OpenBuilds Style Build

[WoodKnot]

OK, I think a hugely embarrassing start, everything from the outset seems to be wrong.

Maybe I had made far too many assumptions and that is my fault!

I assumed that the ball screws would be slightly longer than the rails

I assumed that the ballscrew fittings would sit within the C Beam?

It all looks wrong! I don't think I am being a numpty but mine looks nothing like the others, even before I start.

Having parted with the best part of £800.00 just to get started, it's not a good start.

Will post some images in the next post as they are on my phone

WoodKnot

[A_Camera]

OK, I think a hugely embarrassing start, everything from the outset seems to be wrong.

Maybe I had made far too many assumptions and that is my fault!

I assumed that the ball screws would be slightly longer than the rails

I assumed that the ballscrew fittings would sit within the C Beam?

It all looks wrong! I don't think I am being a numpty but mine looks nothing like the others, even before I start.

Having parted with the best part of £800.00 just to get started, it's not a good start.

Will post some images in the next post as they are on my phone

WoodKnot

This is really bad news, I am sorry. I thought you bought 45x90 extrusions, which was discussed elsewhere as far as I can remember. The BF/BK12 fits nicely on those, using 6mm T-nuts, even though the most optimal would be 46mm center distance but the 45mm is fine. Anyway, I missed somewhere in the debate that you did in the end go for something else than 45x90. Hopefully you can use the ones you have now to something else and can buy 45x90 for your project. I normally buy my extrusions from Germany, a company named DOLD Mechatronic. They have good prices and good quality.

Too bad you wasted some money but I guess the lesson learned is to look at technical specs and drawings. While I never spend time on designing a machine in a fancy 3D modelling software, I do spend a lot of time on looking at drawings and specifications before I order something to avoid mistakes like this. Anyway, this is the reality of DIY world. We make mistakes and we learn and hopefully never repeat the same mistakes. Mistakes are made by the professionals also even in the industry, but their DIY is called prototype. The difference is that the prototype never gets sold, so we only see the nice final product, believing that it always looked like that.

Yes, normally the screws are longer than the rails, but with some smart design you can still have the same movement length as the rails allow, assuming the differences in length are not that large.

Regarding the alignment, I assumed you meant parallel alignment between rails and rails, and between rails and screws, plus squaring. As Jazz says, that's what you should spend time on, not aligning the rails with the extrusions. Of course, it helps if the extrusions are also aligned and squared to each other, but the main thing is that all the rails and screws run aligned and are squared to each other.

[WoodKnot]

This is really bad news, I am sorry. I thought you bought 45x90 extrusions, which was discussed elsewhere as far as I can remember. The BF/BK12 fits nicely on those, using 6mm T-nuts, even though the most optimal would be 46mm center distance but the 45mm is fine. Anyway, I missed somewhere in the debate that you did in the end go for something else than 45x90. Hopefully you can use the ones you have now to something else and can buy 45x90 for your project. I normally buy my extrusions from Germany, a company named DOLD Mechatronic. They have good prices and good quality.

Too bad you wasted some money but I guess the lesson learned is to look at technical specs and drawings. While I never spend time on designing a machine in a fancy 3D modelling software, I do spend a lot of time on looking at drawings and specifications before I order something to avoid mistakes like this. Anyway, this is the reality of DIY world. We make mistakes and we learn and hopefully never repeat the same mistakes. Mistakes are made by the professionals also even in the industry, but their DIY is called prototype. The difference is that the prototype never gets sold, so we only see the nice final product, believing that it always looked like that.

Yes, normally the screws are longer than the rails, but with some smart design you can still have the same movement length as the rails allow, assuming the differences in length are not that large.

Regarding the alignment, I assumed you meant parallel alignment between rails and rails, and between rails and screws, plus squaring. As Jazz says, that's what you should spend time on, not aligning the rails with the extrusions. Of course, it helps if the extrusions are also aligned and squared to each other, but the main thing is that all the rails and screws run aligned and are squared to each other.

