First Time CNC Builder Needs Help!

[Walter23]

Hey Y'all!

I'm about as new as they get. I stumbled across a couple homemade CNC Machines on youtube and have been obsessing over the idea of building my own ever since. I am usually pretty good at building custom things. I'm a perfectionist, and I have a background in mechanics, electronics, and many years of just tinkering under my belt. I think that my main struggle with this project will be not so much with the construction, but the theory and workings of the driver board. I turn to you for help.

I've been through just about every 'Instructable' there is on DIY CNC, but they all seem to be very thorough in the design of the machine, but very lacking(IMO) on the building and connecting the driver. My plan of attack is to build a cheap prototype using printer stepper motors(unipolar), a home-made breadboard driver, my pc's parallel (printer) port, and a basic axis guide. This stage is not so much for building a finished product, but more to help me understand the fundamentals and learn from trial and error. Possibly only X,Y axis, only capable of cutting soft material like Styrofoam, and about 12"X12"

After this stage is complete, I will then make a more functional desktop CNC with more capable stepper motors(bipolar), higher precision axis guides, and more fail-safes in the driver design. 3-axis, able to cut plastics, wood, and soft metal, and about 2'X2'. Then maybe if all goes well at that point I'll move onto large scale...

So now that I've laid out all the background I can think of, let's get to the point where I'm stuck - building the Driver. My goal with the driver is to make it as basic and simple as possible without sacrificing efficiency or safety(within reason). Obviously if leaving out a resistor will fry the board/motors after 5 minutes, it's going to be more cost effective to add the extra components to the original design. This will hopefully serve as a universal basic driver design for others just getting into the hobby as well. Easy to build, easy to understand, easy to modify.

I'm not sure what the bare minimum components are, but what I'd like is a list of these components with model numbers, quantity required(per axis or total), prices, and best place to purchase.

Then I'd like a schematic that is easy to read, with plain English instructions on building it, setting it up, and running a basic test after finished using Mach3(as it seems to be most common)

I already have a Breadboard, AC Adapter (Selectable 3,4.5,6,7.5,9,12V), Resistors, Digital Multimeter, and basic tools, 2 unipolar stepper motors(from printer), and parallel cable.

If someone more experienced would help me out with this I would appreciate it very much. After completion, I plan on making a video tutorial start-to-finish as to 'give back' and help others like me get started.

Also, two more questions, I assume I need a 'chip' of some sort, and I stumbled across this: L293D Motor Driver IC (SKU: L293D) - HVW Technologies

Is this any good? From the little I comprehended about building a stepper driver I think I need a 'chip' of some sort for each axis. This would do unipolar or bipolar if I'm not mistaken and it seems affordable and capable. But I have no idea what I'm talking about when it comes to these things...any thoughts?

Secondly, while pulling my old Epson Stylus C60 apart, I wondered why I couldn't use the existing 'chip' and connections and simply mount the motors where I see fit and control it as though it were any other board. Is this possible, if so how? If not, why? Details please!

Lastly, I will be using Mach3 as it seems to be the software of choice. If this is a bad idea, speak now or forever hold your piece.



Sorry for the novel, and thanks in advance to all those who are willing to help!

~Walter~

[ger21]

Look here
PMinMO.com

Or just build a HobbyCNC board. HobbyCNC | CNC for the desktop machinist and hobbyist.

[doorknob]

Secondly, while pulling my old Epson Stylus C60 apart, I wondered why I couldn't use the existing 'chip' and connections and simply mount the motors where I see fit and control it as though it were any other board. Is this possible, if so how? If not, why? Details please!

Theoretically you could do exactly that.

Practically, there may be some pitfalls that could undermine that approach.

