? There is a reason people want Tormach to provide BT30 spindles, not ISO20 ones. I think if you continue on this road you better have bulletproof numbers to show to your customers.

    Other questions I think your customers are going to have:

    Where's the control box going?
    How are you routing wires? Just hanging free?
    Is this running from LinuxCNC? Mach4?
    What's the actual controller you're using?
    What are the axis drive motors? If they're steppers, do you have a plan to offer servos?
    What's driving the anemic max feed and rapid rates? I wouldn't think slightly larger ballscrews and motors to match Tormach's specs would affect price all that much.
    You should probably be specific about compressed air requirements for the power drawbar as well, and maybe add a note about the increase for an ATC down the road so people can plan for it.

    I also wouldn't pull the trigger on that Kickstarter until you've gotten a working design finalized. You may already meet Kickstarter's prototype requirement at this point, I dunno, but in terms of meeting customer expectations and ensuring operating cash flow it's a much safer road to do digital design and as much physical prototyping as possible before taking people's money. I agree with cameraman that the Kickstarter route is not necessarily a good idea from a financial and long-term strategy standpoint. I think it's a very dangerous model that predisposes small businesses to fail for a variety of reasons that I'm happy to discuss at greater length (and I think I did in the SwissMak thread).

  3. Likes cameraman liked this post
  4. #263
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,495
    Post Thanks / Like
    Likes (Given)
    3090
    Likes (Received)
    672

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    hmm maybe something we can for the 5 axis version we plan to make eventually
    The reason why our y travel isnt the greatest is because we plan to stow the tool changer in between the gantry and as a by product of that the table can move to you and the real travel is about 390 mms. but since the tool changer needs about 160mm of y travel to access both rows it kinda cuts the usable y travel. Then again maybe with the side mounted tool changer we wouldnt need to sacrifice 160 mm of y travel. still our footprint right now is about 800mm x 700mmm smaller then tormach 770m and 440 right now
    It wasn't a criticism*, just an observation... I.e. because your Y travel is not that great i.e. as in magnitude it would therefore be easy to tip the whole thing on it's back and create a horizontal mill*.

    __________________________________________________ _____________________________________________


    * Would make most of the issues you are having with a "Base" disappear. [Also bear in mind how easy the current configuration IS to push over onto it's back until you devise something to bolt it down too.]. Even a Tormach at 1100 lbs if unbalanced can fall and crush someone against a wall or floor etc.

    [A larger European horse weighs about 1200 lbs, and a large pony or cobb weighs as much as your machine (660 lbs), so just remember some dad might (one day ) buy this thing for their Kid you don't want it falling over and killing them... (cast iron is not as yielding as a large pony falling on your kid. Again the horizontal arrangement is much safer IMO.]. In the USA kids used to get killed by pulling over giant flat screen TV's and Ikea' book cases and shelving have attachment points to walls for similar reasons.].

    Doesn't the Chinese government pay for shipping / most of it IF this is set up out through a Chinese company (based in mainland China ? ).

  5. #264
    Join Date
    Oct 2014
    Country
    UNITED STATES
    State/Province
    Maine
    Posts
    771
    Post Thanks / Like
    Likes (Given)
    65
    Likes (Received)
    368

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    We are tying to avoid some of those mistakes by only doing the final assembly and testing on the machines. The machining has been out sourced for kickstater batches to reduce possible hold ups in production. We know it will take about 3 to 4 months to cast ,machine, heat treat,paint and ship all the parts needed for the mill. We would need 2 months to test and assemble the 5 pre production units we plan to make to make sure the machining is done correctly and test our assembly procedure. In addition to that we would need about 2 months to DFM the castings and to design and test the options. so add about 2 months of slack and accounting for things that could happen we are looking at a delivery window of 10 months at worst 9 months at best. We also limited the amount of low cost units to make sure its not suicide to do this.
    Your cast/machine/HT/paint/ship timeline is way too aggressive in my opinion, especially since you're trying to make this inexpensive. You have 50 units at 220# each, or 11,000# of cast iron. This will cost a fortune to ship by air, but will be cheap to ship by sea. Allow 6+ weeks for that to the East Coast. Having built a decent amount of machinery, you will find things you would have done differently when you're building the first one. It's really not until the third one that your design is "right." A lot of things are going to be simple things like needing an extra hole somewhere. If your schedule allows, it might not be a bad idea to hold off on doing all your machining until you have built the first one. Adding machining is easy on the computer screen and much more difficult when you have to do rework on 50 units.

