584,860 active members*
5,251 visitors online*
Register for free
Login
IndustryArena Forum > Machine Controllers Software and Solutions > Fanuc > Fanuc <--> Heidenhain encoder converters on Fanuc 21 control?
Results 1 to 8 of 8
  1. #1
    Join Date
    Mar 2005
    Posts
    17

    Fanuc <--> Heidenhain encoder converters on Fanuc 21 control?

    Hi all - I recently acquired a [gutted] Chiron FZ08 vmc that reportedly had a Fanuc 21i control on it, which was removed by the original owner. It was expected that there were also likely Alpha amps for the spindle and 3 axes, also which were removed. I'm assuming they were Alphas, due to the remaining mount hole, the mating drive connectors and fiber inputs, plus a couple of Fanuc logos in the cabinet. The axis and spindle motors are all red-caps, such as A06B-0162-B076#7000 (aM6/3000) for the x-drive. I was expecting to either replace the encoders for just incremental or use some of Pico's pulsecoder to incremental converters, if that were the need. I was surprised to see so many Heidenhain logos inside the cabinet on top of a bunch of what appears to be converter boxes. There was no indication on the provisioning paperwork that the mill had scales (glass or otherwise), nor have I found any. The count of boxes is in threes; so the spindle probably ran direct to the amp. Any idea what the setup may have been? I'm retrofitting to LinuxCNC, and haven't specifically chosen replacement drives yet (but may pull from my Tolomatic stash, which requires only incremental encoders, no halls) at least just for testing. Although absolute encoders would be ok, I'm not married to them at the moment.
    The encoder wires run in and out of raceways a lot, so it's a little difficult to fully trace, given the wiring of the engraved boxes, including their translation, seem to indicate that they are "1 input, 2 output" devices that may split the encoder signal. 3 of the engraved boxes and 3 of the larger Heid boxes are on one side of the cabinet, while 3 separate larger Heid boxes are on the right hand side. I haven't found an obvious model or part number on these.... Note also that the encoder lines actually run outside the cabinet and loop right back in (meaning there is an "outside" patch point to interrupt the encoder signal and perhaps....connect it to another controller or something? It is an actual loop, not just a split output.)
    Here's a couple pix of the toys:
    Attachment 206394

    Attachment 206396

    I can snag more pictures and start tracing back the loom if useful, but if anyone knows what I could do to find a pinout as well as power the assembly to start scoping out signals, I'd certainly appreciate it. I would be ecstatic to find out that the Heid boxes are converting from traditional incremental to FAnuc pulse stream, but wouldn't hold my breath. Once I get back to the shop I should be able to open up the back of the Z axis motor to get an encoder part number....

    EDIT: I happened upon some "Huberboxes" on ebay; and a tighter search on Huberbox found some more info for those engraved boxes, which are indeed splitters, such as found at their site here: http://www.huberbox.de/index-Dateien...temabzweig.htm

    Thanks,
    Ted.

  2. #2
    Join Date
    Aug 2007
    Posts
    793
    some of Heidehain encoders have sin output. these boxes are interpolators
    in this pdf you will find info how to count pulses for your purpose
    Attached Thumbnails Attached Thumbnails exe602.pdf  

  3. #3
    Join Date
    Mar 2005
    Posts
    17
    Guhl - thanks for that link; I'm familiar with Heid desktop interpolators, but hadn't seen the individually boxed ones before. I'd be really surprised if Chiron modified the Fanuc servos for Heid rotary encoders - which means I'm probably not looking hard enough to find the scales. Combinations of the interpolators and the Huberbox point to quadrature output on one side and a Fanuc output on the other - it may have been that the servos were tied directly to the drives (traditional), but the scales fed back to the control, as well as the optional external loop. The purpose behind it is still unclear, but starting to make a little sense. Regardless, it looks like I'll at least have quadrature from somewhere, yay!

    Ted.

  4. #4
    Join Date
    Jun 2011
    Posts
    0
    Hallo, i have a similiar problem, a colleague have a FZ08W with the original Fanuc 21M control. The machine lost its absolut position because of battery fault. Now we have the problem to enter the correct abs position into the parametres list. But my colleague is completely desparate with the machine. Last time with this fault the Chiron man came for 4 hours and want nearly 3000$ for this.

