Stepper Motor Selection

A stepping motor will require a stepping motor driver. While it is not really necessary to use the same driver manufacturer as the motor manufacturer, when first working with a stepping motor obtaining the driver designed to work with the motor is a good idea. The major advantage I've found in using a stepping motor is that programming control in an open loop is often a straight forward procedure. Now you will have to choose a dumb driver that just takes simple commands like step and direction or quadrature signals to move your table or you could go with a more complicated networked controller that must be programmed with all of the network parameters along with all of the motion control parameters like acceleration/deceleration curves along with top velocity and distance to travel.

Now for what size motor you will need will depend on a collection of kinematic calculations and desired parameters. The moment of inertia will be dramatically different if your 3 kg load is one beaker centered on the table or if that is 3 kg is at the edge of the table. I would assume the hollow cylinder moment calculation for the beakers and the solid cylinder calculation for the platter mass will be sufficient for you to determine how much torque will be required to spin this platter loaded. But naturally you will have to consider how to couple the motor to the table and how much of a mechanical advantage or speed this coupling will provide.

Now only from my personal experience, I would be surprised if you needed anything larger than a NEMA23 size motor to move a fully loaded table. But if you never wanted the table to move very fast and you have a included a lot of gear reduction you may be able to do this with a small NEMA 8.

Thanks redfred...That gave me a decent idea on how to approach this.. I was planning on controlling the stepper from an 8051 and the driver I had in mind was ULN 2003 IC. It's simply a current booster and i'll have to control the stepper motor(i.e deciding no. of steps and direction) mainly from the 8051 pins.. So this is achievable right?

No. Likely the 8051 microprocessor will be more than capable to work as the controlling logic. I expect that this will be capable of doing micro-stepping as others here have suggested. Atmel has some very good application notes on stepping motor control but I'm not sure about that specific platform. But I certainly would not use a that driver package.

A ULN 2003 IC is a seven gate Darlington transistor array that has for a maximum current per gate of only a half ampere. Now I won't say that this cannot run a small motor, but with kilograms of mass to move you will likely wish to be able to sink and source several amperes of current along with turning OFF individual channels. All of these modes a ULN 2003 can do but not on an individual channel basis. Also with the relatively high voltage drop across two collectors to emitters that is a Darlington, this array will dissipate some significant power. If I was designing a stepping motor driver from scratch, I would consider instead using discrete power MOSFETs. This can be easily designed to be directly controlled by the I/O of any micro-controller.

Thanks again redfred... I just want to make sure that I have heard you right: Are you saying that, 8051 may not be the right processor for this app and that there are better versions from ATMEL? And coming to the driver part, I am running short of time so I doubt I'll be able to make a driver of my own at this stage... So if i were to buy a driver, let's say the one which comes with NEMA 23 size stepper, would it cost an awful lot... If so is there ANY cheap and quick alternative.... Thanks a lot for your time..

Well the ATMEL link I provided references a different processor. It also covers a very little bit of stepping motor fundamentals, ATMEL has a better application on stepping motor fundamentals . (The underlined text is a hyperlink to another web page )

As far as a quick and cheap alternative, I think Andy Germany offered a reasonable idea of E-Bay. But you remind me of a sign I saw at a local machine shop. I don't remember the exact wording but it went something like this...

This shop can fabricate any part to these standards:

• The highest quality precision part fabrication available by our equipment.
• The fastest part fabrication turn around available.
• The lowest price available per part.

The customer can expect no more than two of the above standards at any one order.

You get what you pay for.

Thanks redfred!!! That link was a good read... As for the motor purchase, I guess I'll consider ebay once I decide on what to buy..Unfortunately that shop you mentioned is a little out of reach for me

If I understand correctly, you are planning on mounting your aluminum plate directly on the stepper motor shaft. If you do, then I wouldn't try anything smaller than a size 23 (2.3" diameter), and you will definitely want a microstepping driver. Standard stepper motors have 200 steps per revolution, or 1.8° per step. Without microstepping, the motor is going to try to jerk 1.8°, and the inertia of your disk plus load is not going to allow that rapid motion. If it did, then the beakers would probably slide on the tray. Good microstepping drivers can divide each full step into many little steps (64 and 128 are common numbers), vastly reducing the jerk.

In my opinion, a better solution would be to mount the plate on a separate shaft, with top and bottom bearings plus a thrust bearing to support the weight. Stepper motors do not normally include a thrust bearing. Mount a worm gear on the separate shaft, and a worm on the stepper motor (or on a shaft connected to the stepper). The speed reduction of the worm gear will effectively produce the same elimination of jerk, but it will greatly reduce the torque required of the stepper motor. Thus you have eliminated all major forces on the stepper motor bearings, and it will last muuuch longer. This of course means that you could use a smaller stepper motor, but I'm partial to size 23.

