RC Aircraft Hobbyist

[cambridge_cnc_stuart]

Hello to all,

I am new to the world of CNC Routering, if that's correct spelling / grammer. I have a few years experience on 3D Printing if that would help at all.
I would very much like to build a CNC router to use my Dremel with a bed size of approx 1000mm x 600mm. The bed being made of 18mm MDF. Ive been told that for a router of this length I should be looking to use belt drives. I am not looking for the land speed record for this router but neither do I wish for the "Escargot" of all routers :hysterical:

I am aiming to cut bulkheads and wing ribs for Large RC aircraft / Jets, Carbon Fibre Plates for Multirotors.
One that can cut Balsa Wood, Ply and Carbon Fibre sheet, Ideally Up to a MAX 10mm thick.

I have downloaded the plan pack from Bluurmax cnc but they would need adapting from imperial to metric. I have looked at some software and this does not seem unrealistic.

As for electronics I am open to suggestions. I have a budget around £600 / £1000 but naturally keeping the overall cost down would be advisable, as this year I have purchased three turbines, four jet airframes and made a Vacuum Former and modified my 3d printer three times.

I look forward to hearing from the knowledgeable on this matter

Happy CNC'in!

[m.marino]

How accurate and how fast do you want to be cutting? The design links to a gantry router that is acrylic and MDF. Not the greatest materials to work with. You might want to look at some of the designs here. There are a few designs that use ply laminate with epoxy that have a reasonable life span but it really depends on how long you plan on using your machine and what accuracy you are looking for and longevity. Acrylic doesn't like stress and MDF like's water a lot.

Design depends on what you are looking for and what you want out of it. Gladly help as I can.

Michael
MM0MSU

[cambridge_cnc_stuart]

Hi Michael

Thank you for the reply.

I hear what your saying. I don't mind making it our of square aluminium channeling and Plates. As for accuracy. As it would be used for making RC aircraft parts. I'd like it fairly accurate. I have zero experience of speeds when it comes to these sort of things. As I am a newbie. Any and all advice is welcomed

[m.marino]

Strongly suggest reading through 4 or 5 of the different build thread than and getting an idea of what different folks have done. I am in the process of getting things ready to move so my CNc gantry is currently down and when i bring it back online it will be using a 2.2Kw spindle to allow both a large range of tooling and reasonable cutting ability in all the materials I work with and be able to add some additional ones as well.

Best of luck and Your link points out that I really need to put up the design of my gantry router and how to set it up for different sizes and build materials.

Michael
MM0MSU

[A_Camera]

I think that rigidity and accuracy is CONSIDERABLY more important than speed, especially for aircraft. In your case, balsa is not an issue, but carbon fibre definitely is tough and demands a really rigid machine. I would not build a CNC from anything less than aluminium. Belt drive sound good in theory, but I think that in reality, if you want quality and accuracy and no backlash then you can't just take any belt... If I'd plan to build this machine I'd still use ball screws and ball nuts. I don't think you will have issues with speed, unless you have unrealistic speed dreams. With steppers you should be able to get about 7000 to 9000 mm/min maximum speed, at least if you are using fixed gantry moving table design (which is better regarding rigidity), and probably something between 4000 to 6000 mm/min if you go for moving gantry design. Remember that maximum speed is normally not the milling speed, that depends on the material and many other things, so in all honesty, I don't think there is any advantage in belt drive in your case, even if you are limited in top speed.

...but I think the biggest issue is your budget limits. I really think that you have to double that (or maybe even more) to get a good enough machine to mill carbon fibre.

[Zeeflyboy]

Belts would be fine, but you have to have them far thicker than you would have thought... that then eats into the cost saving. For the price of some c7 rolled screws from china these days I don't really see why you'd use belts. It's not like your axis is too long for screws.

Dremels are also a pretty bad choice in almost every regard, not least that when cutting carbon fibre they will blow the dust around even more than it already will be, and CF dust is nasty stuff for humans and machinery. My first machine had a proxxon router which is similar to a dremel but with better bearings, still blew through two of them in fairly short order once I asked it to start machining FR4/G10 and CF.