Thanks A Camera -

Now that i have had chance to get myself orientated this morning - i have checked what everyone else's has used and they are all running 4080!

If I go 45x90, that means all my support extrusions will not fit either!

Most are using much thinner ball screws and also running 12 or 15 rails, where i opted to go 20 to beef it up even more.

I have been looking at the 'carriage' (not sure if that is the correct terminology) that they use on the ball screw to connect to the side plates, this just seems to be a flat plate, nothing like the parts that have come with my ball screws

I need to take some time this evening and re-evaluate the situation.

Thanks

WoodKnot

[AndyUK]

I have been looking at the 'carriage' (not sure if that is the correct terminology) that they use on the ball screw to connect to the side plates, this just seems to be a flat plate, nothing like the parts that have come with my ball screws

Eurgh, the months I spent trying to remember the right terminologies. The moving piece on the ballscrew (a container for a series of ball bearings) is known as the nut. The moving bits on the linear rails are carriages.

[WoodKnot]

Eurgh, the months I spent trying to remember the right terminologies. The moving piece on the ballscrew (a container for a series of ball bearings) is known as the nut. The moving bits on the linear rails are carriages.

Thanks Andy -

Yes, just figured that out!!!!

So................... something has to give!

The least cost option is just to replace the ball screws by the look of it?

I am now trying to figure out which size of thread and flange nut will fit in the 40mm channel.

I intend to keep the 4080 profile and the 20mm linear rails.

I must be able to find something to run in that channel, as everyone else seems to as it is the most common C Beam profile

WoodKnot

[A_Camera]

Thanks A Camera -

Now that i have had chance to get myself orientated this morning - i have checked what everyone else's has used and they are all running 4080!

If I go 45x90, that means all my support extrusions will not fit either!

Most are using much thinner ball screws and also running 12 or 15 rails, where i opted to go 20 to beef it up even more.

I have been looking at the 'carriage' (not sure if that is the correct terminology) that they use on the ball screw to connect to the side plates, this just seems to be a flat plate, nothing like the parts that have come with my ball screws

I need to take some time this evening and re-evaluate the situation.

Thanks

WoodKnot

Are all your profiles 40x80? I tried to catch up a bit and went back in history to try to figure out what you actually bought for parts, but to be honest, I failed. I don't know which parts you bought, or if you just bought a drawing or a complete kit with some modifications of your own. Somewhere else you posted a link to this machine from cnc4you:



Is this similar to yours? Now, I have no idea about the dimensions here, but it looks like mostly 4x120. As you can see, the BK/BF of the X screw is placed over one groove, the picture shows clearly that the BK/BF is fixed through drilled holes in the extrusion between the grooves. That is also possible to do as a solution, but if you have only 40x80 all the way then you have bigger issues. Anyway, it is not clear to me at this stage what you actually bought, how the parts are meant to be installed or what the actual problem is, other than the fact that the BF/BF won't fit into the grooves. In any case, if 45x90 is not working, perhaps 40x120 would work. It adds some costs, but surely, not nearly as much as £800 because I am pretty sure you could save and use most of the stuff, depending on the design. Never the less, it would be a good idea if you posted some pictures, it would help a lot in the discussions. Anyway, if you buy new extrusions I can recommend you buy it from DOLD Mechatronic. They have good prices and fast delivery. Some extrusions won't cost you an arm and a leg, and you can easily cut them in your workshop.

I think that the 20 rails are MUCH better than 12, which in my opinion is only suitable for 3D printers, so in my opinion that's a good choice. I also think that using a smaller than 1605 screw is not a very good thing. 1204 may work, but not on such large machine. I am using 1204 on one of my 3D printers. 8 mm ACME is a joke for this sort of thing, it is suitable for 3D printers, but not a large CNC.