    My design comment is that your Z-axis car spacing is much greater in X than it is in Z. With a single screw drive in the center, this doesn't provide much resistance to rotation about Y. Your kinematics are flipped from the normal double column design, which contributes to this. The usual way to fix this is to spread the cars more in Z. Alternatively, one company, I want to say Kern, uses a neat three rail design.

    Lastly, way covers and sheet metal always end up being more of a pain than you think they will.

  6. Likes Vancbiker liked this post
  7. #265
    Join Date
    Sep 2009
    Location
    barcelona, spain
    Posts
    2,246
    Post Thanks / Like
    Likes (Given)
    509
    Likes (Received)
    1369

    Default

    VulcanMachine should do one working machine.

    1.
    Demonstrate working milling, in steel, making a steel round of say 10 mm D in 304 SS square stock of 11 mm square.
    Measure D in x and y, dti error in roundness over a ground v block.

    2.
    Then demonstrate working rigid tap, and or inverse-mode or feed/rev.

    3.
    Then cut 2 parallellograms in SS of choice, flip one, micrometer the error.
    50 mm in size is fine.
    Slow is fine.

    After 1,2,3 they have a machine.

  8. Likes barbter liked this post
  9. #266
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    4,181
    Post Thanks / Like
    Likes (Given)
    778
    Likes (Received)
    2242

    Default

    Quote Originally Posted by DanielG View Post
    ......My design comment is that your Z-axis car spacing is much greater in X than it is in Z. With a single screw drive in the center, this doesn't provide much resistance to rotation about Y. Your kinematics are flipped from the normal double column design, which contributes to this. The usual way to fix this is to spread the cars more in Z. Alternatively, one company, I want to say Kern, uses a neat three rail design......
    My thoughts too. Real machine builders would do an offset third rail or use a dual screw drive.

  10. #267
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Maryland
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    12
    Likes (Received)
    5

    Default

    Quote Originally Posted by cameraman View Post
    Greg Jackson Chief engineer and founder of Tormach, (very early days of Tormach) did not have a proper base for the machine, but had a very elaborate plywood structure... REALLY elaborate piece of engineering in wood. [I bought probably one of the first Tormach's sold in the USA. Mainly as I was enthusiastic about Greg's idea/ concept.]. But one had to build this plywood monstrosity yourself from plans and would take just waaaay too much time. So I bought a rigid table from McMaster Carr (sp) that was rated for about 6000 lbs and was impervious to oil etc. Andf just bolted through that. The machine weighed 1100 lbs. Then Greg started recommending the table I picked to others as a quick fix to tide people over. Later they did ship a more elaborate steel table and tray and all that + pump units etc. Made in China, but comparatively large in space but not that super rigid or anything.


    I've done a fair amount with various large scale filled epoxy systems and I would not recommend that for "Home use" … If you have seen "Peeps" on YouTube trying to build their own granite machines (with some success) it seems like such a PITA and creates such a mess and requires some skills there. Also not 100% sure everyone is that skilled at mixing accelerator with resin and other materials + variations in temperature and humidity from Canada to New Mexico at different times of year in people's garages etc.


    Some Epoxies , the good stuff is really expensive, especially in volume, (that's why I was wondering if for a bench top horizontal) you could ship granite slab from an architectural masonry / stone cutters yard. Basically a thin grave stone / thick kitchen granite slab , doesn't need to be metrology grade.


    Sand ? Yeah why not... Probably most folks could handle quick setting concrete for a base if they had too (accelerated Portland cement ).

    You can sell the machine without a stand / base but you do have to have a disclaimer that you are not responsible for injury or mishap.

    Greg Jackson had the argument that his machine was like a knee mill in rigidity but he just severed the the "Knee" and base so the force loop is much smaller / more rigid he would claim. I.e. saying that all that weight in the base is a waste/ pointless. For me one of the great things about a good knee mill or a bed mill is that it DOES have a decent base even for a light machine (compared to some).