If Carbon fibre is on the menu, you really need to think about dust management which really ultimately ends up with cutting underwater. For that, a spindle that doesn't blow air everywhere (ie a water-cooled spindle) is really ideal... bonus is that they are also far more quiet. Even HEPA filters on Vacs can't guarantee you'll trap all those pesky CF fibres. A company I did some work with had no end of problems with their prototype electronics for months, eventually traced it back to them having a new CNC machine that they had been using to prototype CF frames for drones - despite the vac shoe and hepa filter on the shop vac, enough still escaped to settle on their PCBs and components to cause serious misbehaviour. They switched to cutting underwater and had a top to bottom clean of the lab, no problems since.

[cambridge_cnc_stuart]

Wow A_Camera & Zeeflyboy
Thank you so much for your reply and information. Very informative and the more information I get the better informed decision I can make. I am still at the design stage. Had come to the view I shall be using aluminium extrusions.
I have also been looking at a dust shoe.
I have also looked at the maximum size I need. This has been reduced slightly to a working area of 750mm X 450mm but taking into account losses I'm still thinking an overall footprint of 1000x600???
Also I have been looking for designs and have found it difficult.

Now this maybe the wrong at round, but to look around the software I have purchased Mach software......

Realistically, I think the majority of projects will use Balsa wood or Ply too 5mm thickness. I can't really imagine needing too cut CF. I already have 8 Multirotors. For the sake of safety, I think the safest way I'll be too outsource CF cutting.

Regards

[A_Camera]

Zeeflyboy is right, you should be getting a better spindle than the Dremel. I have a Dremel tool also, but I think the air flow is not the worst with that, but the extreme noise and the crappy bearings. They are also pretty weak. So, my opinion is that you should add a real spindle and a VFD to your budget. Of course, even if you give up on carbon fibre, adding a dust shoe is a very good idea, regardless which sort of spindle you use or which material you mill. Regarding dust and carbon fibre milling, one way of avoiding the dust is milling in a water bed. Have seen some people doing that, but you need to build the rig, so it may not be that simple to do.

Personally I gave up on Mach3 and using UCCNC from CNC Drive. I am also using their UC300ETH, which works also with Mach3. My drivers are DQ542MA and they work very well.

[cambridge_cnc_stuart]

Thank you Guys for your replies and information. I am getting an idea on what I should and shouldn't do.

How does this sound? I will make the frame from aluminium extrusion of say 30mmx60mm long sides y-axis and 30mmx30mm X-axis. I'm thinking of using two 23nema Steppers on the y-axis unless this is overkill for the length. A 23nema on the X-axis and a 17nema on the Z-axis
I am leaning more towards a moving gantry design. What is the best design / way to allowing the gantry to move.?
A Spindal of around 2.3kw That's as far as I have gotten so far.

A few questions now that I wonder about!

1, Do CNC routers use similar electronics to a 3D Printer?
2, Do the need microswitches for end stops?
3, I see a lot of electronics in boxes with fans?? Do they produce that much heat?

Any opinions/ advice greatfully received

Thanks once again

[Clive S]

I will make the frame from aluminium extrusion of say 30mmx60mm long sides y-axis and 30mmx30mm X-axis. I'm thinking of using two 23nema Steppers on the y-axis unless this is overkill for the length. A 23nema on the X-axis and a 17nema on the Z-axis
45x90 is a better size as the slots fit nicely with the BK and BF bearing etc. (assuming you are going with ball screws)

nema 17 are not good enough for the Z with a 2.2kw spindle.



1, Do CNC routers use similar electronics to a 3D Printer?
2, Do the need microswitches for end stops?
3, I see a lot of electronics in boxes with fans?? Do they produce that much heat?

1. Yes but need more power suggest, using drives AM882 run at 68v
2. its always a good idea to use limits and homing switches sensor type are cheap.
3. It is usual to put a fan in the box to get the heat away

It will be difficult to build on the budget you have stated I would estimate about £1500 to get a decent machine built by yourself.
I take you have read through some of the build logs on here. Also its never a good idea to buy kits of electronics as they are never matched.

You would be doing yourself a favour if you could draw the machine out in cad first.

[Zeeflyboy]

Just out of interest you don't state what version of mach you bought... If mach4 you need to be sure whatever motion controller you use will actually work with it.

General Basic principles of the electronics are similar to 3D printer, but obviously beefed up due to the higher power requirements. Power supplies tend to be higher voltage (at least 48v) and motors have their own dedicated stand alone drivers. Due to VFD noise one needs to be a little stricter regarding good earthing practises. Most 3D printers are also stand alone, most CNC's require a computer running something like mach/ucnc etc so considerations about Ethernet/USB/parallel port and the various advantages and disadvantages of each need to be considered.