If I were you I'd try to make something out of the parts you have, even if in the end it will take longer time. What's been said several times and in several forms is that you must make as a minimum, some sort of plan on how you want the machine to look like. If you do that in a CAD software, that's fine, but it is my understanding that you don't want to do that, so the absolute minimum is to take a piece of paper and a pencil and start sketching, measuring and making drawings. I think Jazz misunderstood my comments, because it sounds like he thinks I am against using software, which I am not. In fact, I do use several, but not for machine design and machine assembly simulation. Even though I am not using any machine modelling software, the mistakes you made would never happen to me, not because of CAD or no CAD, but because I do read specifications, look at technical and mechanical drawings, check dimensions, make sketches and measurements and so on. I think that it is ABSOLUTELY necessary that you do that as well, otherwise your build will be extremely expensive and may end up in the garbage if you just ignore things and assume that it will work out somehow. I don't know what sort of things you have made before, but you seem to have a very large and luxurious workshop, especially compared to mine, so surely, even if you only worked with wood before, you must have made many drawings for whatever you made. I am not a wood worker, and my last wood project was two very simple birdhouses, but I made drawings even for those before I bought the material and started cutting up. So yes, drawings are VERY important according to me, but machines were made long before CAD software was invented, so that is not necessary at all. The type of table CNC we normally build is actually a pretty simple machine compared many others which can be made without and CAD software, so that is not required, but careful measurements and drawings are definitely necessary.

[WoodKnot]

Are all your profiles 40x80? I tried to catch up a bit and went back in history to try to figure out what you actually bought for parts, but to be honest, I failed. I don't know which parts you bought, or if you just bought a drawing or a complete kit with some modifications of your own. Somewhere else you posted a link to this machine from cnc4you:



Is this similar to yours? Now, I have no idea about the dimensions here, but it looks like mostly 4x120. As you can see, the BK/BF of the X screw is placed over one groove, the picture shows clearly that the BK/BF is fixed through drilled holes in the extrusion between the grooves. That is also possible to do as a solution, but if you have only 40x80 all the way then you have bigger issues. Anyway, it is not clear to me at this stage what you actually bought, how the parts are meant to be installed or what the actual problem is, other than the fact that the BF/BF won't fit into the grooves. In any case, if 45x90 is not working, perhaps 40x120 would work. It adds some costs, but surely, not nearly as much as £800 because I am pretty sure you could save and use most of the stuff, depending on the design. Never the less, it would be a good idea if you posted some pictures, it would help a lot in the discussions. Anyway, if you buy new extrusions I can recommend you buy it from DOLD Mechatronic. They have good prices and fast delivery. Some extrusions won't cost you an arm and a leg, and you can easily cut them in your workshop.

I think that the 20 rails are MUCH better than 12, which in my opinion is only suitable for 3D printers, so in my opinion that's a good choice. I also think that using a smaller than 1605 screw is not a very good thing. 1204 may work, but not on such large machine. I am using 1204 on one of my 3D printers. 8 mm ACME is a joke for this sort of thing, it is suitable for 3D printers, but not a large CNC.

If I were you I'd try to make something out of the parts you have, even if in the end it will take longer time. What's been said several times and in several forms is that you must make as a minimum, some sort of plan on how you want the machine to look like. If you do that in a CAD software, that's fine, but it is my understanding that you don't want to do that, so the absolute minimum is to take a piece of paper and a pencil and start sketching, measuring and making drawings. I think Jazz misunderstood my comments, because it sounds like he thinks I am against using software, which I am not. In fact, I do use several, but not for machine design and machine assembly simulation. Even though I am not using any machine modelling software, the mistakes you made would never happen to me, not because of CAD or no CAD, but because I do read specifications, look at technical and mechanical drawings, check dimensions, make sketches and measurements and so on. I think that it is ABSOLUTELY necessary that you do that as well, otherwise your build will be extremely expensive and may end up in the garbage if you just ignore things and assume that it will work out somehow. I don't know what sort of things you have made before, but you seem to have a very large and luxurious workshop, especially compared to mine, so surely, even if you only worked with wood before, you must have made many drawings for whatever you made. I am not a wood worker, and my last wood project was two very simple birdhouses, but I made drawings even for those before I bought the material and started cutting up. So yes, drawings are VERY important according to me, but machines were made long before CAD software was invented, so that is not necessary at all. The type of table CNC we normally build is actually a pretty simple machine compared many others which can be made without and CAD software, so that is not required, but careful measurements and drawings are definitely necessary.