    ______________________________

    Got some hay to hit , Later / Ta.
    gotcha how was the first tormach ? how did they do their first unit sales?
    I see hmmm my buddy has space allocated in the current stand that would allow us to fill it something. I may go fill it up and see if it has any effects.
    Yea we will have to that about the liability as well as purchase liability insurance

  11. #268
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Maryland
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    12
    Likes (Received)
    5

    Default

    Quote Originally Posted by Pariel View Post
    I'm on your email list and saw your press release this morning. A few thoughts from someone who has owned a few hobbyist machines and designs machines for a living:

    I think you're backing yourselves into the same corner Swissmak did here, although maybe not as badly. Realize that your supply chain is much, much longer than his since you're using castings, and that if those castings fail you are gonna be screwed if you let them get all the way through machining and delivered to you. Are you planning on being at your machining vendor (who hopefully is also doing the casting) to inspect the castings pre-machining? If not, your worst case window is at least one more round of casting at +4 months, plus all the lost money from bad castings. And am I understanding correctly that you plan to ship these by air? I would seriously reconsider that, for sure. Even a correctly packed, partially filled container will be much, much cheaper (although it does add to your timeline). IMO it seems like, as someone aptly put it in the SwissMak thread, you're 'seeing how everything can go right'. The way to succeed is to ensure that when everything goes wrong, you don't have to close shop.

    From a technical standpoint:

    I just don't think a 5HP spindle is realistic on this machine. Have you done the FEA (or even some hand calcs) to estimate the deflection you see when using all that power? Compare to the "7.5HP" of a Haas mini mill, which weighs four times as much and is still considered a pretty non-rigid machine.

    I would also go back and do some serious cost analysis on the ISO20 spindle. What's the total cost to your customers look like? There is a reason people want Tormach to provide BT30 spindles, not ISO20 ones. I think if you continue on this road you better have bulletproof numbers to show to your customers.

    Other questions I think your customers are going to have:

    Where's the control box going?
    How are you routing wires? Just hanging free?
    Is this running from LinuxCNC? Mach4?
    What's the actual controller you're using?
    What are the axis drive motors? If they're steppers, do you have a plan to offer servos?
    What's driving the anemic max feed and rapid rates? I wouldn't think slightly larger ballscrews and motors to match Tormach's specs would affect price all that much.
    You should probably be specific about compressed air requirements for the power drawbar as well, and maybe add a note about the increase for an ATC down the road so people can plan for it.

    I also wouldn't pull the trigger on that Kickstarter until you've gotten a working design finalized. You may already meet Kickstarter's prototype requirement at this point, I dunno, but in terms of meeting customer expectations and ensuring operating cash flow it's a much safer road to do digital design and as much physical prototyping as possible before taking people's money. I agree with cameraman that the Kickstarter route is not necessarily a good idea from a financial and long-term strategy standpoint. I think it's a very dangerous model that predisposes small businesses to fail for a variety of reasons that I'm happy to discuss at greater length (and I think I did in the SwissMak thread).
    |
    only the prototype castings are air shipped. We have an agent that does the checking for us. The production units are sea shipped after we confirm the test batch is ok. The agent will check all the casting pre machining as well to ensure that there are no surface defects and such. The foundry we are using are used to making machine tool castings.
    Yes we have My friend did the sims and i think the worst it was at was about 50 to 70 microns of deflection at crashing loads with the z at max height. we assumed this would be the worst amount of deflection since in our design the z is the weakest at the top and the strongest at the bottom.
    As for the 20 taper vs the 30 taper. The cost of a Chinese iso 20 toolholder on aliexpress is about 38 to 40 dollars with 10 dollars of shipping. A bt 30 is about 40 to 50 dollars.

