Journey to resurrecting C controller...

    Hi everyone,

    approx 7 years ago i've found a more or less abandoned C70-Controller with FS02-Arm with no documentation at all. As i always wanted to play around with robots, i started to look around to get this one working again. This board and especially its moderator KWAKISAKI were of the upmost help in these days: The Batteries of Controller and Arm were completely depleted and JT1 was stuck. Two new Lithium-Batteries were soldered but the problem with JT1 was frustrating. It turns out, that simply removing the servo and turning the servo "by hand" (with pliers) got it back rotating freely by magic.

    Last problem (hunting for features :smiling_face: ) was the idea to upgrade the motherboard to a version with ethernet. Already in touch with kwakisaki, without the previous problems would not even been solved initially, it would have not been possible to even think about this step due to absolute lack of knowledge and further support. An asbolute shoutout to KWAKISAKI: THANK YOU!

    As times go by, always having the now working C70/FS02 in mind and what to do with it (laser engraving, milling... RoboDK became more and more interesting), a C80-Controller with FS06L-Arm crossed my way. Offered for a reasonable price and with the sellers statements, that the unit was in use until first quarter of the current year (2022 by the time of writing) and everything would be in place except the Teach Pendant (standard is a so called Type 2 TP on these controllers). Manufactured May 2002. I got in touch with - kwakisaki, who else :smiling_face: - and it seemed possible to adapt the TP from the C70, although a Type 1 TP, to the C80 by means of a simple Adapter. The key for the Control-Pad was plugged in, but the key for the Controllers cabinet was missing and therefore closed.

    We came to a deal and now this "thing" found its way into my garage.

    For the Teach Pendant: This...

    must fit into this:

    For the C80, the req. Plug for TP Type 2 would include

    Housing:

    MHO 06 L20 ILME Ilme MH-44.27 Tüllengehäuse niedrig, 1xM20 seitlicher Kabelabgang, für 1 Bügel (am Unterteil) | stex24.de
    Jetzt ILME Tüllengehäuse, Gr.44.27, Alu, 2 Bolzen MHO 06 L20 bei Stecker Express online bestellen | ✓ Keine Verpackungseinheiten
    www.stex24.de

    Inlay:

    https://www.stex24.de/cddm-24-ilme-stifteinsatz.-gr.44.27.-24ppe.-crimpa-10a.-250v-113508

    Pins:

    CDMA 1.0 ILME Ilme CD-Crimpkontakt Kontaktstift (M) 1mm² AWG18 silber | stex24.de
    Jetzt ILME Stiftkontakt, 1,5mm², Silber, Crimpa. CDMA 1.5 bei Stecker Express online bestellen | ✓ Keine Verpackungseinheiten
    www.stex24.de

    For the Type 1, it is an Japan JL05 Housing - Mouser might have it, but it has to be ordered in Japan :frowning_face:

    https://www.jae.com/direct/topics/topics_file_download/topics_id=88548&ext_no=06&index=0&_lang=en&v=2022082514203615043909

    So, my idea is to "Universal" my TP Typ1 with putting an available Connector into the Cable and have "two" Adapters, one to the C70, one to the C80. :smiling_face:

    Kabelsteckdose & Steckergehäuse 24-polig Rundstecker
    Preiswertiges Set - Stiftgehäuse und Buchsendose inkl. Crimpkontakte. Marke Thoughcon. Betriebspannung 250V. Online bei Kalitec bestellen!
    www.kabelschuhe-shop.de

    Parts are on order, its xmas, lets wait...

    Ok, now the setback: Opening the cabinet without key is easy: Removing the Front-Plate the inner locking-plate can be reached and with unscrewing the 10mm-nut the door opens (its hold by the main switch as an additional safety feature, but turning it to "over-off" releases the door (unit is european-version, the american-version seems a bit different concerning the mains-switch).

    First look: Ok, not much dust, but... fu.k, mainboard 1GA/1HA missing:

    ...tbc...

    ... a bummer. However, the parts-list showed a possible variant, the 50999-1991 mainboard i pulled out of the C70 a couple of years ago. Checking with, you guessed it right, kwakisaki, it was worth a try, even with a note, that this board was only forseen for controllers until may 2001, a year before my unit was built.

    And: It was to expect, that the Controller/Arm is in a total interlock/error-state, due to the still missing Teach Pendant.

    Ok, lets try. Prior everything i exchchanged the batteries in the arm and the controller. They were still original, cells are Maxell ER17/50 (or in general type 17505, what is more common today). They were stamped 1202 what in best case can be translated to Februar 2012. But, despite the fact that, IIRC, Lithium-Cells can be stored up to 10 years without significant self-depletion (correct me if i am wrong) - these puppies were totally flat. Re-Builing them with 17505-cells is no big deal, however, throwing the Kawasaki Parts-Number 50750-1018 into Ama.on brings up nicely(?) ready-to-use batteries with the correct plug.

