[iceblu3710]

Ok so here is my theoretical setup:

Mill: Fireball v90
-20TPI skrews on x & y, 5TPI on z

CNC Kit: Kelinginc x2 kit
-Except using the KL23H276-30-8B 282 steppers Torque Curve

I was using Mr. Beans Axis Calculator but have no idea where these numbers are coming from, if someone can walk me through it It would be greatly appreciated.

What I understand sofar:
-From the stepper Torque curve I think 4000pps is the fastest I should run it as that's 115oz-in (0.8Nm)
-20TPI is 0.05in per revolution
-1/64 microstepping I believe microstepping is shifing the current in a wave like form in the coils the achieve greater resolution than just having one coil on and the rest off right? If thats so 200/(1/64)=12800, so it takes 12800pulses to rotate the stepper shaft once?
-If the above is true then (12800/4000)/60=0.053RPM but it can't be

Axis Calc says:
200steps/rev, 1/64stepping, 20tpi skrew, 1:1gearing v= 18.75RPM with a 4000pps input or 375IPM, 0.0016in resolution.

I was expecting in the 10-15IPM range no the hundreds... Im trying to get the best accuracy combination out of a 20TPI table and that stepper kit, if it really will achieve 375IPM with those numbers then I could side mount the steppers and have a 11:1 belt system that at 4000pps would be 38IPM and an awesome 0.0001in resolution.

Anybody help?

[Stepper Monkey]

There is some likely a bit more reading you should do before you try setting up a stepper system, as while there are a couple of easily corrected simple errors in the formulas you used, I think there is also some misunderstanding of some of the fundamental underlying design concepts as well.

I don't have any good links about stepper design fundamentals off of the top of my head, anybody got a couple of good intros/FAQ pages to recommend for him to get started on?

First, lets fix the simple math error;
It would be 12800/4000*60 = rpm (or 18.75 RPM, or 0.9375 IPM)
(you need to multiply, not divide, as 4000 is per second, to get per minute would be times 60)

I won't get into the rest of the equations, as the arithmetic is correct but none of the suppositions or conclusions are fundamentally useful. They show a number of serious misunderstandings, and implies several more that aren't immediately apparent in the numbers.

So lets start with some basic formula;
Resolution is TPI x steps x microsteps, for instance 20tpi x 200 x 8 = 32000 steps/inch of travel or .00003125 inch resolution.
That is overly tight at only 1/8 microsteps. At 1/64, if the hardware could handle it (and it can't) resolution would be in the billionths.

Now to some concepts;

By any measure, a 20 tpi screw on a router is considered VERY slow. With your theoretical setup, less than 1 IPM. Routers generally have 5 or 10 TPI screws or faster to achieve the high IPM required to get proper feed for the materials they are designed to cut. It isn't about conveniently fast speed, it is a hard requirement to operate at very defined speeds appropriate to the bits and materials you are using.
Secondly, some stepper drivers can be for designed up to 1/256 microsteps, but it doesn't mean you can use them for every design type of steppers out there. That said, ALL steppers of the design type of construction used for CNC machining cannot accurately accept anything above 1/8 to 1/10 microstepping. There is no gain above that whatsoever in accuracy or resolution, just a serious loss of speed as you have to send a crapload of signals to get it to move very far as you can see by the above math. You may want to consider half steps even.
Third, resolution does not equal accuracy. They are entirely unrelated. Higher resolution on the same machine doesn't improve its mechanical properties. Too high of resolution simply slows down the machine unnecessarily.
Fourth, the torque chart is rated there in half steps. It is confusing for them to have done it in quite that way. 4000 pps in that case is equal to 20 RPM. You should look at RPM, not PPS. That table also is only really good for that specific configuration of microsteps, current, and voltage. It can be radically different with a different setup (BTW, mad props for getting far enough into the theory to look at the torque curve! Most folks just look at biggest oz/in rating and think thats all there is to it)

There is more to be said about that stuff, but thats a lot of the main important stuff. I want to cover what isn't shown by the numbers though;

There is a lot of math covering what speeds and feeds are required for any given combination of bit and the material it is cutting. It is VERY important to any sort of proper cutting (and extremely important to accuracy and finish) to run bits at their proper speeds. You need to determine your IPM feed rate needs properly before you can get a target as to what you need, regardless of resolution. Start there first, you are jumping ahead of the game doing any math before you know this number.
Secondly, if you do need the high accuracy and slow milling speeds you mentioned, a gantry router like the Fireball isn't designed to handle it and the spindles used on a router aren't appropriate for that type of work either. Routers tend to be much bigger, faster cutting and much lower accuracy than mills, also less rigid for cutting softer materials. Do you actually need a mill instead?

What are you material cutting and accuracy requirements?