    My colleague want better to give away the machine instead of using it because the fanuc control have little or no programm help. you must type each line separately, no cycles are possible and other problems.

    If I can convert the Fanuc Pulsecoder output to 1Vss and use it in a Siemens or HH control we can have a much better usable machine.

    I can ask him, if he want to sell the control separately and retrofit it with Siemens, HH, or maybe Mesa.

    I would better entering the position into the control, but no one from Chiron gave us information about the single steps for it.

    As I know from the last one, you must enter a code into the control, switch it off and on and after that you can enter the right values into the PMC. But I dont know anymore, which code in which mode of the control.

  5. #5
    Join Date
    Mar 2005
    Posts
    17
    Uli12us-

    The servos in my FZ08 are also absolute (I was chatting with Peter in another thread about them) - Jon Elson has a Fanuc Serial -> Hall/Incremental converter board for about $150: Pico Systems : Fanuc Serial Encoder converter (pulse coder) [fanquad] - $150.00 - this would let you fall back to using the servo motors, but not the Fanuc drives which require FSSB input, so changing controls will be "an adventure" if you already have a completed system. My machine arrived without control or drives, and I'm a LCNC guy at heart, so retrofitting to that control and new electronics isn't such a difficult commitment.

    I did determine that the Heid boxes, (confirmed by finally finding the part numbers on the boxes, as well as the electrical diagrams of the Chiron) were 1v p/p converters to incremental, followed by splitters. The scales (after conversion) thus output quadrature incremental, however I haven't stripped everything out yet (I'm currently cleaning the toolchanger) to determine if the index pulse is at center of scale or edge of scale. Heid provided two options, apparently. I'm hoping that the index is at an edge, as that would make auto-homing easier (a known direction to limit), versus having to manually jog in the needed direction for homing.

    I'm not planning on using Jon's boards, nor trying to retain the serial nature/absolute nature of the existing Fanuc servo motors (at this time). I opened the pulsecoder and located A/B/Z TTL incremental on the coder, although it's not routed out to the connector, just test points. I don't mind hacking the pulsecoder to be incompatible with the Fanuc-specified pinout, but I know I'm a rare breed in that case. Maybe some day when that serial train is fully revealed to the public, including the zeroing/reset commands, I'll revisit it, but at this time, incremental is perfectly fine for me. The drives I'm installing will take hall+incremental, so I'll let the drives deal with the motor, and I'll feed the scales into LCNC. (Mesa hardware most likely, as I have a lot of that on hand)

    I touch a lot of Fanuc controls on a weekly basis (not a ton, but enough to remind me how little I like integrating with them, although many are very appropriate for their implementation and user base), so I'm quick to avoid them entirely. (I've had a couple Siemens 802 installs and it was just "Drop-in magic", comparably). The last time I had Fanuc absolute's go blank, we had to get the machine to within a few inches of "home", set a parameter to "1" in the control (actual parameter# and mode process varied per control model), reboot, then jog each axis until a home dog switch tripped. Returning the parameter to "0" stopped the process and made the axes happy again. This took about 5 or 6 tries. On my Mazak with Mitsubishi controls, there's a couple absolute drives for the toolchanger and those batteries die pretty quickly (it's a rarely used machine these days, due to lack of work), and I ended up receiving three different reset procedures to try, none of which worked, requiring a tech to come out (only $500, but still....) - and even then his procedure (luck and multiple attempts) took a while to fix the problem. Since these units are designed to be "out of sight", they end up being "out of mind" for both operators, owners and techs; when they go down, everyone has to scramble as it's not a part of daily activities. I'm ok with homing a machine at power on any day.

    Best of luck,
    Ted.