Here are a few to be aware of:

- You may need to do a lot of reverse engineering to figure out how to drive the existing stepper drive electronics, only to run into a brick wall. For example, it is possible that a consumer electronic device would use custom circuitry for which you would not be able to obtain published data. The stepper driver circuitry could possibly use one or more custom and proprietary ASICs (application specific ICs) that do not have externally-accessible pins that you would need to access to drive the steppers from your PC, or that would require inaccessible microcode to initialize and operate, or may need unusual voltages, or that may have built-in diagnostics and fail-safe routines that must be satisfied before the device can perform its initialization, or any number of other insurmountable hurdles. Of course it is possible that everything that the printer's stepper drivers are doing uses standard, off-the-shelf, and well-documented parts, in which case you're golden - but you can't count on that going in.

- The printer's internal circuitry might possibly use interconnects or surface-mount electronic devices or complex, multi-layer circuit boards that are optimized for high-volume, automated machine assembly, but which are difficult to rework manually, even if you have access to specialized equipment such as a hot-air soldering station.

- Some of the existing driver circuitry is probably optimized to handle the leadscrew geometry or belt drive needed for precisely driving a print-head, while other existing driver circuitry is probably optimized for driving a coarser paper feed mechanism. It's possible that neither of those will be properly adapted to drive the leadscrews that you will be using, with the number of steps per rotation and torque that you will want for your test bed.

Etc.

So, it might work, or it might not.

It's an interesting idea, though, so good luck with it...

[Walter23]

Gerry,

Thanks for the links, but what I'm looking for is the building experience and customization ability with my board. I want to start simple, but be able to build off the fundamentals. This is why I am choosing not to go with a pre-arranged kit. I know it's basically the same thing building it from individual pieces, than assembling a kit, but I already have most of the components(I think), plus this way I understand what I'm doing-the most valuable part!

Doorknob,

Thanks for your reply. That makes complete sense, I just figured maybe I'd luck out and have a usable bored that just needed to be configured in Mach3. After posting I started tinkering around in the program and testing my ports. They all test good, but when hooked up to the printer no magical axis movement.

On the upside! I get two stepper motors, some gear reducers, shafts, pulleys, parallel port with pins, and a parallel cable out of it!

Now if only I could get a jump start on this whole driver build...

[doorknob]

Also, two more questions, I assume I need a 'chip' of some sort, and I stumbled across this: L293D Motor Driver IC (SKU: L293D) - HVW Technologies

Is this any good? From the little I comprehended about building a stepper driver I think I need a 'chip' of some sort for each axis. This would do unipolar or bipolar if I'm not mistaken and it seems affordable and capable. But I have no idea what I'm talking about when it comes to these things...any thoughts?

Based on a quick look at the data sheet for the L293D, I suspect that it is not an ideal "chip" for you to use in this application. While it could possibly be used in this application, it has only very basic motor-driving capabilities, and so it would require additional components and complexity to adapt it for your use.

For example, when Mach 3 drives a parallel printer port, it will output step and direction signals on the port pins. The L293D does not have the built-in capability to translate step and direction signals into the sequence of motor signals required to drive a stepper motor in response to those signals. Rather, it appears that the L293D only performs the function of taking logic-level signals and translating them into the higher voltage and current signals needed to energize the coils of a motor. So, you would need to add logic circuitry between the parallel port and the L293D to handle that translation. Such logic could be handled by a simple microcontroller such as a PIC. There are a number of published PIC stepper driver designs on the net, and so perhaps one of those could be used with the L293D.

Another problem with the L293D is that it does not appear that it has been designed to handle microstepping output voltages. So, your driver would probably be limited to full-stepping, or possibly also half-stepping, but would not do 1/4 steps or 1/8 steps or similar. Whether or not that would be OK for your application is something that you would need to decide.

[Walter23]

Doorknob,

Thanks for another helpful reply! I guess I'll stray away from that one. I've been doing constant reading in between tinkering and waiting for replies to my post. I just stumbled on PMinMO.com Stepper Driver Information which is on the site Gerry referred me to. It's helping me to understand things a little better.

I think I should be able to figure things out on my own in the long run with enough reading and research, and of course the good ol' trial and error. I was just hoping someone out there would look at what I find difficult and be able to set me straight. It may be one of those things that isn't just cut and dry, but it doesn't hurt to ask!