[cambridge_cnc_stuart]

Sorry I hadn't appreciated there was more than one version, everywhere I read it stated Mach3, so that was what I looked for.

Does anyone know of some plans that I can follow and save making an expensive mistake

Thank you

[cambridge_cnc_stuart]

Well, the more I read, the greater my confusion gets.

Is there anyone / where near Cambridge or surrounding area that I can see a CNC and ask questions?

Ideas Please

[A_Camera]

Thank you Guys for your replies and information. I am getting an idea on what I should and shouldn't do.

How does this sound? I will make the frame from aluminium extrusion of say 30mmx60mm long sides y-axis and 30mmx30mm X-axis. I'm thinking of using two 23nema Steppers on the y-axis unless this is overkill for the length. A 23nema on the X-axis and a 17nema on the Z-axis

It does not sound good at all to my ears. 30x60 and 30x30 is for 3D printers, not for a CNC. Rigidity and mass are important even if you give up on milling CF. My machine is mainly built using 45x90 extrusions and is definitely not overkill, not even for plastics which I mainly use it for. Wood is not different. I am also milling aluminium but would not start milling CF with it, I think that would be too painful. You should not use any NEMA17, again, they are good for 3D printers or other things where forces are considerably less. Even the Z requires a lot of force and pretty high mass to move up and down at a fast speed, so don't go for a weak solution. There is no advantage in using a small motor, the price difference is negligible.


I am leaning more towards a moving gantry design. What is the best design / way to allowing the gantry to move.?
A Spindal of around 2.3kw That's as far as I have gotten so far.

Good luck. That's what most people chose because it is the sexiest design. The only advantage it has over a fixed beam type is that it looks more impressive when you show it to your friends and that it takes a bit less space. Fixed beam is easier to build, easier to square and align in every angle, AND provides better rigidity. A 2.2kW spindle is pretty heavy, even the 1.5kW I am using is too heavy for NEMA17.


A few questions now that I wonder about!

1, Do CNC routers use similar electronics to a 3D Printer?
2, Do the need microswitches for end stops?
3, I see a lot of electronics in boxes with fans?? Do they produce that much heat?

Any opinions/ advice greatfully received

Thanks once again

To answer your three questions:

1. Yes, the electronics in principle the same, but a CNC requires much more powerful motors and better drivers. Also, the 3D printers I looked at, are not driven directly from any software, you generate an STL file and transfer that to the printer using a memory card or some other means and print that specific file. A CNC uses a PC which generates G-code and sends that code to the motion controller. The software can be Mach3, Mach4, UCCNC and so on, some are more expensive than others, but they all communicate with the CNC using the parallel port or through a motion controller using USB or Ethernet, like the UC300USB or UC300ETH or others. Those motion controllers require a driver utility, a plugin, which the software you are using uses to communicate with them. As an example, all the UCx00 from CNC Drive have a Mach3 plugin but no Mach4 plugin. Of course, all can be used with the UCCNC software, since it is made by the same company.

2. Yes. It is a very bad idea to leave those out from the design because they are actually protecting your machine. I don't think there is anyone on this earth who never hit the limit switches. My machine runs at 9000mm/min and smashing into the sides at full speed is a pretty nasty thing. Even with limit switches it means a sudden stop and a loud bang, so it is scary, but it saves the machine from committing suicide.

3. Yes and no. The heat depends on your drivers. Some drivers get very hot, others not at all. My drivers are never hot and I don't have a fan in my electronics box. I do have a large fan spinning at low RPM in my PSU box and that fan is temperature controlled, the sensor is on the rectifier. At normal temperature the fan can't be heard, at 35 degree Celsius the rpm is increased to 60% and then gradually if the temperature continues to rise. I don't think I ever had the fan at 100% rpm. The VFD has a fan as well, and that is also spinning at low RPM unless the heat is increased. I have never had the fan at 100%, so that is also pretty quiet. Steppers can get hot, up to 60 degrees, but that's normal. I have no cooling on those. Note that I have all those in three separate pretty large boxes. The VFD has it's own, delivered by the factory, but the PSU and the electronics is my own design, so those use the boxes of my choice. Many people build everything in a smallish box, which I think is a bad idea, but it is up to each to decide what they prefer. In those cases there is probably a need for fans because the air inside will be heated and the lack of natural air flow is prevented, so you must have a fan (at least one) to generate air flow and cool the contents. Also many people use cheap Chinese switching PSUs and all those have a small fan which is spinning all the time, generating high noise. Also, if you have a powerful machine you may need individual PSUs for each stepper, generating even more noise. So, in short, there are always fans involved somewhere, but the number of them and how much noise they generate depends on the design.

