megrimm
Posts: 28
Joined: Mon Jul 16, 2012 1:34 pm

Thermo CRS F3 Robot Arm Controller

Mon Jul 27, 2015 10:08 pm

Hello,

I am completely new to this (robot arms) so bare with me.

I have a Thermo CRS F3 Robot Arm that was gifted to me. I DO NOT have the controller (C500C). Would it be possible to build a controller with the RPi to control the arm?

Here are some minimal specs:
http://www.robotsdotcom.com/images/F3.pdf

If i had the controller i would use RAPL-3 language to control it as so:
http://www.retisoft.ca/instruments/crs/ ... notes.html

I am actually not even sure where to begin to control this thing so any advice would help tremendously. Basically the at the bottom of the arm there are a bunch of wires hanging out. I am assuming with them, as long as I can power the thing, I can control the motors.

Thanks for any help or guidance!
mark

BMS Doug
Posts: 4423
Joined: Thu Mar 27, 2014 2:42 pm
Location: London, UK

Re: Thermo CRS F3 Robot Arm Controller

Tue Jul 28, 2015 7:35 am

megrimm wrote:Hello,

I am completely new to this (robot arms) so bare with me.

I have a Thermo CRS F3 Robot Arm that was gifted to me. I DO NOT have the controller (C500C). Would it be possible to build a controller with the RPi to control the arm?

Here are some minimal specs:
http://www.robotsdotcom.com/images/F3.pdf

If i had the controller i would use RAPL-3 language to control it as so:
http://www.retisoft.ca/instruments/crs/ ... notes.html

I am actually not even sure where to begin to control this thing so any advice would help tremendously. Basically the at the bottom of the arm there are a bunch of wires hanging out. I am assuming with them, as long as I can power the thing, I can control the motors.

Thanks for any help or guidance!
mark
Hi Megrimm,

As you say, that mass of cables will be connected to the motors and associated rotary encoders. Figuring out which wire goes to which motor (and the voltage and current requirements of that motor) is going to be the important part.

You will need a motor driver board for each motor within the arm (figuring out which driver board is right for which motor is why you need the motor specifications), then the Pi can control the driver boards.
Doug.
Building Management Systems Engineer.

megrimm
Posts: 28
Joined: Mon Jul 16, 2012 1:34 pm

Re: Thermo CRS F3 Robot Arm Controller

Tue Jul 28, 2015 1:40 pm

Hello,

This is very helpful for getting started. Thanks!

mark

User avatar
B.Goode
Posts: 14129
Joined: Mon Sep 01, 2014 4:03 pm
Location: UK

Re: Thermo CRS F3 Robot Arm Controller

Tue Jul 28, 2015 1:47 pm

It looks like a great toy to be able to hack around with.

But can I just say.... That thing weighs as much as I do! I imagine it could do some damage to people or property if you set it in motion and can't control it. Do be careful...

BMS Doug
Posts: 4423
Joined: Thu Mar 27, 2014 2:42 pm
Location: London, UK

Re: Thermo CRS F3 Robot Arm Controller

Tue Jul 28, 2015 1:58 pm

B.Goode wrote:It looks like a great toy to be able to hack around with.

But can I just say.... That thing weighs as much as I do! I imagine it could do some damage to people or property if you set it in motion and can't control it. Do be careful...
I agree, the response time on the movement is frighteningly fast, you will want to be sure that you don't smash up bystanders, property or the arm while figuring out the connections.

The motors within the arm are probably stepper motors (for more precision when positioning).

Speed and Workspace
Axis Workspace Max Speed
J1 (waist) ± 180º 240º/second
J2 (shoulder) - 135º / + 45º 210º/second
J3 (elbow) ± 135º 240º/second
J4 (wrist rotate) ± 180º 375º/second
J5 (wrist pitch) ± 135º 300º/second
J6 (wrist roll) unlimited 375º/second

With 6 joints, if each had a 4 wire stepper motor and 2 wires for the encoder you would have 36 wires to figure out, is that about right?
Doug.
Building Management Systems Engineer.

megrimm
Posts: 28
Joined: Mon Jul 16, 2012 1:34 pm

Re: Thermo CRS F3 Robot Arm Controller

Tue Jul 28, 2015 2:08 pm

yeah reading some of the docs there is a lot of information concerning "safety" so I am aware of the damage it could cause (to me especially!)