  6. #6
    Join Date
    Feb 2008
    Posts
    644
    Quote Originally Posted by laserted View Post
    Uli12us-

    I'm not planning on using Jon's boards, nor trying to retain the serial nature/absolute nature of the existing Fanuc servo motors (at this time). I opened the pulsecoder and located A/B/Z TTL incremental on the coder, although it's not routed out to the connector, just test points. I don't mind hacking the pulsecoder to be incompatible with the Fanuc-specified pinout, but I know I'm a rare breed in that case. Maybe some day when that serial train is fully revealed to the public, including the zeroing/reset commands, I'll revisit it, but at this time, incremental is perfectly fine for me. The drives I'm installing will take hall+incremental, so I'll let the drives deal with the motor, and I'll feed the scales into LCNC. (Mesa hardware most likely, as I have a lot of that on hand)

    Best of luck,
    Ted.
    You might note that LinuxCNC master as of a month or so ago supports
    Fanuc serial encoders (and SSI encoders) with Mesa hardware/Hostmot2 firmware
    (A big thanks to Andy Pugh for the driver)

    You need a RS-422 daughterboard of some kind with a TX/RX pair per encoder

  7. #7
    Join Date
    Mar 2005
    Posts
    17
    Peter - I was not aware of that; might be worth keeping the serial pins accessible too, then. I knew there was some I2C/SPI stuff going on a while back, I didn't know anyone else was pursuing other formats. I haven't pulled anything from the master repo in about a year; stable releases have been so frequent I haven't had to. Do you recall if that driver relies on your SmartSerial (limited to 5i2x and similar?) or would it run on an older 7i43? (of course, looking at your website - wow, you've been busy! - it might be worth my while just to pick up a 7i90 and not worry about it. ) Are the bit files as exciting to edit as they were in "ye older days" or is there an easier method? I recall having lots of unused "enable" outputs when I was using the SV6 etc firmware, due to the hardware I was connecting to. It's not really a big problem with 72 i/o sitting there. My major consumer of I/O is the toolchanger, which since it's semi-intelligent (mechanically) anyway, I'll roll a Modbus solution for it; it needn't be running in the RT layer.

    Regards,
    Ted

  8. #8
    Join Date
    Feb 2008
    Posts
    644
    Quote Originally Posted by laserted View Post
    Peter - I was not aware of that; might be worth keeping the serial pins accessible too, then. I knew there was some I2C/SPI stuff going on a while back, I didn't know anyone else was pursuing other formats. I haven't pulled anything from the master repo in about a year; stable releases have been so frequent I haven't had to. Do you recall if that driver relies on your SmartSerial (limited to 5i2x and similar?) or would it run on an older 7i43? (of course, looking at your website - wow, you've been busy! - it might be worth my while just to pick up a 7i90 and not worry about it. ) Are the bit files as exciting to edit as they were in "ye older days" or is there an easier method? I recall having lots of unused "enable" outputs when I was using the SV6 etc firmware, due to the hardware I was connecting to. It's not really a big problem with 72 i/o sitting there. My major consumer of I/O is the toolchanger, which since it's semi-intelligent (mechanically) anyway, I'll roll a Modbus solution for it; it needn't be running in the RT layer.

    Regards,
    Ted
    All firmware modules (with one exception) will work in all FPGA cards.
    The exception is the resolver interface which will not fit in the older
    5I20 or 4I65 Spartan2 based FPGA cards.

    Making new firmware basically means creating a new pinout file
    and re-compiling (not trivial the first time but not terrible either)

    Forgot to say that one advantage of using the serial interface is that it has much
    higher resolution than the quadrature outputs (65536 counts/turn vs 8192 on aA64 for example)
    because it uses interpolation from sine/cosine A/B signals

Similar Threads

  1. Replies: 10
    Last Post: 06-12-2013, 12:26 AM
  2. fanuc encoder
    By mr pitt in forum Fanuc
    Replies: 0
    Last Post: 09-21-2012, 09:52 AM
  3. heidenhain or fanuc??
    By gallzer in forum Uncategorised MetalWorking Machines
    Replies: 3
    Last Post: 11-13-2011, 10:54 AM
  4. Fanuc G Code to Heidenhain - Best method?
    By boolay in forum CNC (Mill / Lathe) Control Software (NC)
    Replies: 3
    Last Post: 09-22-2010, 10:45 AM

Posting Permissions

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