    Control box location will either be bolted to the back or side.
    We are using 2 cable chains to route the cables
    No we running a stand alone controller that is compatible with the standard fanuc gcodes.
    this the controller we are using its missing the estop and a ac power meter.
    20190613_121156.jpg
    Yes we do plan to offer servos.
    The anemic feed rates are driven by the motors not the ballscrews. our thrust at those speeds are
    1357.16N Feed
    904.778N Rapid
    though we may be able to increase rapids to about 200 ipm. we are just waiting for the gen 2 castings to arrive so we can test this.
    hmm as of right now the power draw bar needs 70 psi to run correctly its a double stack design. ( we have ran the machine on a harbor freight 50 dollar air compressor it was not fun but its doable)
    With that in mind yes we do see the dangerous of it but since this will be the 2nd functional prototype i think we are going to be better waters. we already rebuilt the gen 1 prototype about 3 times and have gotten assembly time down to about 8 hours.

  12. #269
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Maryland
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    12
    Likes (Received)
    5

    Default

    Quote Originally Posted by cameraman View Post
    It wasn't a criticism*, just an observation... I.e. because your Y travel is not that great i.e. as in magnitude it would therefore be easy to tip the whole thing on it's back and create a horizontal mill*.

    __________________________________________________ _____________________________________________


    * Would make most of the issues you are having with a "Base" disappear. [Also bear in mind how easy the current configuration IS to push over onto it's back until you devise something to bolt it down too.]. Even a Tormach at 1100 lbs if unbalanced can fall and crush someone against a wall or floor etc.

    [A larger European horse weighs about 1200 lbs, and a large pony or cobb weighs as much as your machine (660 lbs), so just remember some dad might (one day ) buy this thing for their Kid you don't want it falling over and killing them... (cast iron is not as yielding as a large pony falling on your kid. Again the horizontal arrangement is much safer IMO.]. In the USA kids used to get killed by pulling over giant flat screen TV's and Ikea' book cases and shelving have attachment points to walls for similar reasons.].

    Doesn't the Chinese government pay for shipping / most of it IF this is set up out through a Chinese company (based in mainland China ? ).
    hmm ill try to tip the current mill. If a 300 lb ex o lineman cant tip it doubt a little kid could.
    Thou good catch on the tipping issue ill double check the center of mass calculations and see where it is . thank you

  13. #270
    Join Date
    Oct 2014
    Country
    UNITED STATES
    State/Province
    Maine
    Posts
    771
    Post Thanks / Like
    Likes (Given)
    65
    Likes (Received)
    368

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    My friend did the sims and i think the worst it was at was about 50 to 70 microns of deflection at crashing loads with the z at max height. we assumed this would be the worst amount of deflection since in our design the z is the weakest at the top and the strongest at the bottom.
    How did he model the block-rail interface? If you make it a sliding (or no-separation, or whatever your package calls it) contact you will overestimate your stiffness both side-to-side (which is obvious) and in rotation (which is less obvious).

    The best way to do this, in my experience, is to use a general contact that lets you specify stiffness in all 6 directions (X,Y,Z,A,B,C). You set the Z stiffness to 0 (it's able to slide along the rail) and plug in the other five from the catalog stiffness values. Bearing car stiffness is non-linear, however. For rough work, we'd just pick a straight-line approximation. For more accurate work, you run the initial simulation, extract the forces/moments and come up with a straight-line approximation that is more accurate in the region of interest. Most of the manufacturers make it difficult to get the force/deflection and moment/deflection curves. IKO published good curves though, and between-manufacturer variation is minimal, with the exception of IKO, who use a different roller arrangement. Ball rail has a bit more design variants, but you should be able to find good curves.

    Some of the software packages (looking at you SW Simulation) are severely limited in this regard, even though the back end can handle it, the front end can't. We used ANSYS for the analyses.

  14. #271
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Maryland
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    12
    Likes (Received)
    5

    Default

    Quote Originally Posted by DanielG View Post
    Your cast/machine/HT/paint/ship timeline is way too aggressive in my opinion, especially since you're trying to make this inexpensive. You have 50 units at 220# each, or 11,000# of cast iron. This will cost a fortune to ship by air, but will be cheap to ship by sea. Allow 6+ weeks for that to the East Coast. Having built a decent amount of machinery, you will find things you would have done differently when you're building the first one. It's really not until the third one that your design is "right." A lot of things are going to be simple things like needing an extra hole somewhere. If your schedule allows, it might not be a bad idea to hold off on doing all your machining until you have built the first one. Adding machining is easy on the computer screen and much more difficult when you have to do rework on 50 units.