...and yes you are correct... it is really heavy! I can lift it myself which I am sure could easily throw my back out! Although i have only had to move it twice (cross fingers).

as far as encoder driver boards go, i have had quite a hard time finding technical specification documentation (i have user guide, c500c controller guide, rapl-3 programming guide etc) specifically the motor tech. I am assuming i can just open the thing up to figure out what motors are used in order to match driver boards. i am also assuming the driver boards are going to be a little more advanced than what i can get at sparkfun, etc or maybe not?

m

rotwang
Posts: 243
Joined: Sat Dec 07, 2013 1:12 pm

Re: Thermo CRS F3 Robot Arm Controller

Tue Jul 28, 2015 5:33 pm

megrimm wrote:yeah reading some of the docs there is a lot of information concerning "safety" so I am aware of the damage it could cause (to me especially!)

...and yes you are correct... it is really heavy! I can lift it myself which I am sure could easily throw my back out! Although i have only had to move it twice (cross fingers).

as far as encoder driver boards go, i have had quite a hard time finding technical specification documentation (i have user guide, c500c controller guide, rapl-3 programming guide etc) specifically the motor tech. I am assuming i can just open the thing up to figure out what motors are used in order to match driver boards. i am also assuming the driver boards are going to be a little more advanced than what i can get at sparkfun, etc or maybe not?

m
Some quick googling turns up a reference to "brushless motors", which may indicate that this thing uses DC Servo's (which seems about right given those rates of move), in which case you are about to enter a whole world of fun.
From a safety point of view, establish the envelope which this thing can reach and install a kill switch which cuts all power from well outside that zone. And always use the kill switch before entering the zone (that's why industrial bots are surrounded by substantial cages with interlock switches on the gates).
Servo drivers will need a lot of tuning to get a decent response without overshoot, so read up on control loops and stuff, start off slow and no matter how trivial the adjustment, stay out of the zone until after you hit the kill switch.

megrimm
Posts: 28
Joined: Mon Jul 16, 2012 1:34 pm

Re: Thermo CRS F3 Robot Arm Controller

Wed Jul 29, 2015 7:16 pm

Hi,

I opened up the bottom area and basically found 2 of these:

http://www.ebay.com/itm/TAMAGAWA-SEIKI- ... 0863576745

tamagawa sa 35-11/24bit-LPS-5V ts5643 n 100

.... and yes, it looks like a whole world of not really too fun.

so my questions: whats next? how do even begin to drive this thing?

thanks!
mark

BMS Doug
Posts: 4423
Joined: Thu Mar 27, 2014 2:42 pm
Location: London, UK

Re: Thermo CRS F3 Robot Arm Controller

Wed Jul 29, 2015 8:38 pm

megrimm wrote:Hi,

I opened up the bottom area and basically found 2 of these:

http://www.ebay.com/itm/TAMAGAWA-SEIKI- ... 0863576745

tamagawa sa 35-11/24bit-LPS-5V ts5643 n 100

.... and yes, it looks like a whole world of not really too fun.

so my questions: whats next? how do even begin to drive this thing?

thanks!
mark
We start with lots of googling.

This appears to be a PDF of a scan of the specification document for the rotary encoder.

The Tamagawa website has links to pdf files for their products, but the servo you have is probably not their current item, maybe the wayback machine will give us more relevant information.
Doug.
Building Management Systems Engineer.

AlEwing
Posts: 1
Joined: Tue Jan 12, 2016 12:47 am

Re: Thermo CRS F3 Robot Arm Controller

Tue Jan 12, 2016 1:00 am

Hi, Mark,
How is your project proceeding? I am interested in alternate programming of CRS/Thermo robots.
I have not done what you are trying, but I do have two functioning CRS/Thermo F3 robots operating with their C500C controllers.
Also some CAT5 and CRS255 robots, with similar controllers.
I have programmed the controller-robots via CRS's ActiveRobot interfaces, but now mostly use the C500C serial interfaces directly (allowing me to program with current computers).
Let me know if I can be of any help, with info on the C500C interfaces. And let me know if any of your trials worked.
Al
LabSense Technologies LLC
http://www.linkedin.com/in/alanewing

megrimm
Posts: 28
Joined: Mon Jul 16, 2012 1:34 pm

Re: Thermo CRS F3 Robot Arm Controller

Tue May 03, 2016 2:49 pm

Old thread but just an update:

I HAVE NOT made any progress since last year. The arm has basically sat in my basement since a little after I started this thread. With that said I do have a little time off this summer and was thinking about trying my luck with it a little again this summer.