    After sliding in the old 50999-1991 mainboard and connecting computer first power-on, with switches still standard (SW8-0):

    Ok, basically, its alive :smiling_face: Next step was a boot with SW8-1:

    After this, i was able to do a SAVE:

    c80.as.txt

    ... and with SW8-0 it boots to:

    This is my current status. What i still have is a backup of the old C70. The former firmwares are written into the controllers housing and, for the Zeroing, i have the values from the factory:

    Additionally 4 sheets of the factory-setup:

    As this thread is intended to document the way to ressurect further C or D-controllers in the future, we will try to document all required steps.

    Next steps would be the adaption of a TP to the Controller and upload of the required Firmwares.

    Controller is Model C80F-A002, Serial FS0060269, manufactured 05/2002.

    Arm is Model FS006L-C, Serial FS0060269, manufactured 2002.

  • Nation December 28, 2022 at 12:14 PM

    Approved the thread.

    Nice initial posting, lots of useful images and information here......... :top:

    What a bugger about the CPU and Teach pendant, I hope the seller will continue to support with obtaining missing items as this clearly could never be sold as a 'working/complete system' with missing parts.

    So the good thing is your 'old' CPU board appears to boot up and allow for initialization, that is good sign, even if only temporary till you get a teach pendant connection/MFP Type 2, there are still roads that can be travelled.

    1. With the info I provided, you should be able to identify on the TP connection:

    - Emergency stop circuits.

    - Teach Lock circuit.

    So by using estop buttons/switches/jumpers you should be able to make good the safety circuit to allow for motor power and some automatic AS commands and therefore send instructions to move robot via KCWin32 and possible load in joint exercise programs (written offline) to test the arm.

    Obviously you would need to consider safety of your environment and suitable mounting position of robot for testing.

    This would be valuable time well spent to test the functionality.

    Having the original setting sheets is also good as you have default information that could be input, but the main thing is to try:

    - Get the controller in an empty default condition for your arm using minimal settings.

    - This will allow for initial testing and confirm working/non working conditions.

    Using KCWIn32, carry out ZROBOT to see if your arm is available, if available then no SV firmware download required and should be possible to make minimal settings along with teach pendant emergency stop and teach lock connection jumpers to send motion instructions and load data to controller.

    Next steps: Ive changed the arm to 91 acc. controllers spec-sheets (FS006L-B001).

    After next boot controller greeted again with Error 1019 (Checksum). Ive cleared this one by setting CHSUM to 1 (Effective).

    After clearing the Error-State through the "Clear"-Button on the Control-Panel, error was cleared, but now a 1420 came up:

    According to the Troubleshoot-Manual this one pops up mainly when the AS-Software (which is on the 1HA pulled from the C70-controller and now in the C80) or data of ID-code in AS-Sfotware is different from the 1GM PC or AS Software does not correspond to the 1GM board.

    What now brings me to the conclusion that i need a TP for uploading at least a suitable firmware on the 1HA to correspond to the 1GM-board...

    Very probable, your arm is C version and that firmware appears to just have upto B, it was worth a shot.

    Without the CPU board, no data is retained anywhere.

    So your CPU that you are using is using it's internal firmware, so more than likely that would need upgrading.

    So your mission, should you choose to accept it:

    - Use your C70 controller to flash your 50999-1919 CPU card with the C80 firmware.

    Doing this would obviously carry a risk as you are intruding on your C70 working controller.

    - It requires you removing the CPU card from your C70 for at least 40 mins without battery connection.

    - Risk would be yours or course.

    Hello everyone.

    Just noticed that my last post was approx a month ago and there has been so much improvement and communication between kwakisaki and kawasaki germany behind the scenes that we decided not to chit-chat everything in this main thread but to conclude the highlights and howtos on an ongoing way.

    First of all: The Robot is moving, the motors are working.

    What was the way to this milestone:

    On the last post, we were stuck at this nasty error-code 1420. As i did all the flashing of the card on my C70-controller und putting them again into the C80 for testing, we decided to rule out possible root-causes of the problem: First of all, i got my hands on a New-Old-Stock 1HA-Board, 50999-0020, what initially was rumored to be a successor of the most desired 50999-2384, Ethernet, 4MB, Mainboard. This was meanwhile confirmed by Kawasaki Germany, that the the 50999-0020 can be seen as a full replacement of this 2384.

    44948796oc.jpg

    44948797cn.jpg

    Next thing was hook-up a Tech-Pendant (or MFD: Multi-Function-Display, Multifunktionsdisplay in German) on the C80, because the original one was missing. We had some hints in the Manuals, that there was the possibility to have a Type-1 on the C80, however, standard would be the larger, white, Type-2 variant.

    45119373dt.jpg

    However, the phisical connectors are quite different between my C70-Type-1 and C80-Type-2:

    Connector on C70 Controller for Type-1 TP:

    44938603dp.jpg

    Connector on C80-Controller for Type-2-Teach-Pendant:
    44938601xj.jpg

    Looking into the wiring, all is straightforward, except another pinning for the EmerSwitch and some other layout of the two trigger-switch-circuits.

    The C80 "expects" to have two seperate Circuits for each trigger-switch, whereas the Type-1/C70 has two Trigger-Switches as well but the are pinned out "or", what means if the left or the right button is pressed, the circuit is closed. On the C70-side, there are two inputs as on the C80, but one input is already bridged by factory.