Thanks A Camera

The profiles that virtually everyone on OpenBuilds is using is this - https://ooznest.co.uk/product/c-beam...l-cut-to-size/

A lot of guys in the US purchase from them in the UK.

Never used a drawing for anything i have made.

I measure the dimensions of what i want to make and where it is going to go and then just get on with making it.

So, i need to make a bookcase for an alcove -

Height, width, depth, height of books from large to small, thickness of material being used, allow for rebates etc - all just written down as physical sizes. No drawings

I look at other designs, profile edges, create radius corners, all based on what i think looks aesthetically pleasing!

I will make use of most of what i have - but the ball screws are to short and will cause an issue as they are the exact length of all my profiles.

Were as the others are oversized generally by 40mm or greater to allow them to pass through the plates and be secured.

WoodKnot

[A_Camera]

Thanks A Camera

The profiles that virtually everyone on OpenBuilds is using is this - https://ooznest.co.uk/product/c-beam...l-cut-to-size/

A lot of guys in the US purchase from them in the UK.

Never used a drawing for anything i have made.

I measure the dimensions of what i want to make and where it is going to go and then just get on with making it.

So, i need to make a bookcase for an alcove -

Height, width, depth, height of books from large to small, thickness of material being used, allow for rebates etc - all just written down as physical sizes. No drawings

I look at other designs, profile edges, create radius corners, all based on what i think looks aesthetically pleasing!

I will make use of most of what i have - but the ball screws are to short and will cause an issue as they are the exact length of all my profiles.

Were as the others are oversized generally by 40mm or greater to allow them to pass through the plates and be secured.

WoodKnot

Oh, I see. Well, those extrusions are in my opinion far too weak. I was hoping that you at least use 40x80, but those you went for, I see as 20x20 C shaped. Maybe suitable for some very small, light machine with a small DC motor as spindle, but not for a machine of the size and type of use you are aiming at, even if you'd only work with soft wood, I think that would cause you more pain than joy when using it. I think at this stage I'd byte the bullet and start over, trying to reuse the strongest parts (rails and ball screws), not the weakest ones (extrusions), but in any case, I think it's high time to start sketching on paper or in a computer.

Also regarding CAD, I think you must give up any idea about not wanting to use one. I understand that you don't want to spend a small fortune on a CAD software, but there are very capable and useful software out there even for free. FreeCAD is what I am using and as the name says, it's free of charge, unless you are like me and want to donate money to those developers. There are other free CAD and CAD related software out there, and if you are aiming at ever using your CNC then you simply must learn some CAD software, without that, you will have really no use for a CNC. Even a simple freeware like F-engrave needs some understanding and configuration, as well as the CNC software (whichever you plan to use) needs some serious configuration, input file with some G-code containing the correct setup for the material and the cutters, spindle motor and many, many things. It all sounds simple and looks as a simple work, but it isn't. You can, if you are good at low level programming (like I am), create G-code without CAD/CAM, or any other software, using only Windows Notepad, but it takes time and requires a good understanding of G-code, but it is not practical and it takes time, especially if you are not a professional programmer or have many years of experience and knowledge. I started with manual G-code programming and actually created fairly complex items because I was in a hurry and for me it was easier and better this way, while I was trying to find my choice of CAD which I'd be happy with, but I have over 40 years of experience in programming, of which 15 in assembly and C, and the rest is C++, Ada, Ada+ and so on, so with my background it was the right choice, but if I did not have that background, I would have never started with manually programming in G-code. So, while I still stand for what I said, machines can be designed without CAD software, I strongly recommend you to seriously consider some CAD software because you will need it.