I do not have the C500C controller that AlEwing mentioned in January unfortunately so I have no way to program with the RAPL3 language etc.

At this point I was really thinking about gutting it (other that the motors), figuring out power and powering each servo independently, and controlling the servos with Arduino boards and/or RPI which may seem easy enough? If I just try one motor at a time maybe I can make headway. I think looking at the larger picture of the entire arm was a little overwhelming last year so just getting one motor to work will at least be a first step.

I am looking at threads like this for help but so far I have not found too much detailing controlling an industrial robot arm with an arduino/rpi/etc.

Anyway I have attached pictures in anyone is interested in visualizing what I have here.

cheers
mark
IMG_0145.JPG
IMG_0145.JPG (52.79 KiB) Viewed 9112 times
IMG_0147.JPG
IMG_0147.JPG (50.01 KiB) Viewed 9112 times
IMG_0150.JPG
IMG_0150.JPG (30.05 KiB) Viewed 9112 times

Larsen2000
Posts: 1
Joined: Mon Aug 08, 2016 11:31 am

Re: Thermo CRS F3 Robot Arm Controller

Mon Aug 08, 2016 11:36 am

Hi!
you got PM!

zefroncanada
Posts: 1
Joined: Fri Oct 21, 2016 4:25 am

Re: Thermo CRS F3 Robot Arm Controller

Fri Oct 21, 2016 4:30 am

Im not sure how your progressing with your robot arm...but i do have a controller c500c that Ive just listed to sell...am asking 600.00 - let me know if interested.
Michael B
w a y 2 c u t e 2 0 0 0 2 0 0 1 @ y a h o o . c o m

megrimm
Posts: 28
Joined: Mon Jul 16, 2012 1:34 pm

Re: Thermo CRS F3 Robot Arm Controller

Fri Oct 21, 2016 1:41 pm

hello,

slow progression...

i just found out last month that my arm is missing the umbilical drawer/baseboard assembly. the part i need was cited as:

10-0030 Base Board Assy.
BASE BD REV 0 R-EA1420-001-00

the arm is supposedly not operational, without major mods i assume, without this part.

so im just sitting on this thing until a solution presents itself with no need for immediacy.

thanks for the controller offer though! I will keep in mind if i get correct part or other ideas for moving forward!

cheers
m

BMS Doug
Posts: 4423
Joined: Thu Mar 27, 2014 2:42 pm
Location: London, UK

Re: Thermo CRS F3 Robot Arm Controller

Mon Oct 24, 2016 7:02 am

megrimm wrote:hello,

slow progression...

i just found out last month that my arm is missing the umbilical drawer/baseboard assembly. the part i need was cited as:

10-0030 Base Board Assy.
BASE BD REV 0 R-EA1420-001-00

the arm is supposedly not operational, without major mods i assume, without this part.

so im just sitting on this thing until a solution presents itself with no need for immediacy.

thanks for the controller offer though! I will keep in mind if i get correct part or other ideas for moving forward!

cheers
m
Thats probanly a staging point and sub controller board, your earlier plan of figuring out one motor at a time still has merit I think.

Start with the least dangerous, where the movement won't be able to cause any damage to the surroundings (wrist rotate perhaps).
Doug.
Building Management Systems Engineer.

steibeldj
Posts: 1
Joined: Sat Dec 17, 2016 10:25 pm

Re: Thermo CRS F3 Robot Arm Controller

Thu Dec 29, 2016 10:23 pm

megrimm ,

I contacted you on eBay. My email is dennis.steibel @ realized.com. Let's discuss your robot and motion control project. I am doing the same thing as you are with a different approach for the moment.

bleduc
Posts: 1
Joined: Wed Jan 04, 2017 4:01 am

Re: Thermo CRS F3 Robot Arm Controller

Wed Jan 04, 2017 4:21 am

megrimm,

I am an engineer that worked at CRS. Please be extremely careful. The F3 is very powerful.