Let us know what jobs you actually intend to do with it, and we will help walk you through the minefield! It is really commendable that you are diving into the details and the math, a lot of people never really try. It isn't hard or magic but there is a LOT more to it than you have access to right now, so keep at it and it will all make sense before long!

[philbur]

This one might help:

http://www.cs.uiowa.edu/~jones/step/

Phil

I don't have any good links about stepper design fundamentals off of the top of my head, anybody got a couple of good intros/FAQ pages to recommend for him to get started on?

[iceblu3710]

Thanks for the massive post Stepper Monkey, I should stop posting before midnight as it appears im quite stupid when tired.

Mr. Beans axis calc gives very different numbers than what your saying but theirs alot of strange things the program does, such as giving massive IPM calcs and the motor rpm never changes even when you set a higher velocity which is impossible.

Resolution and mechanical accuracy are related if you have a machine thats the most ultra tight $150k commercial unit that has 0.0001 accuracy and you drive it with a stepper and 5tpi screws you will only ever be able to get 0.001 out of the machine as you cant turn them any less. 1/(200x5)=0.001
Obviously though thats a retarded example most mills under $5k are probably happy to get 0.001in accuracy.

I have no idea what screws are on the fireball v90 as the guy has not emailed me back for a week and two emails, I just took my numbers from a x2 mill thread that the table apparently has ~20tpi threads and a coarser z ~5tpi. And as for spindle considerations im only mounting a 30,000rpm wolfgang eng or similar spindle as all my work is on pcb's.

Feed requirements:
Im not sure the range of bits to use but have looked at the usual websites and bits such as the thinktink.com fish-tail chip-breaker solid carbide router bit - 1/8 in. shank are recommended for commercial units. Most rules to maximize bit life say feed rate should be %4.5 bit diameter per revolution so:
0.001in @ %4.5 & 25k spindle = 1.125ipm
0.008in @ %4.5 & 25k spindle = 9ipm
0.012in @ %4.5 & 25k spindle = 13.5ipm
That would be to maximize bit life, Every youtube video I see uses 20-60ipm and most use 45deg v-bits though im sure the bit life calculation is similar.

So my goal was to figure out the set of equations I would need to get an electrical resolution of 0.001in, a feed rate around 60ipm for rapids, and then figure out what type of gearing/screws would work best from there.

New math examples:
Ill be using say max 6000pps this time
Shaft RPM @ 1uS = (200ppr) 30rpm
Shaft RPM @ 1/2uS = (400ppr) 15rpm
Shaft RPM @ 1/45uS = (800ppr) 7.5rpm
30rpm with 5tpi screw = 6ipm
15rpm with 5tpi screw = 3ipm
7.5rpm with 5tpi screw = 1.5ipm

Now those feed rates would be cutting values as a higher pulse rate would cause the stepper to loose more torque and not have enough guts to drive the gantry well enough. Even the maximum rates of 8000pps is 55oz-in, 40rpm, 8ipm which is nowhere near a fast rapid, a 4x4in board would take almost an hour at those speeds

And to answer the final question of accuracy requirements and do I need a mill instead:
I was originally going to buy an x2 but had to re-think what im going to actually use the device for as you can't get what I want from one machine in particular. Originally I thought of everything I wanted to do which was, Shell hardwoods, cut 0.060in 6061-al and brass not milling just engraving and cutting out shapes for custom enclosures and metal cogs for clocks. Also wanted to cut plexiglass for cogs, and mill hilly accurate PCB's for QTFP and 0604 smt components.

I realized that I would be better off buying a PCB only mill as I will be making pcb's 4-5times a week while wood, metal and cogs 2-3times a month. And latter on down the road buy something like the x2 for the monthly work.

For a pcb mill my requirements are a minimum 4x5in table but 8x10in or larger would be better. Ill be using a Wolfgang type spindle no matter what as I need 25-35k rpm out of it and they claim 0.0004in runout which is awesome. The parts I will be milling are 0603 surface mount (0.063x0.032) and QTFP which can change but ~0.015 between pins. Trace width's I want are 0.01in and larger but 0.015in is absolutely necessary, Isolation paths are 0.01in and larger. Essentially I need 0.015in accuracy on the axis.

Im hoping if I get a gantry style mill their pretty rigid when running a 35k rpm spindle at minuscule penetration as most machine only deform when your trying to push through material and that's where most of your backlash comes from right?

Thanks for the replies, I shall await your response.

[Stepper Monkey]

When you get into exceptionally small carbide bits the math does tend to change, and with 90 degree included angle (45 degree half angle) engraving bits especially you can run them very fast without issues - people recommend different optimal feed rates based on personal experience but they are all at least 2x and often 4x-5x or more than you would think over using larger bit calculations.