    My design comment is that your Z-axis car spacing is much greater in X than it is in Z. With a single screw drive in the center, this doesn't provide much resistance to rotation about Y. Your kinematics are flipped from the normal double column design, which contributes to this. The usual way to fix this is to spread the cars more in Z. Alternatively, one company, I want to say Kern, uses a neat three rail design.

    Lastly, way covers and sheet metal always end up being more of a pain than you think they will.
    Just to clarify that is to cast and ship the parts over to us the current supplier is asking for a 2 months lead. so i added a month it account for any Chinese holidays that may be in the way.
    break down is
    45 days molds
    1 weeks to cast and heat treat
    1 weeks to machine
    1 week to send it to the port
    3 days for processing time of the payment.
    We may go the long beach route and have it trucked to us from cali.
    Since from my dads shipping records on average the transit time from shanghai to long beach is between 21 to 25 days ( gathered from 2004 to now ) and about 25 to 35 days from shanghai to his Baltimore warehouse.
    We are building the 2nd one and if the kickstarter is success full we cast this version which version 3
    capture1234.jpg
    same gantry better designed table and base allow for easier bolt access on the table.
    hmmm ill look into adding a 3 rail near the ball screw.
    We just added in the waycovers and i hope they arnt that bad.

  15. #272
    Join Date
    Jan 2016
    Country
    UNITED STATES
    State/Province
    New Jersey
    Posts
    191
    Post Thanks / Like
    Likes (Given)
    62
    Likes (Received)
    80

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    |
    Yes we have My friend did the sims and i think the worst it was at was about 50 to 70 microns of deflection at crashing loads with the z at max height. we assumed this would be the worst amount of deflection since in our design the z is the weakest at the top and the strongest at the bottom.
    That may be your worst case, but without analyzing the case at

    As for the 20 taper vs the 30 taper. The cost of a Chinese iso 20 toolholder on aliexpress is about 38 to 40 dollars with 10 dollars of shipping. A bt 30 is about 40 to 50 dollars.
    Maybe all your customers are willing to buy from Aliexpress, but you're in the small business price range at this point, so stuff that's available in the US is probably gonna be a factor. Make of it what you will, but I've spent years on hobby CNC boards and most of those guys only go to Alibaba as a last resort.

    Control box location will either be bolted to the back or side.
    We are using 2 cable chains to route the cables
    I would show those in your marketing materials, and spec out what's in them, if I were trying to drum up interest.

    No we running a stand alone controller that is compatible with the standard fanuc gcodes.
    this the controller we are using its missing the estop and a ac power meter.
    That's impressive. Any chance it won't come covered in Chinese?

    Yes we do plan to offer servos.
    Standard? Either way, my point is, be specific about your product specs.

    The anemic feed rates are driven by the motors not the ballscrews. our thrust at those speeds are
    1357.16N Feed
    904.778N Rapid
    though we may be able to increase rapids to about 200 ipm. we are just waiting for the gen 2 castings to arrive so we can test this.
    Well I'll leave the ballscrew rigidity analysis to you, but you've already got people paying $5-7k for this, what's another couple hundred bucks to get decent acceleration? Especially on the X-axis, which looks pretty light.

    With that in mind yes we do see the dangerous of it but since this will be the 2nd functional prototype i think we are going to be better waters. we already rebuilt the gen 1 prototype about 3 times and have gotten assembly time down to about 8 hours.
    I believe the concern was more about your customers than you. What are you gonna do if someone sues you after losing a finger (or worse)? Product liability insurance is your best bet, they will likely have requirements you have to meet in order to issue you a policy.

    Quote Originally Posted by DanielG View Post
    Some of the software packages (looking at you SW Simulation) are severely limited in this regard, even though the back end can handle it, the front end can't. We used ANSYS for the analyses.
    Yeah, if CATIA or ANSYS is available both will blow SW or anything else really designed for basic 3D CAD out of the water, if they're willing to put in the time (and money).

    That said, they can likely do that analysis by hand. If I were them I would analyze each section that's mounted on a linear bearing separately using FEA, plug those solutions into an analytical spreadsheet using manufacturer's rigidity specs to get the deflection at the rails; repeat as necessary.