The C500 controller has many safety features that reduced the chance of injury, and even then you needed to treat this robot with respect. You are creating a dangerous situation if you build your own robot controller from a Raspberry Pi. The other issue you will run into is that this robot requires high current from a responsive power supply, something you would need to build and it too will require care.

Please think about finding a C500C controller and the missing parts, and keep an e-stop with you at all times.

Blair.

randalpope
Posts: 1
Joined: Sat Mar 04, 2017 6:15 pm

Re: Thermo CRS F3 Robot Arm Controller

Sat Mar 04, 2017 6:17 pm

Any updates on your project? I am starting in the same situation.

silverguy
Posts: 1
Joined: Thu Mar 16, 2017 10:22 pm

Re: Thermo CRS F3 Robot Arm Controller

Thu Mar 16, 2017 10:46 pm

hello,

I'm considering buying a CRS F3 with C500C controller. Anyone know where I can get software and documentation?

Thanks,

Don

drdownen
Posts: 1
Joined: Sun Apr 02, 2017 8:21 pm

Re: Thermo CRS F3 Robot Arm Controller

Sun Apr 02, 2017 8:37 pm

Thermo CRS F3 encoder is a Tamagawa TS5679N166AD4755 which from research tells me it a serial absolute encoder based on a spec that was written for the TS5679N132.pdf and the document from Tamagawa http://www.tamagawa-seiki.com/pdf/downl ... shusei.pdf. The Motor is a 4 pole AC induction motor determined by the coil windings on the stator. Based on the sticker on the side of the Thermo robot I believe the brake to be 24VDC as they require a 24VDC 1 Amp source on the robot . The concern I have is the Thermo robot has a sticker that states 77 VDC at 10 Amps for main power. This means a "normal" AC servo drive with 120V AC source will create 170 VDC and that is way too much for these motors. The only AC Servo drive I can find without rolling your own is this. http://www.dmm-tech.com/Dyn2_v2.html. If anyone has other ideas or thoughts on how to control these motors please tell me.

to2
Posts: 1
Joined: Wed Nov 22, 2017 11:14 pm

Re: Thermo CRS F3 Robot Arm Controller

Wed Nov 22, 2017 11:25 pm

I just bought an f3 robot and I am planing to donate it to tech hub in africa but I can't find the software anywhere, I can't event test it.
Can anyone please help me to test this robot?

Spirit532
Posts: 3
Joined: Sat Apr 21, 2018 9:03 pm

Re: Thermo CRS F3 Robot Arm Controller

Sat Apr 21, 2018 9:54 pm

Reviving a very old and dead thread here, but I'm currently working on reverse-engineering a KUKA KR3 robot, which is just a rebadged CRS F3 robot with a KUKA-made controller.

There are two, three-axis integrated brushless servo drives inside the robot itself(inside A1/2 and A4). The drives were designed in the year 2000 by Agile Systems, now defunct(bought by Allied Motion), but I have been unable to find anyone who has worked at that company, let alone someone who might still have the 18-year-old documentation about an obscure custom product. Attempts at reaching out to Allied Motion and Thermo Fisher(who bought CRS) were met with silence.

Both servo drives communicate with the robot by means of the same two RS485 lines(four wires, two diff pairs), unidirectional to and from the robot.
The drives are do their own motion processing, and they should be PID tuned from the factory to the brushless motors and harmonic drives(reducers from the actual HDS GmbH as far as I'm aware).
They work in a "target-based" architecture, where the controller sets the motor current and target position, and then the servo drive itself handles the motion control, PID, encoder and AC drive, reporting the actual position back to the controller at regular intervals, essentially easing off the load on the controller.

This, however, makes them exceptionally difficult to drive, because one needs to know the entire command set for addressing the six axes from a single serial line, and on top of that parsing the output information, which is essential due to homing requirements(which are kept by the robot using the baseboard's 3 batteries) and error messages that the controller may send(such as overcurrent/overtorque).

