First Build-What comes first... Ball Screw? Spindle? Linear Motion System?

So I'm building my first CNC system. I have purchased 3D printers and done lots of research on CNC router tables etc. But as I try to start, each time I think I have made a decision I realize to buy my ball screw I need to know the weight of my spindle and gantry. To buy my spindle I need to know if my ball screws and linear motion rails are big enough. So I try to use the calculators I always see but they require information I don't have yet.

How can I start to build this without all the information?

My specs are this:

Steel table and Gantry (I will fab)
Cut area: 4'x5' (4'x4' for wood but eventually I want to put a 4th axis at the end so I can turn material and do end work.)
I'll do 34 Nema closed-loop hybrid steppers
I would like to use Ball screws
I will cut aluminum so this needs to be a stout machine

But how can I get past my Fusion 360 steel table design without knowing what ounce stepper since I don't know wight etc. etc. etc?

PLEASE...Don't post if you want to just say you can't you don't, etc. I a looking for answers, not all the things you think I am doing wrong. If you want to point out things I should do because I am missing something GREAT! But too often I just see responses from people that know everything but won't offer any real help. I am not trying to be a jerk at all, seriously. I just need help from good people willing to help a first-timer on his project. I appreciate your help more than you know!

Given that you are not trying to build a spacecraft where every ounce counts, this really does not need to be over engineered.

You can cut aluminum (or steel) with a Dremel tool, but not very fast. I guess I would start with how much electric power do you have available to run the spindle motor. This will dictate the maximum spindle power/size. Once you have chosen a spindle, look at the published shipping weight of it, that's close enough.

Take a look at the rail load spec's for various rail/car combinations. Just keep these spec's in the back of you mind for the moment.

Then add up the mass of all of the bits & pieces that make up the gantry assembly, including the Z axis hardware and spindle. This gives you the moving mass. Then decide on the acceleration rate that you want to achieve. F=MA Force = Mass x Acceleration. This will dictate the ball screw/axis motor combination required. Don't forget to look at the torque curves of the axis motors. Stepper torque drops off rapidly as the RPM increases.

All of this will give you a starting point for doing the calculations.

Or based on your spec's, you could throw out everything I said above and go with some 2005 or 2010 ball screws, 20mm rails, and around 1200 oz/in axis motors and build the machine. :)

Hi,
I'm not in favour of closed loop steppers, they cost more but you gain little over good low inductance open loop steppers. 34 size steppers tend to have high inductance and therefore have
very modest to speeds, whereas low inductance 24 size steppers go much quicker, with less torque authority, perhaps but way faster. Either way that is a decision you can make closer to the time.

If you really want closed loop performance,and who doesn't, then get AC servos (Delta or DMM recommended)...don't mess around with steppers at all.

I would proceed as Jim has advised.

I would build starting with the ballcrews and linear rails, then the frame. Its likely that your budget will run out before you finish.....and its better to have a working (ie all axes working) machine
to which you have to put a miserable little spindle because it all you can afford vs having a brand new spindle but a machine that won't work because its got no ballscrews.

Craig

Thank you so much for these replies. This is EXACTLY what I needed to find out. I have plenty of power since the house I just bought was owned by another one of us... He put a second panel in for a garage shop 200 amp... Lol I want to build one at home first then a bigger one at my shop.

Interesting about the servos as well. I guess I assumed that only 34 would give me the ability to run with the torque, speed and carry the weight of a steel gantry. If 23 works I'm all for it.

Funny you say don't over engineer because that is what I felt like I was doing each time I tried to start calculating. I literally almost said F it and just bought screws to start the process.

What should I expect in terms of speed etc. for aluminum and for wood? Aluminum would not be crazy production stuff just more about... Hey I need to make this part out of half inch plate can my machine do it even if slowly... I'm someone that has been a general contractor for 20 plus years and have been building electronics machines welding etc for a long time so I know I can make a fairly nice machine. I just don't know how high to set my expectations. I just don't want to build it then say why didn't I this and that... Then have to do it again to get the machine I really want.

One more question. Can 80x20 etc actually be rigid enough for a machine like I want? I see them but keep thinking it doesn't look rigid or heavy enough.

Thanks again guys.

Hi,

Quote:

---

Interesting about the servos as well. I guess I assumed that only 34 would give me the ability to run with the torque, speed and carry the weight of a steel gantry. If 23 works I'm all for it.

---

The extra torque of the 34 is indeed desireable but so often they run out of torque over a few hundred rpm. There are many who've posted on the forum that have used 34's thinking the high torque
was the answer only to find it wouldn't go beyond 50ips....and they replaced them with 23/24 size.

If you want speed and thrust get servos.......but even entry level servos are still double the price of a good open loop stepper and drive.

Rigidity is key, it doesn't matter what you are cutting there will be cutting forces. ANY flex in your machine counts as inaccuracy and worse....vibration. If wood requires 'A' units of
rigidity, then aluminum requires '3 x A' units and steel requires '10 x A' at a minimum.

You may be wise to consider reducing the size as its much easier to make a rigid small machine than a rigid large machine, or alternately make one machine for wood and another much smaller more rigid machine
for metals. Many newcomers fall into the trap of trying to make one machine do all things.....it is possible but WAY expensive or the compromises you are forced make end up with a machine that doesn't
really do anything well.

Bang for your buck goes to a welded steel frame. Heavy rectangular sections welded together can make a great job. Downside is that you will have to stress relieve after welding and then machine the critical surfaces flat.
Heat treatment is not cheap and nor is machine time on large format mills, but still often the best result for the money spent.

Craig

You can get servos from about 50W to at least a couple hundred kilowatts. So size availability is not a factor. You can get anything you need to run the machine. For a 4x5 with good performance 400 to 750 watt would be more than adequate. With a nearly flat torque curve through out their operating range they far out perform steppers. Also given their intermittent overload capability of at least 200% gives you very high acceleration capability

The down side of servos is the additional cost, A 1123 oz/in NEMA 34 closed loop stepper & drive & cables is going to cost around $315, where a similar sized servo motor & drive & cables is going to run around $500 for a 750W system. (Pricing from Automation Technologies web site) Like Craig, DMM or Delta servos would be my choice also. I have both on my machines.

If it's in the budget, then servos are the only way to go.

8020 extrusions are certainly adequate for a wood router frame, but for the added stress of aluminum machining a welded steel frame would be better. I think you can make a welded steel frame without needing to do any post machining on it, you can always shim the rails to keep things straight.

Hi J22- All that you want can be done. What is your tradeskill level and resources? Can you weld steel for instance? NEMA23 will do your job easily. If speed is not an issue use 5mm lead and you will have a few 100kgf at your disposal. Use 10mm and you will have adequate forces and good speed. N34 are way too big for what you want and as discussed if you use asian servos which are steadily coming down in price then you get high torque and high speed all in one go. Most builders that talk about what your talking build an adequate machine but since you have mentioned aluminium you will need to have a deep think about the Z axis. This is what lets people down the most. If you start designing from the outside and leave the Z till last you end up with no geometry left for a super stiff Z. So start at the Z and make it super stiff then work outwards to design your machine. Find a build thread in here that you like and follow it closely. As they say there is deep knowledge on this site..

The answer to your first question is that you need all the information up front, which is impossible unless you copy a good design. So you have to solve in a spiral manner down the rabbit holes. The whole problem is interconnected. You solve part of it, move onto the next bit and that changes the bit you just did. But now you are closer to the final design. So you go round and round refining and specifying until it all makes sense....Good Luck Peter