Remember that PPS isn't really a measure you should be using for stepper performance. It get complicated and involves a lot of things like voltage and inductance, but RPM is really what you should be looking at. Whether you are at 2000pps at full steps or 16000pps at eighth steps, the motor is still running at the same RPM which is more indicative of performance.
Secondly, remember that a stepper at a 24v supply might drop off to very little torque above, say, 20 RPM - while the same stepper driven by 48 volts might have more torque even at 40 RPM. At 72 volts, the stepper could keep the same torque well through 60 or 70 RPM.

In a related note, you likely need a HELL of a lot less torque than you think. There are calculations to figure out exactly what you need, but in effect if you can supply more torque than the amount of side loading it takes to snap your bit, the weak link in the chain isn't your steppers. Even 20 oz/in multiplied by 5 or 10 or 20 times by the mechanical advantage of the screws is still enough to do some seriously heavy work and or damage things. Don't worry about dropping below the 115/oz range - with the screws you mentioned that would mean around two hundred pounds of motive force at its lowest ebb, well beyond what you need or can use unless you plan on sitting on the worktable while its cutting...

"most machine only deform when your trying to push through material and that's where most of your backlash comes from right?"

Mechanical deflection and backlash are unrelated. Backlash is simply a problem with play in the screw/nut interface causing lost motion when reversing direction. (When the screw changes direction, it spins a little before taking up the slack and catching the screws opposing thread faces, causing the nut to "miss" some of the screws motion). This can happen in any size or rigidity of machine.
A machine flexing or vibrating under load is a bad thing, and an unrelated one to backlash. You are correct, however, in that you will have very little loading just milling PCB's and therefore mechanical deflection is not a big problem.

You likely need a system with far faster speeds at far less torque than you think, with a higher voltage supply and lower inductance motors. Trust me, for this stuff, the maximum 25,000 or 35,000 pulses that Mach delivers each second can actually be the limiting factor!

[iceblu3710]

The KL23H276-30-8B steppers are the lowest inductance motors in the NEMA23 listing at 2.2mH, I haven't found any steppes since that provide anything lower.

The KL4030 can only handle 40v but if I threw on some high speed switching mosfets I could easily handle anything up to 250v, 2.4ns acting could handle ~200kHz input. Other than that Its a $10 upgrade to the KL-320-36 36V/8.8A supply.

I wont be using Mach, primary Linux EMC for everything. I dont know if it has any limiting factors as I didn't even know Mach had a pulse limit.

So I guess the new question is should I be looking elsewhere for motor/controller packages? I was checking out the UHU servo controller but its easily $300 for the three boards and I cant find servos anywhere as ebay has very few listings.

Also can you recommend a gantry style mill that will work the best for pcb milling? I don't mind modifying and will definaty need experience and tweaking to get my wanted accuracy but would prefer something close to the objective right from the beginning. I haven't found any reviews of the fireball as he just started making them and the yahoo group has no talk on its boards. I have $2000 to spend and sofar my only ideas for low cost are the fireball and keeling kits, or a K2CNC all aluminum machine is freakin sweet but $1200 base price is hurting, add a t-slot table and remove the tool attachment and its still $1400+300 shipping.

[Crevice Reamer]

Just FWIW: The Keling kits are almost the same price for the kit as for the individual components. You can put your OWN kit together. Most of the motors need to run at 48 volts or higher for best performance though.

The motor you have chosen would run very nicely at 48 volts. But it also requires 4.2 Amps. The 4030, even at 36 Volts, can only put out 3 Amps.

CR.

[TheNotchJohnson]

Im interested in 3D milling. I work on motorcycles alot and want to be able to cut parts for certain applications. I am hoping to take aluminum castings and have a cnc machine do automated milling. I am very good with cad and have good math skills. I figure with the right resources I can build a cnc to suite my needs. I just dont want to invest in a design without confirming its going to work properly, in a mathmatical sense. At this point im looking for technical data and resources that will help me better understand the calcualtions on speeds and feeds, rpm and torque values, ect. maybee some charts explaining requirments on different alloys, things of that nature.
For example, when I am choosing a recirculating ball nut shaft, how do I choose the right diameter and pitch, ect. or when gearing up a stepper motor, how to determine the right ratios for the design.
The issue I am having is finding a resource for this information, if you know of a fourm, website, or even a book with such data, let me know. Thanks.

[Stepper Monkey]

The KL23H276-30-8B steppers are the lowest inductance motors in the NEMA23 listing at 2.2mH, I haven't found any steppes since that provide anything lower.

My fastest machine is currently running Pacific Scientific steppers at 0.8mh, which is good for a very nearly flat torque curve out to extremely high usable cutting speeds.

Keling isn't the only source for steppers. 2.2mh is indeed very good for almost all general use, but if you do need lower inductance there are a number of companies that have winding configurations of much lower inductance than Kelings selection, and much more efficient motors to boot.