Here is a video about FreeCAD and Blender, the guy uses FreeCAD (just like I do) for designing parts and he assembles those parts using Blender, which is also free. I think that this is a very good idea, so I actually installed Blender and will have a look at it for myself.



I am also not interested in spending a fortune on CAD/CAM and modelling software, but of course, those who work professionally and actually selling machines, these are necessary tools, and commercial ones are expensive for us mortals, so for me the free alternatives are suitable. As I said, FreeCAD is free, but donating is fine, you can actually donate any sum and to be honest, it is a professional software, so those guys and dolls deserve my money every time I update the software. I just installed Blender and that is also freeware (donate ware) and a quick test shows that I like it, but will test more and if I ever use it for real I will donate them as well, just like I donate other software makers. But it is not compulsory, and those software don't have any limitations if you don't donate, so that's good. Many commercial alternatives have free versions, but every time I tried one, I run into limitations, so I gave up on them because even they claim that you get the full version, they are ALWAYS cripple ware, or have some serious limitations in usage time, saving possibilities, maximum sizes or other things which can be frustrating or make life difficult.

In short, you should start consider some sort of CAD and accept the steep uphill it may cause before you get started for real with one.

[JAZZCNC]

Would love to see the comments from Jazz when he see's these!! What a pile of ........................

I've seen many machines which use Delrin nuts and ACME screws that work perfectly fine in for the right applications, however not very good for CNC for obvious reasons plus they wear out very quickly.

Too bad you wasted some money but I guess the lesson learned is to look at technical specs and drawings. While I never spend time on designing a machine in a fancy 3D modelling software, I do spend a lot of time on looking at drawings and specifications before I order something to avoid mistakes like this. Anyway, this is the reality of DIY world. We make mistakes and we learn and hopefully never repeat the same mistakes. Mistakes are made by the professionals also even in the industry, but their DIY is called prototype. The difference is that the prototype never gets sold, so we only see the nice final product, believing that it always looked like that.

No sorry, the mistake was not listening to the advice given and making very large assumptions without a clue to what was being bought.!

If Woodknot had listened to the advice I gave and done the research on the design and components used, then made at least a 2D drawing to help with dimensions and fit then this wouldn't have happened.
Also, if he had started a Build thread sooner and posted the components in this thread rather than buying first and starting threads all over the Forum there was a much higher chance one of us more experienced builders would have caught this before it happened.

I don't agree with you regards not designing using 3D models.!
All my machines are 100% designed with 3D models, even the ones which never get built and I rarely make mistakes that actually make it to the actual machine as they are all caught in the CAD model. It actually saves wasted money on R&D as a large chunk of the problems are caught on the Model so parts are not being scrapped, (the Camming of parts is another story and I make mistakes regularly) 3D models allow me to see everything and how it fits together so this doesn't happen and the Finished Model is an exact replica of the real finished machine. So much so that all the machined parts are taken directly from the model into CAM software.
Also, if design changes are needed for any reason, it's often a simple tweak and the effects of the change are reflected and obvious instantly.

3D Modeling, even a basic model, can and DOES save you time and money in the long run even for a one-off machine.! This case is a classic example.?

If WoodKnot had modeled just a basic model of the machine using off the self Models for Profile and Bk bearings etc which are provided by suppliers then it would have become obvious they wouldn't fit and this expensive mistake would have been avoided.
Also, it shows and removes the chance of any hidden surprises further down the road which at the moment he will only find out about at the time he comes to this part of the build, which is often too late and again becomes expensive.!

Also, if the machine was modeled any plates and parts which he required machining by outside sources he would have models to give the engineer and reduce costs.

Lastly, at some point in CNC then sooner or later you are going to want to use a 3D model so taking the time to learn 3D modeling while designing the machine will pay off for using the machine and creating G-code quicker.