If you have a limited budget, some time, and access to somebody's junque pile, many old laser printers and larger dot-matrix printers had size 23 steppers in them.

Thanks for your input dkwarner.... I was in fact planning on using a separate shaft to couple the motor and the plate. But i didnt realize it'll involve all the components you have mentioned and quite frankly, I am not sure how to go about it now.. I am an ec student and dont have much idea on the type and use of bearings.. So can you please suggest a way I can approach this? I have a pic of the rotary plate design but am not sure how I can upload the file.. the formatting tags allow only webpages and I dont know how I can upload my image.. thanks for your time

Here's what I had in mind:

I've shown a standard thrust bearing. Another, probably better, possibility would be to use a lazy-susan type. I did not show the bearing at the end of the worm shaft, but of course there must be one, to hold the worm in contact with the worm gear. I have a long shaft on the motor for two reasons: 1) It reduces the forces on the stepper motor bearings. 2) It allows part of the motor to extend above the table, which will allow the table to be lower than I've shown, resulting in a more stable overall machine.

What does 'ec' stand for? economics?

It has been almost 30 years since I last made my own stepper driver, so you can get better advice on that from redfred and others.

My most recent motion control projects have used Galil motion controllers and Applied Motion Products stepper drivers. I like Galil because of its easily understood programming.

Good luck on your project!

Thanks a lot for that sketch dkwarner... EC, here in India is short for Electronics and Communication.... So I havent had any experience with bearings and gears.. We have learnt about how stepper motors work and how to program them, so that made me think that I'll be able to take up this project... I will have to read up on them before i fully understand your design... Thanks again for your input!!!

There are lots of chips around specially designed to drive a stepper motor. They need to usually only be given a step (clock) and a direction signal, which can come from simple electronics (a 555 for example) or even from a printer port on an old laptop or PC.

There are even very good freeware OSs( Ubuntu) and free driver software (EMC2 for example...)

It depends exactly on what you want/need to do......

Using PICs or similar to drive the driver chips is even easier than programming them for the signals, as I said before, all you need to do is make a clock and a direction output.....

You need to detail us your requirements and how much/little you need to invest mx....in fact, just tell us everything we will all do our best to help you further.....

Thanks Andy Germany... My application is to build a rotary table on which i intend to place beakers which are to be rotated... The basic idea i had in mind was to use a stepper motor driven by a driver ( initially i tot i could use ULN2003 IC but I found out from this thread that it wont suffice, so am not sure which one to use) and controlled by 8051.. Coming to my investing limit, I was looking at something around Rs.500...But I guess the stepper motor I need would itself cost more than that.. So I am hoping the driver and the rest of the components are less expensive and am hoping I can get some funds from my college to be able to buy the motor.. Hope that covers the info you were looking for ,let me know if you need any more information.. Thanks again!!

I buy all my stepper motors on ebay, I don't know the exchange rate to Rs. anymore, but I rarely pay more than €10 per motor. I have sometimes bought 10 motors, of which one was damaged and paid €16 for all together.....

I have even (almost accidentally) bought a 3 axis controller, 3 small stepper motors and some sensors all for €1 and €5 Postage.......

You just have to bid low on many things that you need and not accept paying more. A good dose of Scottish blood is VERY helpful in this matter!!!

I personally would go to a scrap yard and buy an old turntable from a record player, it has usually a simple bearing that will be good enough for the loads you are talking about (often two ball bearings one on top of the other.) and a rubber disc attached to your stepper motor to drive it around, or even more accurately a toothed belt system......

A small motor will be enough if geared down.....

There are so many ways to do all this and cheaply......you should have enough for a bottle of champagne as well!!

If you need some stepper driver chip numbers, just let me know.....

Thanks for the tips Andy Germany..... Look forward to using them once I decide on what to buy.. I will ping you if i'm stuck on the stepper driver chip numbers...I have given an overview of my project below.. Do have a look..Thanks again..

I read it, sounds fine.....

I will wait for your questions rather than bamboozle you with further ideas.....

OK. At my school (50 years ago) Electronics automatically included Communication (as opposed to Electrical, which concentrated on power generation, distribution, and utilization.

We need to know your principal objective. if your main objective is to get experience with steppers and drivers (a very worthy objective), then let Andy help you. If your principle objective is to build the beaker turntable and do some motion control, then it would probably be faster to use a DC motor. Add a pin at each beaker location, and one or more appropriately located microswitches ( or magnets and hall effect sensors) to tell the controller when the table is approaching the stop position for a beaker. One switch would suffice, but others could be used to ramp up speed at the start and ramp down at the stop (to avoid slippage of the beakers or splashing of the contents). One disadvantage of the DC motor is that it has no holding power when stopped, but that problem is eliminated by the worm gear, if it has any reasonable ratio (10:1 or greater).

