Is this a dumb idea?

I have been looking around at various prototyping and manufacturing equipment and they all seem to have some things in common. A number of stepper motors are driving a cutter/printer/extruder head of some kind. I wondered why we have to put a motherboard and stepper motor controlers on each machine.

If there was a Mendel in my office, a 3 axis plasma cutter and a 3 axis mill in my garage and possibly a powder printer as well, why couldn't a single set of control electronics be carried around and plugged into each of them? the software would have to be able to handle various different configurations but that doesn't seem insurmountable. It would also have to have connections for various accessories, from bed heaters to oil pumps for cooling cutter heads for the different machines

Not being an engineer I don't know the issues involved in creating stepper drivers that could handle widely varying motor sizes. On first glance it doesn't seem that this would be impossible and might have some advantages for someone with multiple machines.

Those with more knowledge may now feel free to mock me for my ignorance. :-)

> Not being an engineer I don't know the issues
> involved in creating stepper drivers that could
> handle widely varying motor sizes. On first
> glance it doesn't seem that this would be
> impossible and might have some advantages for
> someone with multiple machines.

Even if expensive to have all that stepper motors, it makes possible to have it working, because if we do as you suggest, we would have such a complex system that would not work for sure.

Like if you have at your kitchen a machine that would do everything instead of many you have. If that only one breaks, you would be in problems... but now if just one breaks, you can keep working with others.

For RepRap, we need modularity, which implies that like using many times the same pieces, like Arduino. Like now there are 3 arduinos on RepRap: 1º arduino on main board + 2º ardunio on extruder board + 3º arduino on heated bead.
And as you can see, heated bed was the last one to be developed, if we were waiting for finish all the system just to use one arduino, we wouldn't even have the actual Mendel

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

... it's mainly the differences in needed torques, dynamics, speed and accuracy, what's defining the usability of a machine.

For using one set of controller and motor-drivers for all different 3-axis-machines in your hose, you would have to implement a complete set of parameters and voltage/current-switches for every different setup, or if you connect the unit to a machine and use the wrong configuration, it could destroy either the motors or the electronics.

And in sum all the different controllers and drivers configured to the specific machines will be cheaper than an 'omnipotent' controller with all the specific configs and hardware-defines ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Only recently active here, I have been using NC for over 30 years, and have owned my CNC for at least 15.

Much has changed in the last 7 or 8 years relating to motor control. This was driven [no pun intended] by the robot war competitions, which have become a popular school activity, replacing what had been the radio electronics club.

A universal controller as you suggest would need to know what step rates each of the motors in a given machine needed. I have been going through my collection of recycled steppers. At best I can find from the manufactures website is a general idea of what the ratings are. In the case of a NEMA17 I pulled from a printer or scanner, The data sheet lists about 15 different torque requirements. My motor was a custom one for the scanner MFG, and the best I can do is to see how close the measured parameters are to the spec parameters.

Currently my CNC, runs at 10 ipm. The motor controller must be able to handle 3.5 amps and the backwash currents, which could be 100s of times this. A reprap needs to move at more like 30 or 40 ipm. Here the motor current may be only One Amp. Some developers here would even like to achieve 100 ipm. Steppers max out at 1000 RPM, so this pretty much limit where they can be used.

Currently the cost of the motor controllers is less than the single quantity parts cost. I was looking into upgrading my CNC using discreet parts, to do microstepping. I can purchase a Gekco board, for less than the parts. This is because the sales volume is high enough that what the marketers call Non Recoverable Engineering [NRE] cost have been spread across the whole project. These cost include power bills and labor rates.

The electronic items may seem to be a gating factor. Especially to someone still in school. When an accountant looks at the system, the parts cost are so low as to almost seem free. This is why I am not an accountant or a lawyer.

-julie

I was wondering why we don't stick with the NEAM 23 we used for the darwin and use shaft couples.

That way you could move the three axis motors to another machine, mendel for reprap, McWrite for light milling PCB routing, etc.

You definitely could do that, fred, but it means that each machine is significantly less convenient and useful. It also means that you can't multi-task by leaving, say, your printer running while you mill those PCBs. When you consider how long the build times are for these machines, that's a major concern. If you get a printer working reliably (to the point that you can leave it running unobserved for 7-10 hours), on the other hand, you can make and sell mendel parts and make back the investment you put into the dedicated electronics and motors.

This was my initial reaction to what seemed like a wasteful ammount of redundancy.

As I have learned more about stepper motors, drivers, and controllers, I now feel less strongly about the topic.

To an extent: such a thing already exists via the "NEMA" standardized footprints.


Why does the Mendel have it's own custom frame when legos would make a more universal building tool? Why use lego's when an iron casting foundry would produce frames of superior performance?

The lack of uniformity is a compromise between waste of redundancy and waste in performance(under or over) for the application.

Which isn't to say that there isn't a market for such a thing. Personally: I don't think any of the reasons given prohibit such a thing from being possible or even viable. You trade wasted initial capital investment for wasted expense liabilities in terms of engineering compromises.

By forcing round pegs through square holes for interoperability reasons you end up tying up expensive CNC mill servo's and geckodrives on a cheap 3d printer. Better to just find enough full time work for both to justify their own organs no?

Or alternatively: outsource.

Why does every highschool in a state need their own robotics team & support logitistics: because a shared lab + equipment results in lost productivity and travel expenses which don't justify the savings.

All that said: there are a lot of things that can be done with a set of NEMA 17s and many of them are never in use at the same time.

There may be a rational market for such a thing, and there is certainly an irrational market for such a thing(people who prefer convienence over the right tool for the job).

At the end of the day: the know how to drive several stepper motors is the most important thing, because with that comes the ability to make a situation specific judgement call on when it's suitible to share support circuitry/stepper motors.