My CNC Router Build Adventure

[JayneV]

Hi everyone,

It’s been a long while since my last post. Was dealing with some health issues and also super busy with work.

Yesterday I placed an order for some aluminium to finally make a start on the CNC. I was told it may take up to two weeks for delivery due to supply issues and flooding this past week. It’s taken me this long to get started, another couple of weeks won’t hurt.

We recently replaced the solar panels on our roof and I decided to keep all of the mounting hardware from the old panels which included a bunch of extrusions (see attached image). I couldn’t bear to see those all go to the garbage. I’m wondering if these extrusions would be suitable to use in the base of the machine as T-Slots for securing work pieces if they are laid flat at intervals with strips of MDF in between? The top of the MDF would sit higher than the extrusion so it can be surfaced. Or make the base entirely out of these extrusions instead of MDF and use a sacrificial piece of wood under work pieces as required. Any thoughts?

[peteeng]

Hey Jayne - Great your back in the game. Peter

[JayneV]

Thanks Pete.

Made a start on the bench today, I cut the legs to size and retrieved some timber for the bench frame out of the roof that has been sitting in there since the house was built 17 years ago.

[awerby]

Sure, if you stacked those aluminum extrusions up, fastening them down to something flat, you could make a table you could attach a sacrificial spoilboard to with tee-nuts. If you use plywood or MDF, various sorts of fasteners can be used to secure workpieces down to for CNC operations. But it wouldn't have enough stiffness for the structural parts of a CNC router. Steel box beam has the best stiffness/cost ratio, if you're building from scratch.

[JayneV]

Hi Andrew, the machine structure is pretty much finalised, mostly made from various sections cut from aluminium flat bar and plate and some C-channel. The above extrusions would only be for the work area of the machine, originally planned to be a sheet of plywood with a sacrificial layer of MDF on top of that. If I go down the path of using the extrusions, they would take the place of the plywood. I'll post a CAD screenshot when I update model.

[JayneV]

I quickly modelled the T-Slots to visualise how it will fit. The first image is with a plywood baseboard (MDF sacrificial board not shown) and the second image shows the T-slots. There is also a length of the extrusion vertically on the front apron for securing larger workpieces to the apron to machine them on edge. The apron mounted extrusion will need some minor adjustments to the front of the machine so that it sits flush with the front face. Can anyone see any downside to using the t-slots instead of a plywood board?

[peteeng]

Hi Jayne - The t slot is discontinuous so can't generate membrane stiffness when it bends across the machine as the "planks" act independently. The plywood being continuous is potentially stiffer. But then that's the job of the beams underneath to provide the transverse stiffness. There are a few machines that use planks this way and they seem to work. Your design is looking solid either way. I like ply and an insert grid vs planks... Peter

[JayneV]

Hi Pete,

I was having similar thoughts regarding the stiffness of the t-slots. As you mentioned, the underlying structure will be providing the overall machine stiffness. The thought of using the t-slots originated from looking for a place to use them rather than picking the best material for the job, hence my question. I’ll give it some more thought, perhaps I’ll stick with plan A and use plywood, rather than cut up and use the t-slots and find anotger place to use for planks.

[Texasbowhunter]

Jayne I have recently joined and want to say you have done an amazing job designing this unit...Pete has been very informative with me on the design I had envisioned as well...
Unfortunately My design is not to this stage as yours...Mine is in my head and seems to be evolving with more and more research and taking a design all to its own...LOL
Jayne I'm former manufacturing here in the US on some of the most sophisticated Air Craft out there and what I'm seeing is on your supports for the frame of your design You will be fabricating a bunch of these several times...let me suggest, if you have not planned for this, is build a simple jig that all your supports and angles can locate hard against so you can drill and tap the various components to make each one individually...Hopefully from start to finish...after finished you will have one rib (blue in drawing) that is what you had envisioned and you will have the self satisfaction of building many more so they will all be the same or at least within a few thousandths of each other...hard located only at the points that are critical in maintaining...Base, Top, and Face
I'll see if I can find something on the web that explains by a visual means to help understand...
Looking like a great build
Paul

[awerby]

If you fasten all that extrusion to a base made of plywood and cap it with a MDF or plywood spoilboard, it will be stiffer than plywood alone, especially in one direction.