  16. #273
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,495
    Post Thanks / Like
    Likes (Given)
    3090
    Likes (Received)
    672

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    gotcha how was the first tormach ? how did they do their first unit sales?
    I see hmmm my buddy has space allocated in the current stand that would allow us to fill it something. I may go fill it up and see if it has any effects.
    Yea we will have to that about the liability as well as purchase liability insurance
    I just want to preface this with that I really loved / believed in Greg Jackson's "Concept", the idea of a 'Personal cnc" he squarely made it 100% analogous to what computers used to be as giant main frames , that no one at the time thought that a "personal" computer would ever happen, as being written off as absurd.


    Greg Jackson's initial idea was also at a time when kids and teenagers became obsessed with computers but actually didn't know anything about electronics i.e. how their computers are made and don't know how to fix stuff; (just before the Arduino guys) and just before the "Maker movement". So I thought Greg's idea was very very good indeed ! [That's principally why I got behind it as I was dismayed that Kids didn't know how to do anything, or build anything or actually know how anything works (as it's ALL micro electronics made by FOXCONN... (made in China/ Taiwan ), To me that was a worrying prospect that needed to be pushed back on.].


    However (being used to smaller extremely precise swiss machines) I absolutely hated the PCNC 1100. I was absolutely appalled at the lack of engineering but that has not changed significantly over a decade. So the MX's right now or soon to be released have real servos, better control and presumably better tolerances. THAT was the kind of machine I was hoping for "Home use" / small laboratory type cnc for special applications part of a small prototype cell. Greg was very persuasive about selling the idea of an open loop system with stepper motors and small force loop (frame of the machine) and slower machine (for prototype work.). … But that all turned out to be complete BS.

    Greg had massive challenges that he explained to me regarding implementing things in China. Back then his major complaint was that they simply did not understand the concept of quality. They would make very bad prototype components and then as samples to score the "Job" and then in a rather casual and flippant way "we can make it better later ...".


    He found training was the biggest issue (out there) and obviously QC. The first machine I had was damaged in transport by the 4th axis unit I ordered hitting it on the head (in a warehouse) from a great height. In spite of that Greg refused to allow me to refuse delivery of the machine... Nothing has changed there lol On a conventional knee mill you can tram the head as it has the axes to adjust, not so much on a Tormach, (I have other equipment so I could re-set stuff. [I'm not a mill wright but have some good 'farm hacks'] ). Basic things like thrust bearings not installed correctly or VFD's fritzing out etc. that kind of thing... Stuff that still seems to happen today and the basic manifest lack of quality and precision. [At that time Greg only had a really bad CAM program to hand out (Trubo CAD) and then tried to hook up with Russian developers SprutCam, but that had problems + sanctions. Greg basically called out the major CAD/CAM companies no better than drug/ crack dealers with crazy expensive predatory licensing schemes.

    The problem was and kinda is , they could have made an MX type machine ten years ago, but because of how they went about things they kinda had to live with a pretty substantial design freeze that they are just NOW climbing out of. So IMO always best to make a product as best as you possibly can so it's design is GOOD for the next ten to 15 years from now BEFORE you launch.


    Good engineering and good design does not cost more than bad engineering, poor quality and bad design.



    __________________________________________________ _


    The Tormach tooling system was quite popular before the first PCNC came out, based on the R8 collet system. That was a clever pre-cursor rollout.




    This is NOT me / my garage lol ^^^ but illustrates how 'Peeps" were wanting to get into the Tormach Tooling system.




    "Theory" ^^^




    ^^^ "Practice"

    __________________________________________________ __________________

    A lot of good things have been made on a Tormach and people have been able to do things they otherwise would never have been able to do (including Saunders Machine Works / NYC CNC and many others.).

    Being half German I have a bug / Broom up my butt / arse about quality and good engineering.

    Horses for courses.

    __________________________________________________ ____________________


    Tormach did not start from a kickstarter campaign, to Greg Jackson's immense credit he took some serious risks, had a lot of balls to go out to China to implement what he did (back then and in the way he did) and his vision/ concept was and is (still) superb ! Sad that he passed away when he did but as I understand the folks that work for Tormach are now part owners or own 100% .

