Integrated Light-Duty Positioning Solutions

Does anyone know of a good supplier (preferably made in USA) of fully integrated, light duty positioning actuators?

Here's what I mean by integrated:
All electronic and software components are included in one package. I can do the mechanical side, but the electronics, amplifiers, encoders, shields, wiring, PLC's, networking, and programming are out of my area of expertise. I can design whatever mounts, bearings, ball screws, bushings, washers, screws, etc that I need. But on the electrical side, I'd like to have only 2 plugs: one that goes to a wall outlet for power, and the other that goes either to the Ethernet port of a PC or to a manual controller (knob or push-button).

Here's what I mean by light-duty:
I want to be able to drive adjustment knobs for small machines. One application is positioning a work piece (maybe 10 lbs at the most) under an inspection camera. Another application is positioning a strip of protective tape as it is reeled up, so that it doesn't bind. Think applications that you could do with your hand, and a knob.

One other stipulation:
Though I'm tempted to just use an Arduino board and put it together the hobbyist way, I am a little fearful that such a setup wouldn't last in an industrial environment. If anyone out there thinks it would, what's your setup?

It seems like everywhere I look, all I find is piecemeal solutions. You have to buy the actuator, then the controller, then all the cords, then a software package that is right for it, etc, etc. For high-load applications, I would hire a consultant, or maybe even invest in the time to learn how to put it all together. But for light-duty applications, it often turns into a manual knob instead of an automated system... just because the headache isn't worth the time. But if I had an integrated system at a reasonable cost, that would be well worth it.

There are several companies that offer these type of units. IAI one of them. (eye a eye)

The problem with comparing to a hand turning a knob is that you tend to forget all the feedback you get from the hand like force and speed, along with very easy and versatile mechanical coupling! You are also probably forgetting start and stop signals, jam sensing visual detection, smell detection and audible alarm etc. provided by the controller attached to the hand.

I also find it interesting when someone says "Just use an Arduino" Amazing what people can do with them when labor hours don't count, and you need to know electronics, programming, etc.

However you can go to companies like AutomationDirect and get a package of steppers, servos or closed loop VFD controls along with power supplies, premade cables, inexpensive UL listed, reliable PLCs with free programming software and so on. Your first project will be tough, and it will get easier from there on out. I am currently designing a PLC control panel that will be very much standardized across a broad range of products. It will be mated to motor control panels of various types. Since I started completely from scratch the first one is taking a lot of time. However subsequent versions will be a piece of cake.

I guess this helps:
http://www.automationdirect.com/static/specs/sureservolow.pdf

Looks like the cheapest they get is around $1000.

Although that doesn't come with a PLC. Can someone decipher the above link - is a PLC required to drive any servo, or is it just for advanced systems? If a PLC is forergone, does the "drive" control multiple servos or just one?

Your description is not clear nor sufficient. Perhaps you can add sketches. My first look at your post made me think you wanted a source of torque, and delivered in some sort of miracle package with "two plugs". Unlikely you will find that. To "turn a knob" I would suggest something similar to those microstepping motors with integrated controllers found at Schneider Electric Motion USA . Cell control for simplistic operations can be performed with any number of cheapo "micro-PLC" units on the market (I'm doing a project with the IDEC SmartAxis Touch unit at the moment & love it...but Unitronics makes some nice controllers, too). But you won't get away from having to know some basic electronics for the necessary interfacing you must do. If you need some sort of "easy to plug together linear motion device" then I'd recommend the Yamaha Robotics line: great stuff, customizable, easy to work with, not cheap.

CutEdge said:

Does anyone know of a good supplier (preferably made in USA) of fully integrated, light duty positioning actuators?

Here's what I mean by integrated:
All electronic and software components are included in one package. I can do the mechanical side, but the electronics, amplifiers, encoders, shields, wiring, PLC's, networking, and programming are out of my area of expertise. I can design whatever mounts, bearings, ball screws, bushings, washers, screws, etc that I need. But on the electrical side, I'd like to have only 2 plugs: one that goes to a wall outlet for power, and the other that goes either to the Ethernet port of a PC or to a manual controller (knob or push-button).

Here's what I mean by light-duty:
I want to be able to drive adjustment knobs for small machines. One application is positioning a work piece (maybe 10 lbs at the most) under an inspection camera. Another application is positioning a strip of protective tape as it is reeled up, so that it doesn't bind. Think applications that you could do with your hand, and a knob.

One other stipulation:
Though I'm tempted to just use an Arduino board and put it together the hobbyist way, I am a little fearful that such a setup wouldn't last in an industrial environment. If anyone out there thinks it would, what's your setup?

It seems like everywhere I look, all I find is piecemeal solutions. You have to buy the actuator, then the controller, then all the cords, then a software package that is right for it, etc, etc. For high-load applications, I would hire a consultant, or maybe even invest in the time to learn how to put it all together. But for light-duty applications, it often turns into a manual knob instead of an automated system... just because the headache isn't worth the time. But if I had an integrated system at a reasonable cost, that would be well worth it.

Click to expand...