A CNC is much more complicated than a 3D printer, so don't underestimate the challenge you are facing. While a 3D printer works in principal the same way, I don't think they are actually comparable in any other way than both apply synchronized axis motion on 3 axes.

[A_Camera]

Sorry I hadn't appreciated there was more than one version, everywhere I read it stated Mach3, so that was what I looked for.

Does anyone know of some plans that I can follow and save making an expensive mistake

Thank you

Perhaps you can find some free of charge, but I think it may be difficult. Also, DIY means that you WILL make mistakes, so that should be added to the budget.

Here is a link to a current build log:

http://www.mycncuk.com/threads/11266-3-Axis-900x500mm

The machine is about the same size you are planning for. Maybe he is ready to help you out, and maybe you can take a trip to him and have a look... Even if he is not in Cambridge, in your case I think it is worth to take a trip even if it might need a night in a hotel or B&B. He can definitely give you an idea about a realistic budget.

[cambridge_cnc_stuart]

Perhaps you can find some free of charge, but I think it may be difficult. Also, DIY means that you WILL make mistakes, so that should be added to the budget.

Here is a link to a current build log:

http://www.mycncuk.com/threads/11266-3-Axis-900x500mm

The machine is about the same size you are planning for. Maybe he is ready to help you out, and maybe you can take a trip to him and have a look... Even if he is not in Cambridge, in your case I think it is worth to take a trip even if it might need a night in a hotel or B&B. He can definitely give you an idea about a realistic budget.

Thank you A_Camera for taking the time to send me an in depth reply, it really is appreciated. I'm quickly believing I am out of my depth. Drowning in my innocence. I was quite confident to begin with, but that as now evaporated.

Having spent many hours watching videos, reading builds, searching the Internet. I am now totally lost, confused and despondent.

Back to the drawing board.

Thank you once again

[cambridge_cnc_stuart]

Clarification, all my 3D Printers have the option of printing from GCODE but I use them connected directly to a laptop or Desktop PC and can make changes on the fly. I have never used the SD card approach. The software I use if Simplify3d or Repetier Host.

Respect

[Boyan Silyavski]

There is no need to spend more hours watching videos. Here on this forum 99% of the builds are exemplary plus the guys earlier have pointed you to all the details.

Just to resume:
nema23 3nm motors on all axis driven by AM822 at 70VDc
1610 ball screws on all axis, 1605 on the Z, pulleys HDT5 15mm wide 20t from motor to ball screw
2.2kw spindle is the typical compromise between power money and ability to cope with tasks
20 size square rails Hiwin or chinese copy, don't use open builds or v rollers or skate rollers or other crap.
buy all hardware from BST Automation Aliexpress or other reputable seller.
Either use the recommended aluminum profile or make a structure from 80x80x3 or 100x100x3 box section.

Just read the builds in forum, all is there detailed and with pictures. i don't know what is with all to imagine that they are engineers and design every single machine from scratch. Just copy a well known and successful build and spend your time using the machine, instead of designing it. if you are into welding use my first build as inspiration and there are free plans in Sketchup. If you are into aluminum even better, many good builds.

You will need proper motion controller, not 3d printing solutions. Ethernet is what people use nowadays, not USB. CSMIO and Mach3, offline DDCSV1, UCCNC controller and software and so on.

Also you will need a seriously big enclosure, Power supply ets.

Once you start the build thread people will help you there so all will progress fast. But at the end of the day you will need to read some build logs,. Without that it will be too long to explain all. And dont read logs from the Zone as you will get too much useless information in your head.

[Clive S]

Very good post Boyan:applouse: Have a look a joe's log http://www.mycncuk.com/threads/4755-I-ve-built-this-machine-100-different-times-in-my-head%21%21%21 all done without special tools

[routercnc]

Very good post Boyan:applouse: Have a look a joe's log http://www.mycncuk.com/threads/4755-I-ve-built-this-machine-100-different-times-in-my-head%21%21%21 all done without special tools

Also some finished machines here:
http://www.mycncuk.com/threads/10066-Finished-builds-picture-menu