Feel free to suggest any components, designs, or advice you have.

As a side note, the stepper motors from the printer are 4-wire meaning they are bi-polar...which is what I wanted in the end anyway. Not sure how powerful they are though. Is there a good way to check the volt/amp of motors that aren't marked?

[doorknob]

As a side note, the stepper motors from the printer are 4-wire meaning they are bi-polar...which is what I wanted in the end anyway. Not sure how powerful they are though. Is there a good way to check the volt/amp of motors that aren't marked?

I don't know offhand how you would check that.

If your printer were still functional, then you could probably run some printing tests while using an oscilloscope to probe the motor voltage and current used by Epson in their design. If you have already pulled apart the printer, it may still be possible for you to do some circuit analysis or probing to get a rough idea of the specs.

[Walter23]

Ya, the printer's pretty pulled apart. It's just down to the motors and logic board now. I'll just have to do some research...but perhaps tomorrow.

[wantsout]

Take it from one that has built his own cnc from ground up as they say, If I had it to do over I would have built all the mechanal parts and got that right. and as cheaply as I could, go with the NEMA 34 stepper motors at least. and use some off the shelf electronics,because you will be making chips fly a lot faster if you do. I know the build is just as exiting to do but the best part is when you get it running. make that time as short as possible so you can enjoy the progress. I just got my three axis moving today so will try to make chips this comming week. for me that is my goal.

[Walter23]

I'm starting to think the same thing...

The mechanical build phase will be exciting, no doubt. I just wanted to feel a little more pride by building the circuitry as well. But I guess since I'm not winding my own motors I can live with someone else building my board too.

I do however want to order the motors and driver board and get comfortable with all that before even starting the axis build.

Recommendations for boards and/or motors? Amps, volts, model numbers, etc.

I read on some forums that bipolar is the only way to go for the necessary precision/stability with harder materials. But most driver boards seem to only work with unipolar motors. Someone want to set me straight on this?

[ger21]

I read on some forums that bipolar is the only way to go for the necessary precision/stability with harder materials.

Unipolar/Biploar will both perform exactly the same with regards to precision.
The only difference, really, is that if you have an 8 wire motor, you'll get more torque with a biploar drive. (about 30% more)
Actually, most of the higher performance drives are bipolar.

It's almost impossible to specify motors and drives without knowing what your building, and what your performance goals are, as well as the screws you'll be using.

[Fastest1]

Plenty of people can and will help you out. You are on the correct forum. You might want to post in the electronics forum here too as there are some pretty knowledgable people there. Regarding the unipolar bipolar battle, use what is available til later. As long as the controller is for your type of stepper it should be alright. And regarding stepper size, I am running 23's on 4 different machines. All can outpower the tool the spindle is running, size is not always the answer. I vote for the Hobby CncPro kit myself as your first project. I did it that way, loved the results and now that I run a G540 on 2 machines, I cant really tell any difference in machine operations. The G540 was much easier to wire up and get running but by the time I was using the G540, I understood so much more about set up and Mach. Anyway, enjoy the build and the pursuit of knowledge. You have chosen a project with many different challenges, it should keep you busy for a while :-)

Doorknob,

Thanks for another helpful reply! I guess I'll stray away from that one. I've been doing constant reading in between tinkering and waiting for replies to my post. I just stumbled on PMinMO.com Stepper Driver Information which is on the site Gerry referred me to. It's helping me to understand things a little better.

I think I should be able to figure things out on my own in the long run with enough reading and research, and of course the good ol' trial and error. I was just hoping someone out there would look at what I find difficult and be able to set me straight. It may be one of those things that isn't just cut and dry, but it doesn't hurt to ask!

Feel free to suggest any components, designs, or advice you have.

As a side note, the stepper motors from the printer are 4-wire meaning they are bi-polar...which is what I wanted in the end anyway. Not sure how powerful they are though. Is there a good way to check the volt/amp of motors that aren't marked?