  17. Likes metalmadness liked this post
  18. #274
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,495
    Post Thanks / Like
    Likes (Given)
    3090
    Likes (Received)
    672

    Default

    Quote Originally Posted by VulcanMachineCo View Post
    hmm ill try to tip the current mill. If a 300 lb ex o lineman cant tip it doubt a little kid could.
    Thou good catch on the tipping issue ill double check the center of mass calculations and see where it is . thank you
    Not so much that , it's home brew attempts at rigging where people in general get killed. Even lathes get flipped.

    Also if such a tall / heavy machine is on a wobbly stand there can be a resonance for slow moves back and fourth where something could tip/ become unstable.


    If people can buy the machine without a stand then there will be the possibility that they won't bolt your machine down to something. Bad footing + top heavy machine ?

  19. #275
    Join Date
    Jan 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    672
    Post Thanks / Like
    Likes (Given)
    102
    Likes (Received)
    351

    Default

    I'd imagine you'd get quite a bit of twist on the Y-axis trying to do anything off center of the table. Could extend the lower trucks on y-gantry further downward to combat twist. (Or just use a screw at each end instead of one in the center)

    You've got extra linear rail there you can't use anyway. How close to the table are you gonna work? Add a vise and cutting tools and it'll be even more wasted rail.



    Why does it seem like hobby guys build their machines with 80 feet of Z-travel, only to do flat parts, or tiny shit in a vise?

  20. #276
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,495
    Post Thanks / Like
    Likes (Given)
    3090
    Likes (Received)
    672

    Default

    Quote Originally Posted by thesidetalker View Post
    I'd imagine you'd get quite a bit of twist on the Y-axis trying to do anything off center of the table. Could extend the lower trucks on y-gantry further downward to combat twist. (Or just use a screw at each end instead of one in the center)

    You've got extra linear rail there you can't use anyway. How close to the table are you gonna work? Add a vise and cutting tools and it'll be even more wasted rail.



    Why does it seem like hobby guys build their machines with 80 feet of Z-travel, only to do flat parts, or tiny shit in a vise?

    I think you hit the nail on the head there.


    If you take a Makino F5 (bridge style machine Vulcan-peeps aim to emulate ) and build a scale working model 1/2 scale so it has linear travels of half of the original...


    Original (travels) - : X 35.5", Y 19.7" , Z 17.7 " (weight 16,500 lbs)

    1/2 scale (travels) : X 17.75" , Y 9.85, Z 8.85 " (Weight 2,062 lbs )


    The weight and volume scale by a cubic relationship* (1/2 the linear dimensions equates to 1/8th of the volume.)


    1/3rd scale (machine) : X 11.8", Y 9.85", Z 5.9 " (weight 611 lbs 1/27th of the volume.). ~ Same weight as Vulcan machine.


    So indeed Vulcan has scaled the Z by roughly a factor 2 (relative to their 1/3 rd linear scaling of a real machine) i.e. Z is twice as long as it would be if correctly scaled. So it's not really a scaled down bridge mill unless an Okuma M-560V and Makino F5 were 18 feet tall and had really thin walled castings (to preserve original mass) .


    So @thesidetalker if Vulcan wanted to build to scale then the Z travel would have to be about 6".


    Not sure how a machine that is 1/27th the volume scales for cutting power given that the spindle they have to incorporate is much more massive by comparison.

    personally cast iron could be a gimmick in this case and emulating a bridge style that got shrunk in the wash doesn't work due to spindle size and geometry (as well as their need for increased Z). To make a better design sometimes one has to let go of certain seemingly cherished ideas, or things that are marketing related rather than functional ...

    __________________________________________________ _____________________________________________

    * Always been mystified by how and why a 4000 lb Robodrill is able to do what it is able to do, but I guess that cubic relationship makes it all "Right"/ goes lovely.

  21. #277
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Maryland
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    12
    Likes (Received)
    5

    Default

    Hi thesidetalker, our current cad model gives about 20mm of clearance between the shortest tool holder and the table at minimum Z travel. We bumped the z travel over the original prototype because once we put a vise on it we lost too much of the available travel and once we put a drill in there plus a vise, our effective boundary box for cutting shrank to just a few inches. We increased the Z over the previous iteration to allow more room for longer tools, to fit the tool changer, and to be more in line with what other smaller sized machines use as a minimum.