    So i wired the Adapter accordingly - and it works.... not directly.

    First problem was, after a quick test, the TP seemed to be dead suddenly, but it turned out, that ambient temperatures of below 5deg C/40deg F will stop the TP working (or at least lighting up the CCFL, didnt testet it).
    Second, i managed initially NOT to have continuity in my Trigger-Switch-Bridge as described above, what pops a "inconsistent"-state error.

    After all of this was fixed - voila. The TP initially complains about a Software Mismatch, so it was flashed to the original TP-Firmware the Robot was delivered from the factory (what was/is for a type-2 TP, but it works) what was supplied by Kawasaki Germany on request (Thanks again!).

    Thereafter AS and SV-Firmware was flashed to the original states, factory like, and.....

    1420-error again! :frowning_face:

    So, finally, this error was solved by going through the factory-sheet and configuring the C80 as it should. What was finally exactly solving the error is unknown, but there were a couple of inputs needed, which were incorrect after the full-reset (motion-mode, servo-mode, controller-type e.g.)

    Actually the current status is:

    - Floppy needs to be enabled/checked

    - Motors/Joints only turn in one direction - even it turned in one direction, not able to turn opposite.

    - Arm is lying on its side, will be mounted during the next days.

    Will keep you updated!

    Like before, a big shoutout to kwakisaki, who was&is a supporter und motivator to try&do and bring things forward. thanks a lot!

    Edited 2 times, last by cguenther (January 30, 2023 at 8:18 PM).

    Next progress on the Floppy-Drive:

    I got a confirmed working AS-File from kwakisaki to upload and set things correctly.

    Thereafter unscrewed Floppy and measured supply-voltage of 5V. All ok. To keep things short: The Controller tried to access the floppy, which could be confirmed by the short lightup of a red LED on the Serial-to-Floppy Interface-Board on the Floppy-Assy ang putting out "drive not ready error" thereafter. So obviously the Floppy-Drive itself was bad, there was no "drive"-light on the front of the drive light up.

    after checking the spindle-motor of the drive by hand and turned them a couple of times - was very stiff at the beginning, the spindle became more and more easy to turn like you would expect it to be. finally put all together - voila, floppy works. check.

    I was able to get a 50817-1260 Teach-Pendant, what is obviously at least the same Type-II TP as to be on the C80 originally, but with different Connector.

    However, i was told the unit might be unserviceable - and thats correct. The LCD is totally shot, as something hit the display and made it totally unreadeable, Glass broken. However, the Touchscreen still seems to work and the CCFL-Backlight as well.

    I disassembled the unit and the LCD-Screen is originally made by Kyocera, Type KCS072VG1MB-G02.

    First look around showed, that the usual sellers are well aware about the "business"-background of these LCDs. Prices range between 700 and 1200 USD. So i think i need to look around more closely, hoping that i will somehow be able to find one.

    Well, I did pass on details of the supplier in the UK to you that has those pendants and they don't sell anything with broken LCD screens, they just can't guarantee the functionality.

    So may be you should contact them and see if you can strike up a deal.

    - mention my name as a referral, they may offer some nice discount to you.

    Hope you haven't lost too much money obtaining it.

    Hi everyone!

    Thanks to the contact brought up by kwakisaki i was finally able to frankenstein a fully working TP, based on an 50817-1260R, to hook-up to my C80-Controller.

    Actually i now have 5(!) TP's here, some with faulty safety-switches (i replaced all 4 safety-switches on my device with new switches, actually still available from mouser), but i think i now have at least 4 working LCD-screens, 2 working TouchPad's and 4 front-glass-screens, at least 4 TP-Motherboards and so on...

    if someone needs spare-parts, please let me know... :smiling_face:

    Quote from cguenther

    Thanks to the contact brought up by kwakisaki i was finally able to frankenstein a fully working TP, based on an 50817-1260R, to hook-up to my C80-Controller

    Love it, so she's up and dancing now......Cool.


    Quote from cguenther

    Actually i now have 5(!) TP's here, some with faulty safety-switches (i replaced all 4 safety-switches on my device with new switches, actually still available from mouser), but i think i now have at least 4 working LCD-screens, 2 working TouchPad's and 4 front-glass-screens, at least 4 TP-Motherboards and so on..

    Will be able to 'frankenstein' another out of your supply of spares as I am sure someone would be interested in it, that way you could cover some of your cost outlay......

    Maybe make one with a backwards adapter to hook up with your C70, then you'll have the whole type1/type2 pendants compatibility covered.

    Quote from cguenther

    So, finally, this error was solved by going through the factory-sheet and configuring the C80 as it should. What was finally exactly solving the error is unknown, but there were a couple of inputs needed, which were incorrect after the full-reset (motion-mode, servo-mode, controller-type e.g.)

    Sorry f I wrote here, I have a similar problem with a C80 controller, I tried to do do system initialization, and I'm afraid that I have wrong value on motion and servo mode. Can you explain more into details how did you change this parameters? Is there a manual to explain it?

    Thank you very much

Advertising from our partners