For what you're talking about, "light-duty" versus "high-load" really won't make much of a difference in price, at least in my experience. (My experience being that the cost of designing/assembling the system is a much larger factor than the forces involved.)

You say "I can design whatever mounts, bearings, ball screws, bushings, washers, screws, etc that I need" - why not just go buy all of that already together, assembled and ready to go from someone?

That's essentially the same question you're asking here on the Controls side, and the answer is the same as well.

Tony's suggestion is probably your best bet, but an actuator is always going to be of somewhat limited use without peripheral devices to tell you what's actually going on....and now you're back to needing a PLC (and everything else from your original question, really.)

One other thought on this.........

Use your APPLICATION guys for what they are getting paid to do. You don't have to figure out yourself what all goes together. That is what your sales application engineering guys are for. Make them earn that money. You are wasting your precious time trying to do their job. You just need to supply them with the correct application information.

Thanks for the responses, everyone. It has been helpful.

garyhlucas said:

I also find it interesting when someone says "Just use an Arduino" Amazing what people can do with them when labor hours don't count, and you need to know electronics, programming, etc.

Click to expand...

No matter what solution you go with, you'll need to have skill in something. I laugh at some of the Arduino projects that are claimed to be "cheap" because no one is accounting for the time spent. That said, many of the projects Arduinos are good for can be done in a few minutes if you're already a skilled programmer. In fact, for most things, I could put together an Arduino solution much faster than using an off the shelf PLC or other module because I already understand C++, Arduino/AVR, have hardware design skills, wiring tools, etc...

Is it always a good idea? Definitely not, but don't automatically discount them either. I got a pile of Arduino ProMini's here that cost me far less than I'd have to pay for the parts to build them myself. Not to mention the hours of PC board layout and review, etc it would cost me. At those prices, a lot of simple automation that would otherwise be too expensive becomes feasible.

Festo has a couple of lines of positioners.
Light to heavy duty.
Electrical positioning systems | Festo

These threads usually diverge, like this one, to 2 extremes, neither of which is useful/correct and or approaches what was the goal.

First line:
Festo, automation direct, et al... good solutions, well proven.
Minimum units costs 1000$ and up, per piece, and you usually need more than one piece.

Second line:
Something like "and it only cost me 35$ with Arduino/Beaglebone/linux/xxx" or whatever.
1. Totally ignores power supplies, wiring, cabling, docs, proven reliability, and mulstiple drawbacks due to lack of standards, docs, experience, stability of platform in hw and sw, etc etc etc.
2. Totally ignores huge effort in work- hours to actually get something documented/tested.

Similar things happen in PC-based CNC retrofits, where usually everyone talks past each other, ignores realities, and is not too experienced/competent and only puffs up whatever gizmo they have last used.

The A, to a very good question, imho.

I dont know of anything useful and cheap - but something may exist.

It is Very Certain that this type of stuff will be available dirt cheap Real Soon Know.

A huge nr of companies are coming up with all sorts of thingies and plugins, and the whole maker/arduino/etc stuff will grow up, due to extra interest/effort etc.
Cheap PLC also get easier, better and cheaper all the time.

I could do something like this very easily - from a PC based system, but only because I have great PC/IT experience, coming from the SW and IT side of things.
Its "a solution", not The solution, and has the drawbackst that system will be non-standard, non-documented, and non-provable. Ie you cannot "prove" tha the system is reliable or deterministic.
This may or may not be a drawback.

Cost (price point) and value (what you get, fexilibilty, IO etc) and quantities are very much drivers.

A single system: PC based because its easy, fast, and cheap. PCs cost (relatively) nothing.
Unless you need to prove it will work (ie functioning system is not be enough).

10-100 systems:
I would look at cheap PLCs and buy some learning kits. Definitely wont get the best results on the first try, but maybe 2-3 iterations and you would have a cheap, easily-reproduced system that is easy to maintain and verify.
I would expect to spend 2000$ in learning, and unit costs 500 or less, maybe as low as 200$.

HIgh-value systems (industrial plants, etc):
Would go with established system for longevity, liability and long-term support.
Most expensive (in support, purchase and learning and work hours). Slowest to do. Hardest. Poor bang-for-buck.

Example:
I built my first low-unit-cost robots with integrated electronics/sw/features circa 1995.
Some ran for 50.000 24x7 cycles with zero glitches (custom phone-answering app, with menus and stuff).
Unit cost was about 100$/pc, for 7-10 iirc.
(Boot via IPX to diskless motherboard, running dos, and one custom built hw board and sound card used as voice prompt system).

Another example:
I will likely make a robotic auto-tool-loader.
I will use a PC sw, g-code, and low-cost hw generator with lots of IO (20+ inputs).
Why?
Unit cost under 350$ (includes PC hw), very reliable, very fast and easy to do (for me, with over 3000 hours experience with these sw/and hw combos).
A PLC could do the same, but I would need to learn all the PLC issues/problems/limitations first.
Programming the thing takes less than a day, for me, in this particular case.

This is for a closed system - idiot type (A-in, B out).
A generic system is much harder to do.