Shop OBEX P1 Docs P2 Docs Learn Events
Apple II Imagewriter stepper motors — Parallax Forums

Apple II Imagewriter stepper motors

CogburnCogburn Posts: 62
edited 2005-10-11 02:33 in General Discussion
As a high school science teacher, I sometimes benefit when "old technology" is replaced by newer stuff.· Our school·replaced all the old Apple II labs with IBM equipment·recently.· I have 10 or so·stepper motors·I·removed from Apple·II Imagewriter printers.· I tried to use·one with a·BS2 stamp controlled darlington array as described in the Stampworks manual experiment # 26.··Everything went OK until I turned up the pot and then the darlington chip go so hot it melted into my breadboard.· I·have googled the stepper but I could not find reliable information and specifications on the stepper but I think it must be a NEMA 23.· This seems to be a size rather than a particular model however and I cannot determine·the cause of the meltdown.··These are really nice steppers and I suspect fairly torquey.· Is·that a word?· Anyway, would anyone out there·have any specs on these steppers or any insight as to·why the ULN2803 got so hot?· I would love to be able to put these steppers to good use.


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Showing up to school doesn't·mean you are a student any more than crawling up in an oven means that·you are a biscuit.

Comments

  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2005-10-02 11:55
    You might try reading the resistance of the stepper's coil. That would give you a clear indication of how much power is required.

    Obviously, the lower the resistance, the more power required. E=IXr and P=IxE will give you a clear indication of why you burned up your chip. Read the specification on the ULN2803 - particularly the maximam watts that the chip can take. When it absorbs more heat via large electrical current, it melts.

    I suspect you will find that the stepper motors and the BS2 are not damaged - just the darlington array. It is a great 'safety buffer'.

    Try doing a bit of calculation and electronic engineering to determine a more appropriate set of drivers. The ULN2803 was originally designed to drive tiny coils in a dot matrix printer, not large steppers. It does work with the very tiny ones and can work as a stage to drive conventional power transistors, but it isn't Mighty Mouse.

    You can buy individual Darlington Array transistors that can handle the greater power demand directly from the Stamp and with protective diodes inside.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    G. Herzog in Taiwan
  • CogburnCogburn Posts: 62
    edited 2005-10-03 17:30
    The stepper says 3V and 3.5 ohms.· 1.8 degrees/step.· I was using·6 volts to power the stepper.· Do I need a voltage divider setup to·supply the 3V?· Also the uln2803 specs say that the array can be paralleled to supply more current.· Does this mean I can put two·outputs·from the·uln2803 to·each phase line of the stepper if needed?· If so, I suppose both·corresponding inputs to the darlington array from the stamp·would also·need to be connected together?···
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2005-10-04 17:04
    Well, 3.5 ohms is 3.5 ohms regardless of voltage. You doubled the amount of power they draw. Sometimes, a motor or coil will allow a lot of leeway for damage.

    I am not sure if using uln2803 leads in parallel will help you if you turn them all on at the same time. There is a limit for the whole package and I am pretty sure the UNL2803 handles only 500ma or 40volts - whichever comes first.

    Do the math ---> e/r=i 3volts/3.5ohms = something quite more than 500ma on just one coil.
    6volts/3.5ohms = something well over 1 amp on just one coil !!!!

    The whole package cannot handle more than 1/2 amps, internal heat buildup is too great. After that it is just smoke and fire. READ the specification sheet to verify my info. Try Google to get a copy or see if Parallax still carries them and provides the info.

    By the way, Darlington Transistors are indeed High Gain, but they combine two transistors together to do so. The result is that they are the worst cooled transistor design. You might try a solid state relay to handle the jump. You might try FETs. You might try cnventional transistors. But you really need to understand the higher Watts usually mean more heat.