[JayneV]

Hi Paul, welcome to the forum. Thank you for your kind compliments. Much of the credit for the design can be traced back to many of the helpful members here on the forum, particularly Pete. Thank you again Pete

I do understand what you mean about using a jig to hard locate parts to make multiple copies of a part. My brother has a small machine shop with a CNC mill which I was intending to use to make all the parts. I recently decided to only make the more critical and complex parts with his machines and the simpler parts I’ll make at home as a challenge to myself.

Hope to see your design evolve soon, do you have a build/design thread?

Andrew, a sandwich construction the way you describe would eat into the available Z travel which I would prefer not to do. I’ll stick with plan A for now using plywood and MDF and if there problems I will try adding the t-slots, that way I won’t unnecessarily be cutting up the aluminium and potentially wasting it.

[Texasbowhunter]

Jayne...On your design the areas where there doesn't seem to me an exterior skin...Is that going to be open?
If so what would look KOOL would be to house the electronics in there and sheet it with clear plastic of some type...
Very Kool looking effect under low light conditions...
Speaking of lights have you considered where or if your going to add lights...seems that folks are trying to figure out how and where to add additional light for there projects that are being cut...Under Gantry? LED perhaps...
Paul

[peteeng]

Hi Jayne - If you intend to cut a lot of aluminium then an all metal bed is good as it won't be bothered by lubricant. If you have a ply bed then you'll need to resin coat it really well otherwise lub will ruin it quickly... Peter

[JayneV]

Didn’t think about the lubricant, thanks Pete. I don’t “intend” to cut a whole lot of aluminium but I’m sure if the machine is capable then I undoubtedly will cut more than originally anticipated. So many things to consider and compromises to be made when attempting to design and build a machine. This is fun!

[JayneV]

Could someone offer assistance with some terminology please? There are several parts of that I need to have either laser or waterjet cut from various thicknesses of aluminium plate ranging from 12-25mm. To help keep costs down I will ask them to only cut the outline of each part and to "centre punch" the location of all the holes. Final machining will be done on a CNC mill. I have previously spoken with a few local companies that can do the cutting and all of them said it would be cheaper if they didn't cut all the holes. However they did say they can "centre punch" the location of all the holes with the waterjet for little extra cost which would be a great help. I don't remember the exact term they used for this "centre punching" process. I am drafting an email to send out for quote today, could someone help me out with the correct terminology for the centre punching process by a waterjet or laser cutter? It is Sunday today and all the cutting places are closed to call and ask them directly.

See attached image for an example of what I mean. This is a snip from a PDF of one of the gantry beam end plates to accompany the DXF file that will be attached to the request for quote. The PDF is intended to give a few dimensions to cross check with the DXF as well as any supplementary information like the "centre punching" of the holes.

[peteeng]

Morning Jayne - If you are having the billets milled after profiling there is no need for marking the part. This is just extra cost. If you are manually drilling then this is worthwhile (maybe) Up here 16mm is about the limit for laser. The next question is how will the mill hold the part? You have two choices 1) In a vice which means the plate needs to be 2-3mm thicker then the finished thickness. The mill will profile and drill the part then the part will be turned over and the 3mm faced off 2) The part will be held in a vice the holes drilled then the holes will be used to hold the part to a sub plate that has threaded holes that agree wit the part holes then the pocketing and profile is completed. If your doing many parts then 2) is the go as you save metal and time....

For this job I expect that the extra cost involved in laser/watercutting for the billet is not worth it. But get it quoted for your own knowledge. A rectangular billet held in a vice is easy to do. Typically a rough billet is cut (bandsaw, cold saw, jigsaw etc) then the sides faced (milled) so they are parallel and its into the vice and much swarf.

I recommend that you do not use a dxf file for the data transfer. The best practice is to supply a manufacturing detail drawing and a step file. The company should be able to open a solid file such as step or parasolid or the native file if they use the same software you are using. DXF can result in various errors that you do not want. In a formal RFQ the terms used are generally that the drawing is the principle manufacturing reference and the solid is the geometry reference only....