Thanks again dkwarner.. Your last idea has intrigued me and am now approaching this in a whole new way!!!

Anyway to answer your question: Yes I am indeed focussed on completing this project of mine foremost. I am in my final year of engineering and this project is a part of my curriculum.

A brief overview of my project: There is this experiment to deposit thin nano films on glass substrates by a process called self deposition. It is a very simple process and the experiment as such involves dipping a glass substrate in six different beakers( for a specific period) over and over again. Now for one round/iteration of dipping in all six beakers takes 40 mins.. typically for good results around 7 to 8 iterations are needed so the total time for this experiment is around 280-320 mins( around 4.5-5 hrs) At present this is done manually, so it requires someone to be around to replace the substrate in different beakers.. So I am planning on automating this process, so that one just has to start the experiment and return after 6 hrs at the end of the experiment.. Not only does this save his time but it also removes any possibility of human error..

So with this in mind, I planned on having a disc on which I could place the beakers and they would come under a substrate which would then be dipped into it for the required time.. it's then lifted; motor rotates and the next one comes in line and it's dipped into that and this goes on..

So this is my project.. Like I said I'm intrigued by this idea of using pins and micro switches, can you please elaborate a bit more on them..

Thanks a lot for your time!!!

qpoint,

You're starting to bring up the curmudgeon in me. Many people here, myself included, are very happy to assist people doing a project for nearly any reason. But I dislike giving away my professional services for free, particularly when somebody else may take credit for it.

Let me get directly to my point, this is supposed to be your senior project not ours. You will be getting credit for this effort, not us. Now this can be done with a stepping motor, a brushed or brushless DC motor, or even a three phase induction motor. All approaches will have some drawbacks and some advantages. Some will be easy to execute but difficult or expensive to fabricate. Some exactly the reverse. All

When I direct you to some easily obtained free application notes on how to use a micro-controller to control a stepping motor, you apparently do not even look at them. Now dkwarner has offered a simplified component selection of a DC motor and micro switches to which you respond for more details in this design. Inspiring you with a design approach is one thing, handing over a complete tested kit to assemble is something all together different.

Now that I've vented a little, I do have a strong hopefully helpful suggestion for you. Sketch up for yourself a list of the established design goals in this project. Include in this sketch the firm established physical dimensions you must meet; size of table, maximum rate of rotation, maximum and minimum mass of table and load, minimum precision of motion. Now add the extra goals you believe you can easily achieve. This should take you no more than half a day to formalize and list your goals. Now for the second half, sketch up two or three approaches on how to meet these goals and try to list the assumptions you've made. List your perceived advantages and drawbacks of each of these approaches. Now look for suggestions from outside on your approaches. Next step (day), fabricate a test model of your machine and do some mechanical testing to verify your assumed mechanic values of table speed and torque values required. This will either validate your design or reveal the flaws in the design before final testing. Now you will have a realistic idea what your task will entail so you can either order the precise components required or test the parts you readily have to see if they're suitable.

I hope that I've helped you here, and I do wish to help you do you're own design.

Redfred,

I'm sorry I upset you. I obviously didnt want that. I also didnt expect you all to come up with a design for my project.. I am sorry if I came across as trying to do that.. I was just trying to get a clear idea of what dkwarner was trying to say but I realize now I should have given it more thought before asking, which is what I'll set out to do now...

I thank you for setting up a framework for me which I shall implement and as you said I will get back when I have doubts at any stage

As for those links, I dont know what gave you the impression that I didnt even look at them, but to be honest with you, I did give it a read.. I may not have STUDIED the whole document but I did go through it...

Anyway, I really appreciate the help you and the rest here have given me so far and hope I can expect the same soon enough..

Thanks again

qpoint,

First, you didn't upset me any where near as much as the problem I was working on at home at the same time. My blog writings were an attempt to clear my clouded mind but it meant that I had a very short fuse at the time. This lead to my premature rant against you. While my rant did help me relieve some frustrations, you did not warrant it. I apologize and thank you for the pressure relief.

You do have a good grasp of what is needed to design a system and more important a recognition of your own limitations. You will make an excellent engineer. Whenever you need a clarification on this or any other project you are working on I'll gladly help if I can.

I'll make up a couple of sketches and send them on later today.

GA for a good well documented post. I have seldom seen such a well documented idea on CR4.....

Thanks Andy!