    {Have you ever tried to touch a 3.5ohm 1/4 watt resistor at 3 volts? Ouch!!! And then it self destructs. I really don't know why they make such resistors. I can understand a 5 watt resistor in that size, maybe.}

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    G. Herzog in Taiwan
  • CogburnCogburn Posts: 62
    edited 2005-10-05 09:59
    Thanks for your help. I understand what you are saying.· I will look for a driver circuit·that uses·more robust components·
  • bobledouxbobledoux Posts: 187
    edited 2005-10-05 13:09
    In CNC applications its common to drive steppers with more than double their rated voltage. Series resistors are used to limit current flow. There are numerous stepper circuits, driven from computer parallel ports, found on the web. These older circuits were often driven using 74XX logic chips. The output was often based on TIP120-type power transistors. See what you can find on the Crankorgan.com "Piker" controller.
  • Mike CookMike Cook Posts: 829
    edited 2005-10-05 13:21

    Take a look at this post:

    http://forums.parallax.com/showthread.php?p=541628

    I believe that the Apple Imagewriter stepper motors are the same as the HPII & HPIII's, there is a *.pdf for the wire color and specs.

    There's also a link for a driver circuit I used to test these motors and a *.BS2 program to drive them.

    oops.. I didn't post the link for the stepper motor *.pdf there, but it can be found here:

    http://www.allsurplus.net/HP-Stepper.pdf

    Mike

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "OEM NMEA GPS Module" Now available on ebay for only $19.99

    Product web site: http://www.allsurplus.net/Axiom/

    Post Edited (Mike Cook) : 10/5/2005 1:29:35 PM GMT
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2005-10-06 13:34
    The TIP 120 rings a bell. I believe it will work nicely and even may have protection diodes built into it. You may have to drive the TIP with a 2n2222 [noparse][[/noparse]2N3906 will also work, but a bit leass durible] and a 1000 ohm resistor to base·in order to [noparse][[/noparse]1] get full power {full saturation} and [noparse][[/noparse]2] to protect the BasicStamp from excessive current demands.

    I have been thinking a lot about this discussion because it is about something that took a long time for me to learn.

    What I learned was that many of us just think that you have to learn Ohm's Law to design circuits. It is not true. Everything is concerned with driving Power, so you must take a great deal of care to size your circuits according to how many AMPs and/or how many WATTs are acceptiable.

    It is very important from a safety point of view as well as just designing a successful circuit.

    [noparse][[/noparse]The other thing I learned the hard way was to never try to charge a capacitor with AC when you want to use it like a battery. It blows up!!! If you don't know why, please ask.]

    I was a bit of a 'mad scientist' type when I was younger and it really took longer to learn things that way.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    G. Herzog in Taiwan

    Post Edited (Kramer) : 10/6/2005 2:15:35 PM GMT
  • CogburnCogburn Posts: 62
    edited 2005-10-09 16:45
    OK- I have looked around and I have found the attached circuit for use with the basic stamp 2 and TIP120 switching transistors.··I am using a 3v 3.5·ohm·unipolar stepper out of an·Apple Imagewriter printer.··Since I have·no TIP120 transistors,·and I do have a good supply of·IRF520's, could I substitute the 520's for the TIP120's?· THere is also example code included with the schematic and I am attaching that as well.· ·
    300 x 754 - 26K
    2K
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2005-10-09 17:28
    Well, you are changing from a Darlington Transitor Pair to a Power MOSFET, but it looks like an excellent choice from the documentation. If will run much cooler and much more efficently.

    These units will handle 10amps of current and much higher voltages - 100 Volts Max from Source to Drain, 20V Max on the Gate.
    Also, it has a fly-back diode built into it to protect it from the coils back EMF. They appear to be nearly indestructible in this application.

    Personally, I haven't used the MOSFETs yet. They are supposed to be a better design for digital as they switch on and off more efficiently. But, take care to not Zap them with static electricity while building you driver.

    I am not sure that you need any current limiting resistors. The Gate turns on somewhere between 2V min and 4V max. And it looks like it uses a mere 240 micro amps to drive it! These figure are very reasonable. Double check them yourself.

    If you have a copy of The Art of Electronics avaiable, you might take a peek. FETs are quite different from traditional transistors.

    ~~~~~~~~~~~~~~~~~~~~~~~~~~
    I suppose that it would not hurt to include the current limiting resistors in a trial set up. If it works fine, you are assured the protection from excessive current being drawn from your BasicStamp. Still they may be entirely unnecessary.

    But always protect your Stamp first.

    If you are really unsure, you can put an IC [noparse][[/noparse]an octal driver] between the Stamp and circuit. If anything goes wrong, you burn up the IC, not your Stamp.

    You might consider some fuses [noparse][[/noparse]one for each coil] to protect the Stepper from unforseen errors [noparse][[/noparse]like wiring two coils on or having your software turn everything on at one time].

    Regarding the voltage, I still wonder about handling the 3 volts. If you have +5 at the supply and want to drop it down, you might use a high amperage diode [noparse][[/noparse]each diode causes about 0.7 volts to be dropped, so two in series would get you down to near 3 volts. I think a 1n4004 would be heavy enough]

    I suspect the coil wire is generic and really designed for much higher voltages [noparse][[/noparse]typical copper enamel wire and may be good for over 100 volts] but heat will eventually cause it to fail if abused - especially if it is extremely fine gauge.

    As I said in the beginning, I have not worked with FETs and someone else may jump in here and give you better advice. At some point, you should try to do the final engineering yourself. After all, that is why you try these things -- to sharpen your skills.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    G. Herzog in Taiwan
  • CogburnCogburn Posts: 62
    edited 2005-10-09 23:48
    Thanks Kramer-and all the rest for all your advice and help. I will put the circuit together tomorrow post my results.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Showing up to school doesn't·mean you are a student any more than crawling up in an oven means that·you are a biscuit.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2005-10-10 06:39
    Postscript.
    I did look at the Art of Electronics and they have quite a bit to say about Power MOSFETs.

    1. While a current limiting resistor many not be technically required, the gate is the usual point of failure from poor handling around static electricity and/or repeated connects and disconnects. The best protection is to put a current limiting resistor on the gate. So keeping the 1k resistor in place will offer you some longevity, it is not redundant.

    2. Capacitance becomes a big issue because unlike regular transistors, the gate actually doesn't do much or any conducting. Thus at higher frequencies, behavior may change as a 'phase shift' occurs. There are ways around it, but read the text. Clue, bigger MOSFET is not always better.

    3. I mentioned an octal driver, but you really only need a hex driver for 4 inputs [noparse][[/noparse]something like a 74HC07 or 74HC06] to create a protective buffer between the Stamp and your project. One of these inverts the signal, so you have to do everything backwards in software; the other doesn't have the inversion, but there is a bit more of a tiny time delay.

    4. 3Volts switching with a 100V rated MOSFET is possibly a problem. The charts I looked at seem to think no one would go there. You may not be able to get full ON. I am not sure exactly why, but it seems that the relationship between High Voltage and High Power are in tandem [noparse][[/noparse]the more voltage you use, the easier it is to get more amps too]. The data is unclear about the bottom limit [noparse][[/noparse]which is quite common on most transistor specifications as they are selling more bang for the buck, not less]. You may have to resize downward to a lower voltage MOSFET. These come in sets in DIPs, but don't have the protective diode [noparse][[/noparse]so add the diode, a 1N4004] if you need to downsize. Again, you have to consider the power rating of the whole package as they collect heat together.

    As you see, there are many, many alternative designs based on what is available causing a change in direction.

    Hopefully, everything will work with what you have.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    G. Herzog in Taiwan
  • GadgetmanGadgetman Posts: 2,436
    edited 2005-10-10 12:56
    Just adding a comment...

    Whenever I salvage motors from something, I always try to locate the driver IC at the same time, either to read off the name so that I can get a new one from somewhere, or to remove it, and possibly parts of the PCB where it's located(this may also yield the correct circuit diagram).

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Don't visit my new website...
  • CogburnCogburn Posts: 62
    edited 2005-10-11 02:33
    Well, I built the circuit and connected the stepper.· It purrs along so nicely!· It·barely gets warm and I gave it a pretty good workout.· The 520's·seem to be doing·great.· I very much appreciate·the help and information.
Sign In or Register to comment.