My guess is your cheapest approach is to buy some plate from your local supplier (they may even billet it for you on their table saw) cut it up at home then take that to the mill....Peter

By the by - the square corner in the bottom of the part can't be done on a VMC in the same setup. You need to add a small reentrant radius to the corner if a square object is to fit to the corner so a second operation on edge mill is not required... The second op will add $$$

Your manufacturing dwg will need to state what's important about the part. Fits, clearances, tolerances so the machinist knows what to pay attention to. eg if a hole is to be threaded use the pilot hole size in the model. If the machinist uses auto hole finding and you use the nominal hole size eg dia6 for a M6 bolt then the program will drill a 6mm hole and you won't be able to thread it. That's why it needs to be stated on the dwg...

The other aspect of "marking" is that the machinist will have to register to the mark and they would prefer to register to an edge or edges. So if you are billeting a part using laser or some cutting process you would leave a machining allowance on edges say 2 mm for a finishing pass on the mill. Keep the whole thing on the mill much simpler and less $$$

[JayneV]

Hi Pete, thank you for lots of good info. I think I got my brain a bit muddled up with what I want to do. The machining costs are not an issue, my brother has offered his CNC milling machine during its down time. Having said that, I want to try and keep the machine time down to not take advantage of his generosity. Originally, the intent was to build a CNC at home with tools I had available back when we were all in lockdown. The design slowly became more complex requiring accurate machining for some of the parts. Thinking more about it, the only parts that need to be outsourced are the front and end plates which are larger than the capacity of my brother's mill, although if necessary I can cut those in half to fit his machine if outsourcing is too expensive. What I really need is rectangular pieces cut from plate which can be machined into final pieces. I suppose if I make the RFQ as a list of rectangular pieces in the relevant material thicknesses, that should be sufficient. One of my limiting factors is the raw material needs to be a manageable size that I can lift by myself and also fit into the back of my car to transport to my brother's shop The cheapest option would be to buy a full or half sheet of the different thicknesses as required and cut the billets out with a jigsaw or circular saw but there is no way I can handle large pieces.

Thank you for the info on machining inside square corners. I was aware of that detail, but since machining costs will be free, it can be done in two ops. Also, that is not a critical dimension so worst case I can square the corner by hand.

I'm guessing plasma cutting is the cheapest option for cutting the rectangular billets, how much do you think I should increase the dimensions to allow for final machining to size? Would 2-3mm added to each edge be enough to allow for the roughness of cut from a plasma cutter or should I add more, say 5mm?

[peteeng]

Hi Jayne,
Your local al supplier may be able to cut to billets on their table saw. Ask them first. Plasma will be extra $$$ and a poor edge. A large table saw set up for aluminium is the cheapest option which the supplier will have or have access to.
Since you are probably only going to make 2 of the discussed parts I'd machine them from one billet. Less handling and setting up. You have to think along the lines of letting the machine do the most work possible. Once its bolted down in a machine its best to get it to do as much work possible for you.
Since its on a machine let the machine do the work and add a reentrant fillet we don't hand finish these things these days!
To help with part weight you can make your parts half thickness and laminate them together or bolt them together. In terms of a machine design this is a sensible approach as shear loads are small. If the parts are epoxied together they are stiff and damp. This also may help with cost as buying more thinner plate maybe cheaper then buying a small piece of thick plate and some thin plate. You can also tailor thickness this way if you use a thinner plate and laminate it to the part vs starting with a thick plate and making lots of swarf. Just something to think about...

Take full advantage of your brothers offer, that's what families are for Peter

[peteeng]

I also meant to add that if the part is thin enough then laser cutting becomes more viable. If you laminate then milling not needed. This is the economical route. So if you had 12mm cut you can laminate to 24mm and 36mm and the part can have varying thickness as each piece could be a different shape.... Peter

[JayneV]

Hi there, not a great deal to report here, still waiting for the aluminium to be delivered. Today I finished cutting up the timber for the bench frame. Hoping to have some time on Saturday to build it so it is all ready before the aluminium arrives.