Even though I have steered qpoint away from steppers (because he indicated time was getting short), I still like steppers. My first stepper project was a device to rotate a loudspeaker 360° while measuring the sound level, to produce a polar graph of the speaker output. This was a project built (with my help) by one of my physics students about 1985, and we used a Mac 512k with no hard drive as the controller. No hard drive meant that we could use 4 of the output lines on the SCSI port to drive transistors to power the motor. All the control software was written in MS BASIC. Very educational!

As I said in an earlier post, in recent years I've used commercial microstepping drivers, but I just might find time to try building one myself. Is there a chip set that can handle 2 or 3 Amps at 24V input (with appropriate heat sinking)?

Thanks

Dick

The chip I would recommend is the Allegro SLA7078MPR......

Steps from single to 1/16 in microsteps......very easy to use, up to 3 amps (adjustable), at up to 46 volts......

No external transistors, or resistors needed, which is a big plus when designing......and takes away certain possible problems that many other chips have.....

Its available singly from a Hong Kong dealer on ebay at around $18 each + postage, or other suppliers in greater numbers at a lower unit price.

There are several PCBs around on the internet if you need one, or design one yourself.

If you send me a private email address I will send all the infos that I have collected.

If you use 4 amp motors, they will probably not even get hot......but its easy to adjust for smaller motors as well via a poti and a reference voltage.

There is also an input which if held high, disables the output, a sort of eStop.

dkwarner,

I can't thank you enough for that... I was blown away when I read it.. I really appreciate the amount of effort you have put in for this.. Thank you so much.... I will definitely keep you guys updated on this one... Thanks again..

You're welcome! Please read my response to redfred.

I forgot to mention that I changed from microswitches to Hall switches because it is much easier to protect them from spilled liquids.

I obviously know nothing about what facilities you have available for building the project, but I strongly encourage you to do as much as possible of the mechanical yourself, or at least observe as much as possible of the machining processes. When I was studying Electronic Engineering, I was required to take machining classes, and I have been eternally grateful for that.

By the way, your English is excellent! Where did you learn it? I would never have guessed you were from India...

Dick

Thanks for the complement ... I have my uncle to thank for that.. He's been a good influence on that aspect..

I will surely try and be involved in as much of the machining process as possible..It's been a real pleasure interacting with you...

I was inspired by your response to redfred, as I too am interested in interdisciplinary fields.. I am keen on MEMS(Micro Electro Mechanical Systems) which involves a combination of many disciplines including Electronics...I want to specialize in that during my Graduate studies...So it was nice to read your experiences.. I really admire your passion for your work...

Thanks again for everything.. I will update on my project..

Keshav ( and it's a 'he' btw)

Bravo dkwarner on a very well thought out and documented project. I do realize that there will be a lot of follow on work required to make a working model. But you may have done most of the heavy lifting here, dkwarner. I hate to rain on this parade, but did anyone consider what might happen now if qpoint's professor frequents CR4 and sees qpoint's final project that looks just like this? If qpoint submits a final project that is a near copy of this design approach, four years of schooling to a degree might go up in smoke. Now qpoint might successfully defend him/herself against a plagiarism panel but even a well defended mark will not look good on a scholastic record.

qpoint, I would go to the professor that will grade this final project immediately with this information and ask if this exceeds the amount of outside help that you are permitted to use. I might be over reacting but the last thing I want is for our help on this project to delay your graduation.

Good point. Thanks for the GA! I like your idea of showing it to the professor.

On the other hand, qpoint is an Electronics and Communications major, and most of what I've provided is mechanical and conceptual. I did not show a single circuit diagram, nor even a single line of motion control code, and I only mentioned the existence of the vertical motion assembly. I think there is still plenty for him (or her, if that is the case) to do to earn the degree.

There are very few people who do both mechanical and electronic construction (as I do), and I'll do a lot to encourage others to learn both fields. My ability to do both has made it so the jobs always found me - I only once went on a job search, in almost 50 years of work. I have built several machines vaguely similar to this one, where I did the machining, designed and built the electronics, including some of my own custom circuit boards, and wrote the control code. It is very satisfying to see such a device operating successfully in an industrial environment for years!

Redfred,

Thanks for your concern.... I must admit I hadnt thought of that.... But I think dkwarner's point is valid so I dont think I will have a problem with that.. But you are right, always better to take the safe route.. so i will have a word with my prof.. Thanks.

GA from me.

The OP should realise that steppers are generally not build to take axial loads, in fact vgery few "normal" motors are, of any type....

Thank you very much for the valuable information, Actually I am planning to use the stepper motor for building big tower clock ( 3 meter diameter ) ,the movement for the clock can drive two outputs one for hours hand and the other for minutes hand , and both of them depending on one motor ,the clock is 4 faces so it is 4 movements and include 4 stepper motors .but one control system and correctly one clock to be use for all.

Mohammed

jalalmuhammed57@yahoo.com