    It's only a 13" or so of travel, and that's barely over a foot. I don't think it's excessive at all, it seems more like the the minimum we could build this with. We considered making the nose to table distance greater, more in line with what you'd see if a vise was on the table, and for anything less using extended tool holders, but we thought that may of just been complicating things. What is your opinion?



    table-nose-clearance.jpg

  22. #278
    Join Date
    Jan 2016
    Country
    UNITED STATES
    State/Province
    New Jersey
    Posts
    191
    Post Thanks / Like
    Likes (Given)
    62
    Likes (Received)
    80

    Default

    Quote Originally Posted by cameraman View Post
    * Always been mystified by how and why a 4000 lb Robodrill is able to do what it is able to do, but I guess that cubic relationship makes it all "Right"/ goes lovely.
    You're forgetting that reducing the travels also reduces the moments, which reduces the deflection that drives total mass of the machine.

    Quote Originally Posted by VulcanMachineCo View Post
    What is your opinion?
    This should never be about opinion, IMO. I mean you're the one selling machine, so you make the call, but when I sell a customer a machine, I've already got a spec for things like deflection under load, and I design the machine to the spec. That includes going over the range of potential motion to determine edge cases that usually drive the design limits.

  23. #279
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Maryland
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    12
    Likes (Received)
    5

    Default

    Cameraman, while your math is mostly correct (in that, in broad strokes, you can scale it like that), I disagree with your assumptions and conclusions. First our original prototype was a bridge style, so at that time if we were emulating a certain mill, I suppose you could say that included the Makino F5. The current one is different and if I had to guess what I might hear people say we were emulating, it'd probably be the Prototrak Op2 in that it has a similar configuration with the way the Z and X works.

    Next, looking at the spec sheet for the F3 and F5 I'd like to argue that our machine and the Makino are two completely different leagues and not just in size.

    Attachment 258922

    Some things I'd like to point out in their website and documentation is how much they stress the high precision of their machines. They use scale encoders with 0.05 micron resolution and feed rates (not just rapids but actually advertised as feed rates) of almost 800 ipm. A positioning accuracy of 0.00006" and a repeatability of 0.00004". There is no doubt in anyones mind that the stiffness and robustness of a frame on a machine like that needs to be through the roof. Our machine will not be a scaled down Makino F5, just the cost in parts and manufacturing to make a scaled version of that wouldn't probably be anything less than 20k being really optimistic.

    Also note that the Makino F3, which has about 10" less travel on the X than the F5, has relatively the same weight, and that their weight estimates are for the entire package, the frame, the motors, the linears, the stand/base that goes all the way to the floor, the electronics, the enclosure, the tool changer etc., while ours are are currently including the frame and estimating the motors, and estimating the electronics.

    For reference, using Tormachs 770m (since it's a machine in which I could find both weights), they're advertising around 350kg for the mill by itself (shipping weight) and about 650kg for the whole package including stand/enclosure/atc. Syil X5 goes from 300kg for the mill by itself to 800kg for base + enclosure + cast iron stand.

    EDIT: the picture I posted is small. I screen shot it from this link, page 6.
    http://www.bener.com.br/img/produtos...b497bde0a7.pdf

  24. #280
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Maryland
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    12
    Likes (Received)
    5

    Default

    Quote Originally Posted by Pariel View Post
    This should never be about opinion, IMO. I mean you're the one selling machine, so you make the call, but when I sell a customer a machine, I've already got a spec for things like deflection under load, and I design the machine to the spec. That includes going over the range of potential motion to determine edge cases that usually drive the design limits.
    I wasn't asking about what the deflection should be lol, I was referring to a functionality/feature opinion in that needing to use extended tool holders to reach anything below a the 2-4" height bump a vise adds to increase more top end z travel on a small to medium sized cnc. Using extended tool holders increases any deflection and runout issues which is why we thought to avoid it for standard usage in the design. A lot of the larger machines will have spindle nose to table distances of over 4" at min z. Ours is currently a hair above 2" but it was pointed out that it is too low?


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •