CNCmill, 'Paxton/Patterson' workable?


Results 1 to 19 of 19

Thread: CNCmill, 'Paxton/Patterson' workable?

  1. #1
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Lightbulb CNCmill, 'Paxton/Patterson' workable?

    So at my school we have a mini CNC mill, without any software (or license) or 'drawing card'. This mill is labeled as the 'Paxton/Patterson CNC training center' in bold white print on the actual machine.

    Some information and picture can be found here... MILL

    The software recommended is Mill master pro, which the school has no budget and i cant seem to find a abandonware version or otherwise.

    There's a substantial PC to connect with it, through parrallel (serial, printer) port though I'm not familiar with cnc or cam work at all.

    Looking to get this thing up an running! Any info or recommendations are great, more info from me is no problem as i have the machine here in my home... The same way one would fill out a book slip, ive done so with the machine :X

    edit- it seems that the mill was/is of sherline production somewhere along the line of things if that info helps

    Fizile(at)gmail.com

    Similar Threads:


  2. #2
    Registered
    Join Date
    May 2005
    Location
    USA
    Posts
    127
    Downloads
    0
    Uploads
    0

    Default

    To get the mill just up and running you could just use TurboCNC (www.dakeng.com). Now that is just the mill control software where you open .CNC files and run them. You can also edit and create programs by hand using TurboCNC as well. As for creating the code for the machine to run on, you can learn G-code well enough to write the programs by hand (which I would personally suggest doing first) or use a CAM program to write the toolpath code from a part for you. You will also need a good CAD program to actually design the part in as well.

    As for particular suggestions on CAD/CAM programs, its really hard to say - just personal preference to suit your needs. I personally use AutoCAD '06 and GibbsCAM '04 for my designing and machining needs. Those are quite pricey programs though - especially for being on a budget. You could use a lower cost alternative such as BobCAD/CAM, but again its up to the requirements you demand. Check out the Programming and Design forum of this site and you'll get a lot of options and even more help from members on what programs to use.



  3. #3
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default

    I tried working with turbocnc a bit, didnt seem like it was working just right... I attempted to 'jog' the machine after booting into dos via the floppy disk. No response =\ it did detect the device in the port but would not respond in the phsyical world.

    I have autocad 06 as well as solidworks 06, if theres a free cam program for either that would be perfect.

    A walkthrough or tutorial for turbo cnc would be good, i used the one from their website but maybe im missing something? looking forward to working



  4. #4
    Registered
    Join Date
    May 2005
    Location
    USA
    Posts
    127
    Downloads
    0
    Uploads
    0

    Default

    Yeah, you have to kinda tweak TurboCNC to get it up and operational with your unit. Luckily for me I had a buddy that had a setup file that I just loaded and it worked perfect (well almost). Ask in the TurboCNC forum about setting it up for your mill and hopefully they can take care of you. I honestly don't know enough about the program to set it up correctly with the machine. You might just have to go through and start fiddling with things to get it to work, just don't go too crazy.

    Basically any CAM program will take in files from either of those 2 programs you mentioned. Nearly all take in .DXF or .DWG and numerous take in SolidWorks files as well as .iges files.



  5. #5
    Registered dkowalcz's Avatar
    Join Date
    Apr 2003
    Location
    USA
    Posts
    131
    Downloads
    0
    Uploads
    0

    Default

    With TurboCNC (or any other program), just figure out which pins on the port are step and direction for which motors and set it accordingly. There's unfortunately no standard at all for parallel port signal wiring, every machine builder did it their own way.

    In the \utils directory of the TurboCNC download there's a utility called FKEYBIT - run it with one of these command lines:

    fkeybit lpt1

    fkeybit ltp2

    fkeybit lptm.

    You can twiddle the pins on and off with the function keys, and observe what happens - this should take a lot of the pain out of deducing what's what.

    Once you've got that figured out, then it's all about motor tuning. That's a whole 'nother post - I'll just wait until you're there.



  6. #6
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default

    So ive fiddled with fkeybit a bit... So in the lpt1 and lpt2 i get the same outcome of inputs pin 10,11,12,13,15,16 stay 'on' always while all outputs make no change in the inputs or actual machines action.

    lptm is a little different in that input pin 10,13,15,16 are always 'on' but input pin 11,12 only come on when activating any output bit, yet they return to their off state after about 10 seconds when all the output's are also turned off...

    Pretty much in the dark-



  7. #7
    Registered dkowalcz's Avatar
    Join Date
    Apr 2003
    Location
    USA
    Posts
    131
    Downloads
    0
    Uploads
    0

    Default

    OK, I think you needed a little background info, sorry:

    The parallel port on a PC has a logical address set in BIOS. When you boot it up, you'll see a message saying "Press <F1> for setup" or something like this (often it's <DEL> as well) - hit that and enter the BIOS screens.

    Set your parallel port for $378h as it's address, and SPP or BID for the mode.

    Then you can use "fkeybit lpt1" to twiddle with the pins. (lpt1=$378, lpt2=$278, lptm=3BC)

    Understand that there are three basic schemes for controlling stepper motors from a parallel port:

    Scheme 1: Step and direction (most common). Two pins for each axis. Pin A will set the direction to move, toggling pin B takes a step.

    Scheme 2: Phase drive. Typically four pins per motor, one per phase - used on the MAXNC machines and others. A full rotation would look like this:

    0001
    0011
    0010
    0110
    0100
    1100
    1000
    1001

    Scheme 3: RS232 over the parallel port and a "smart box" downstream. This means you're stuck using their software.

    Looking at some sources on the web, I'm pretty confident you have step and direction as the drive scheme.

    Steps are always small, say .0001" or so. So don't expect the motors to lunge forward when you activate a pin - it'll be an almost imperceptible tick. If you put your hand on the motor you can just barely feel it move.

    Another approach would be to take a peek inside their box and ring out the cables with a meter - this should reduce the number of possibilities to deal with. You are using the parallel port right? (DB 25 connector).

    I remember a guy in the mech eng department at my college trying to deal with this very issue - had a training mill that didn't work owing to lost software. He never got it working again, as he knew nothing about computers, electronics, or CNC (I think he was the "low totem pole" guy, so he got stuck with it). If you're in the same boat, you might sell it and buy a CNC taig new - probably be about an even trade. There are plenty of folks on this forum that'd buy your mill. I wouldn't want to walk you through re-wiring the control with geckos, although you could do that as well.



  8. #8
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default

    I found this Yahoo groups thread relating to the topic http://groupyahoo.com/group/Sherline...6?viscount=100 But i dont know how usefull it is to my situation. Maybe very useful? I attempted to use fkeybit after double checking the bios and working tediously through the Scheme 1: Step and direction method. No result =\

    Im connected through this wire, as seen at circuitcity.com http://www.circuitcity.com/ssm/Belki...oductDetail.do

    When i worked with fkeybit i was in windows xp, ill work through the dos floppy disk and see how it goes though it doesnt look good.


    EDIT------------
    I emailed a person having some trouble with the mill in 2003, this is what they responded -

    Hi (name?),


    I was able to successfully fix my Paxton Patterson CNC mill and it has
    been working for a couple of years now. You must have noticed that
    message was dated 2003.

    The most important thing I found out - the answer to that first question
    - is that the Microkinetics Optidriver controller in the P/P mill
    enclosure does NOT use standard LPT signals from a PC. It is designed
    for their controller card to be used in the PC. IF you are going to
    keep the Microkinetics stepper controller in the PC, you need to rewire
    your printer cable. If you use a standard printer cable you will blow a
    fuse in the Microkinetics power supply.


    Here's something I wrote up when I sold my spare Microkinetics hardware
    on eBay. It will help you hook up to the mill from a regular PC.

    1. Don’t use a standard 25 pin to 25 pin cable: This is because the
    OptiDriver is designed to connect to another MicroKinetics card (the
    OptiStep) that isn’t really necessary. Pin 25 on the driver card is +5V.
    You’ll ground that to your PC, blowing the 3A fuse in the power supply
    (ask me how I know!!) If you can find an OptiStep card, you can use the
    additional features it provides: things like limit switch inputs for the
    axes. If all you want to do is turn on a stepper motor and move the
    axis, it will work with just these pins.

    2. To connect this controller to a conventional printer port for use
    with TurboCNC the pinout is

    MicroKinetics DB-25 Description and Pin # (note offset of 1 pin) to PC
    1 MOTOR #1 DIRECTION to: LPT 2 Direction
    2 MOTOR #1 CLOCK LPT 3 Step
    3 MOTOR #2 DIRECTION LPT 4 Direction
    4 MOTOR #2 CLOCK LPT 5 Step
    5 MOTOR #3 DIRECTION LPT 6 Direction
    6 MOTOR #3 CLOCK LPT 7 Step
    7 MOTOR #4 DIRECTION
    8 MOTOR #4 CLOCK
    9 CURRENT 0
    10 CURRENT 1
    11 OUTPUT 1
    12 OUTPUT 2
    13 OUTPUT 3
    14 MOTOR #1 NEG LIMIT INPUT
    15 MOTOR #1 POS LIMIT INPUT
    16 MOTOR #2 NEG LIMIT INPUT
    17 MOTOR #2 POS LIMIT INPUT
    18 MOTOR #3 NEG LIMIT INPUT
    19 MOTOR #3 POS LIMIT INPUT
    20 SHIELD INPUT
    21 MOTOR #4 NEG LIMIT INPUT
    22 MOTOR #4 POS LIMIT INPUT
    23 EXTERNAL GND
    24 OUTPUT 4
    25 EXTERNAL +5 VOLTS DC

    * * * * * * * * *



    I ended up buying a Xylotex 3 axis controller board because the middle
    channel on the Microkinetics controller was bad, and switching to
    TurboCNC instead of the software that came with the mill. You probably
    won't want to do this, unless you can get the school to pay for it.

    Feel free to ask any specific questions you might have.




    Bob
    He seems to have valid info on what needs to be done, the cable is the issue... (my best guess)

    Last edited by File; 04-10-2006 at 08:58 PM.


  9. #9
    Registered dkowalcz's Avatar
    Join Date
    Apr 2003
    Location
    USA
    Posts
    131
    Downloads
    0
    Uploads
    0

    Default

    Yes, that's hugely useful information!!!

    I'd warm up a soldering iron and make your own cable if you can't find one - it's very easy to do. Run pin 2 from the PC to pin 1 on the microstep, 3 to 2, etc... From a quick glance, it seems that only pins (on the microstep) 1-8, 20, and 23 are really needed to get it running again, you can always hook up the rest later.

    FKEYBIT needs to run in DOS by the way, XP will keep you from accessing the parallel ports directly.

    If you're running XP you can look at Mach3 as well for control.



  10. #10
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default

    Alright so my teacher is going to get the cable pieces and solder it up a notch tommorrow after school so I'll have more to report later that day... going to attach pin 2 on one end of the parallel port connector thing, to the pin 1 on the other end... and go on to each. Do i attach pin 25 to pin 1 at the end of that? Am I thinking along the right lines for this?



  11. #11
    Registered dkowalcz's Avatar
    Join Date
    Apr 2003
    Location
    USA
    Posts
    131
    Downloads
    0
    Uploads
    0

    Default

    No, you want to leave the microkinetics pin 25 floating - as in your quoted text above. Hooking +5 straight to your port is generally a bad idea.

    Let me write up a harness list for you:

    Code:
    PC DB 25          uKinetics DB 25
    2                            1
    3                            2
    4                            3
    5                            4
    6                            5
    7                            6
    8                            7
    9                            8
    20                          20
    18                          23
    The PC parallel port itself is pretty simple. Pins 18-25 are all grounds. Pin 2-9 are output, pins 10,11,12,13,15 are input, pins 1,14,16,17 are either input or output. If you can get a good magnifying glass on the connector then you should be able to see the pin numbers.

    You should have had something at least respond if not move before when you were messing with fkeybit and TurboCNC at the prompt. You may want to have an electronics type at your school set up a signal injector for TTL logic and verify that the stepper drives actually move when they're commanded to. It would be no fun to discover that they don't work after all this effort - IMHO things do tend to get set aside for a reason so this is a definite possibility...



  12. #12
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default

    Alright so my shop teacher and I soldered the wiring of a printer cable. We took the wire from one end, which we labled pc, that ran from pin 2 (we checked by running current through each pin, he managed that part... labelled each wire depending on color to each corresponding pin.) and then soldered it to the wire coming from the other end which we labeled MA which connected to pin 3. We followed suit in that manner to the pins described by the post above.

    After cleaning up the mess, and connecting my pc to the machine and using 'fkeybit' to test, no dice. Also tested with turbocnc and Mach2 software without result.

    Maybe the cable needs more work? Only connected the pins described in your 'harness list' is there more needed? Also the person in contact through email described a fuse that burned out... could this be a factor~

    edit- i can take a photo of the cable mod if need be to see if its... ridiculous



  13. #13
    Registered dkowalcz's Avatar
    Join Date
    Apr 2003
    Location
    USA
    Posts
    131
    Downloads
    0
    Uploads
    0

    Default

    Could be that the drives are kaput for whatever reason - open fuse could be one; maybe an output stage is burnt. I'd pop the case and have a look. Maybe the steppers aren't actually plugged in - a lot of interesting things get done putting machines in storage.

    I expect the problem is not your cable, if you used a regular 25 pin DB cable then you probably did blow the fuse as your buddy mentioned. If you want to post a pic that's fine, as long as it's around 3-5' long or less things are pretty forgiving there...

    Again, from the info you'rve given it sounds like a drives issue. If you can get it to the point where applying a 5V square wave (this is an on-off-on-off) signal to the X clock and grounding the X direction on the drive box makes the X axis move, then any software should work. If you can't make it do that, then no software will work.

    Hope this helps.



  14. #14
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default

    Took some pics of the insides of the machine... Some curious things i found were this round orange thing with a wire detached, and the motor power fuse looks a-ok. None of the power plugs seemed out of order. Take a look at this orange thing and the other pics if desired~

    http://img.photobucket.com/albums/v4...e/DSCN0256.jpg

    http://img.photobucket.com/albums/v4...e/DSCN0255.jpg

    http://img.photobucket.com/albums/v4...e/DSCN0254.jpg

    http://img.photobucket.com/albums/v4...e/DSCN0253.jpg



  15. #15
    Community Moderator Al_The_Man's Avatar
    Join Date
    Dec 2003
    Location
    Canada
    Posts
    24221
    Downloads
    0
    Uploads
    0

    Default

    That look like either a MOV (Voltage dependent resistor) or a ceramic filter capacitor, both are used to filter voltage spikes, It will not affect the normal operation if disconnected as it is used to prevent damage by Electrical noise.
    The markings would probabally confirm the type it is.
    Al.

    CNC, Mechatronics Integration and Custom Machine Design

    “Logic will get you from A to B. Imagination will take you everywhere.”
    Albert E.


  16. #16
    Member
    Join Date
    May 2005
    Location
    USA
    Posts
    3920
    Downloads
    0
    Uploads
    0

    Default

    I'm hoping I'm not to late but for a school project what you really want is open and free software. For CNC I'd reccomend EMC. Check out www.linuxcnc.org.

    There ae many reasons for this, one is that he school has no budget and probably never will. Another is that the source code to every thing is there right from the begining, which means there are many educational oportunities here

    In any event it looks like you are already on your way to debugging the machine. Best of luck. I'm wondering though where you are with this machine?

    thanks
    Dave



  17. #17
    Registered
    Join Date
    Mar 2006
    Location
    United States
    Posts
    9
    Downloads
    0
    Uploads
    0

    Default

    I'll take a look at this EMC, I'm familiar with using linux (a bit). As for where I am, do you mean in a suburb of Springfield, MA. Or something along the lines of not yet having the machine respond through my computer~ I used ubuntu for about 6 months without windows, but then realized how badly games ran emulated through linux :X Maybe its better nowadays? It's been a while.

    And yes the cost and ease of use for the opensource software will make it easier to sleep at night knowing there is no software pirating going on through me for the school.



  18. #18
    Registered dkowalcz's Avatar
    Join Date
    Apr 2003
    Location
    USA
    Posts
    131
    Downloads
    0
    Uploads
    0

    Default

    Well, looks like you just need to troubleshoot until you find out what's not working... Try to find a board member in your area, I know that someone who has been down this road before could probably help you out in an afternoon - but for someone who is first starting out (like yourself) it can be daunting.

    As the other poster mentioned, if you re-crimp that cap back on it'll be a minor improvement, but it won't keep it from working if you don't.

    Things you want to check, in no particular order:

    Power supply(s) - there should be power coming in and going out. Look for 5V and some other voltage for the motors, probably on the order of 50VDC or so.

    Drives - see if power and signal are coming in, and if the motors are energized or not. They should "lock up" when they're on.

    Steppers/machine - make sure stuff is plugged in, cables have continuity still after being flexed, and that there's no mechanical binding to keep it from moving.

    Computer stuff - just check to see if the port signals are actually present at the port and at the other end of your cable.

    Good luck. I gather you're not that experienced with electronics, so don't go where you aren't comfortable - get an electrical guy to help out.



  19. #19
    Member
    Join Date
    Jul 2014
    Posts
    58
    Downloads
    0
    Uploads
    0

    Default Re: CNCmill, 'Paxton/Patterson' workable?

    This thread is old, but its the highest in google rankings when you search for info on these Paxton/Patterson CNC machines.

    The big issue to making these work is a simple one. However i NEVER see it directly addressed, though it is eluded to in another thread {which is the clue that got me pointed in the right direction.}

    The optidriver needs a signal of about 3vdc to work. My laptop has 3.3vdc when a pin goes high, and thats plenty.

    Also, as noted many times. The DB25 pinout that the optidriver board uses is also offset by one pin, if you plug in a regular db25 male/male straight thru cable you will blow the 3a logic fuse on the board.

    Since i didn't have a connector on hand, i just cut the cable in half and spliced all the wires to offset the pins.

    Code:
    PC DB-25 Optidriver DB-25
    2                 1
    3                 2
    4                 3
    5                 4
    6                 5
    7                 6
    8                 7
    9                 8
    18                9
    19               10
    20               20
    18               23
    The missing link here is the current select pin. On EVERY one of these pinouts, nobody has addressed the current select pins. Without at least one of these pins selected {to ground} THE STEPPERS WILL NOT MOVE!!!!

    You only need to select to ground pin #9 OR #10 on the Optidriver DB-25 connector. If you ground both you will deliver 1.2a {max current} to the steppers which is fine, but not absolutely necessary for operation.

    FWIW i am controlling using a trial version of Mach 3 right now.

    I used the application PARMON from GeekHideout to monitor the status of the LPT {db-25} pins on the laptop. Once i got the Mach software to where it would reliably flash the "LED's" in PARMON i knew the problem was isolated to the optidriver board itself. Once i grounded Optidriver Pin #9 the machine sprang to life!.

    i don't post here often, if anyone needs to ask me a question directly feel free. Xbunlocked@hotmail.com

    Last edited by Kawgomoo; 08-01-2014 at 02:17 PM.


Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

CNCmill, 'Paxton/Patterson' workable?

CNCmill, 'Paxton/Patterson' workable?