If anyone has more information on the communication protocol or a working F3/KR3 robot(regardless of technical skill), please get in touch with me at rpi_crs@spirit.re or via PM.
If email doesn't work(I try to respond to everyone), try the other methods listed on my site.

aNewBerry
Posts: 3
Joined: Sun Feb 24, 2019 10:52 am

Re: Thermo CRS F3 Robot Arm Controller

Sun Feb 24, 2019 11:11 am

Hello there.
I have been actively experienenting with the F3 platform as well. I too tried to contact the company and or anyone that knew anything about the motor controllers. No one wanted to share anything. No schematics, power connections or anything... geez. So that made me determined to research this myself and see if I can figure this out and share it with the community and fellow hobbiest.

I have a fully functioning unit and one that I have disassembled in order to log the communications to the onboard custom controllers. I plan to “sniff” the RS485 serial links on the functioning unit while I send and receive the relative commands etc to the controllers on the disassembled ones. Hopefully I will be able to put together the relative board addresses and motor commands. I have ordered the parts to tap the RS485 link on the working unit already. :)

I will be posting the voltage pinouts and how I hacked a defective controller (found very cheap) to provide the power that I needed. I think I can help out everyone. Just need to reverse engineer the communications... update to follow

Spirit532
Posts: 3
Joined: Sat Apr 21, 2018 9:03 pm

Re: Thermo CRS F3 Robot Arm Controller

Sun Feb 24, 2019 7:04 pm

aNewBerry wrote:
Sun Feb 24, 2019 11:11 am
Hello there.
I now have a functioning KUKA KR3(F3), but I haven't done anything related to physically reverse-engineering the drives, since I'm actively using the robot and the KUKA controls are quite good.
However, I have found a bunch of documents from later devices by Agile Systems, so feel free to contact me(see above, email or website) and I could send them over.

dieselxj
Posts: 9
Joined: Tue Feb 26, 2019 3:03 am

Re: Thermo CRS F3 Robot Arm Controller

Tue Feb 26, 2019 4:20 am

Hello all, I am brand new to the forum and brand new to robotic arms. Also the brand new owner of a CRS F3 arm with c500c controller and teach pendant. I have not even taken delivery yet. I do have a tiny bit of CNC learning. But my CNC mill went in the ocean during hurricane Harvey. So I purchased this F3 thinking at the worst I could use some of the arms parts to help get my mill up and running. maybe the encoders will transfer and maybe the servo power supply.
What i don't have is the RAPL-3 . This RAPL-3 is a software program correct?? So the owner would receive RAPL-3 on a CD-rom? and load it on a windows computer . This would let the owner write program routines for the arm? and then i guess load the routines into the c500c un which would then run the routines to the F3. Is that how this was supposed to work?
If i read correctly i may be able to get some functionality from the F3 with just the c500c controler and the teach pendant is this correct?
There is something mentioned in the literature about hooking up to the controller with a " terminal mode " Is this using a stand alone computer as basically just a keyboard to talk to the C500C and enter data? If so how is that connection made? If you can actually program the c500c I would like to learn about that?
I have a acquaintance that is a pretty good roboticist, and he mentioned I think the same situation as aNewBerry above about tapping into and logging the signal string going to the servos, in order to reverse engineer a controller. that is above my pay grade and hopefully not necessary since I already have the c500c.

I too was shut down hard on the phone with Thermo. Not very friendly folks up there. After 3 phone calls I finally fibbed a bit, I told the tech I had a F5, and was able to talk to a tech for a few minutes. All I wanted was some info on how to collapse the unit for shipping. the tech indicated that I would most likely be out of luck, but that if I could get some power to the arm there are some control (brake release) buttons on the arm itself? Then the tech went off on the batteries only being good for 5 weeks without power.


Well I guess I am trying to find a copy of this RAPL-3 program if any one has any info on this I am interested.


Also I think I did find some pin out info on the F3 and the controller in some left over PDF's on the net. Well heck I can't find the pin outs again, they are out there on the net, I just saw them today. I will post them if I find them again, I am looking There seems to be a fair amount of info out there, you just have to dig it out of the internet. below are some links. that may be helpful to others.

http://www.uoguelph.ca/~habdulla/course ... 0Guide.pdf
http://web.csulb.edu/~wmartinz/robot/cr ... onment.pdf
http://www-eng.lbl.gov/~dw/groupshare/p ... 0Guide.pdf

I think I can find the pin out info again, I just have to keep digging

Return to “Automation, sensing and robotics”