Staubli CS& V_TRAJSIG

[Puma760]

Checking thread277-123312, Whats the Current Status of the Offline V_TRAJSIG Software, Current Applications now have V_PAINT a graphical interface which helps with the the parameter file of speeds and segment types, the file managments of trajectoires etc.. I belive this could be done better via a Serial commands on aPC on the Serial Line

 

[tygerdawg]

Staubli discontinued their use of Adept Technologies' CS7x controllers and V+ Operating System & language. V_TRAJSIG was a V+ application. Staubli no longer supports V_TRAJSIG.

I think I saw something on the Staubli website about a "VAL_TRAJSIG" application. But the path capability of the TX/CS8/VAL3 robot system is so good, what's the point?

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[Puma760]

Yes, you are correct, but we still have many V+ Staubli Systems that could do with a simple low cost 3D trajectory maker. That is if we require a Joint position every 16 ms for true TCP speed control, what is good about the VTRAJSIG is the smoothing Filters and the 3D Arc profiling that is missing in V+. Also the Real time Trajectory control, Just like old days of old Puma ALTER Control.
The main reason is to try and update Staubli systems using old Vtrajsig, it requires a lot of re-editing of the parameter program per Trajectory for custom speed Etc..
Remember how VTRAJSIG started out, A PC connected to a Puma 700 and calculated a point every 2 mm.

 

[tygerdawg]

Yup, I agree. As powerful and useful as V_TRAJSIG was, the user interface was painful. We tried many times to get the boys over in Faverges to develop something more user-friendly.

The only 3rd-Party user interface for V_TRAJSIG that I ever saw was RELIABOT from SHAFI Inc. But they are out of business now. Again, Staubli & V+ don't exist anymore, so SHAFI probably wouldn't support it either. It was a bit clunky, anyway.

But the bigger issue here is the window of support for Staubli V+ robots is rapidly closing. Those are obsolete arms and the day is coming when you'll no longer be able to get parts or support for them. You should consider upgrading.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[Puma760]

I agree, but i need to keep my customers happy, and has you say, to be up with the avaialability and the cost of a VGB board or the Offline Adept windows. Looks like I must write a windows front end in VB.NET to talk serial to the CS7 and Ethernet of the CS7B. I simply need a 3D Libary DLL Function to calculate the positions of joint angles along a trajectory in the way VTRAJSIG does. I have copies of the V+ VTRAJSIG 41A,42B Source code to look at. The calculation takes so long on the Robot to Calculate, The answer must be that it's done on a PC.

COSIMIR, EASY-ROB, seems a possible as they output V+ Code. VISUAL COMPONANTS outputs their own style of code, which can then be proproceeced in to any code, and for me it could be VAL2,V+ or the Kawasaki AS

 

[tygerdawg]

RobotWorks will do this also:
(1) Select CAD entities to define a path
(2) Create a robot path with your desired path point spacing.
(3) Normally the output is Cartesian point data in the particular robot's native format. Is also a simple matter to output joint angles for each corresponding Cartesian point.
(4) Generates a delimited text file of joint angles for most 6-axis robots.
(5) Build your app from this data.

Learn more from

http://www.robotworks-eu.com.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[Puma760]

Thanks, have download a demo version, will check it out, Will this help on site where each customer would require their own PC. Out of the office this week, glue layeing with a Kawasaki SD130 using a CAD packacge that i written in VB.NET, It's only a 4 Axis Robot so i don't require a 3D Package. This in itself a bouns with the AS Languge that it's as a ARC command in 3D if i need it, that a command greatly missed in V+, and VTRAJSIG does that trick well.

 

[Theatronics]

Does the term “Staubli V+” refer to the 500 family of Unimation PUMA units?

I am rebuilding one and I have it about 95% done. I have collected an impressive set of college and even high school project reports because they all had some tidbit of technical data about the system.

Is there anyone out there who is familiar with the 506C and the CS4-SB controller?

Specifically I am working through a repair/replacement of the 5 ¼ floppy drive unit right now. 3D path control and the Alter/Supervisor interfaces were next on the list.

I will eventually begin upgrades but not before I get this working in ‘vintage’ condition and the document – document – document…..

Thanks
-Mike

 

[Puma760]

Theatronics, Val2, I was brought up on the Puma's. They were manufactured here in the UK. On on the Floppy Side of things. I have written a VB6 Program to emulate the 5-1/4 Floppy on a Standard PC, even the Operating system can be downloaded.I have 10 systems using this software. You could also use the AID port on the back of the VDT terminal to simulate command to read the floppy unit. The Alter command can be used internal and external, the serial protocol for the external alter is simple. When you get the DDCMP Supervisor link working, then the DISK Swicth can be used for saveing and loading of programs.

 

[Theatronics]

Bravo! I have always thought the Puma design was a nice looking system. I vowed to get one some day. Now that I have one my daughter is positively giddy about it. I imagine I would have been at her age also. So, I am in a rush to get it buttoned up and stable. There is a video on You tube of her first VAL II program trying to pick and place a plastic box before I even had the gripper hooked up. Search for "Theatronics Puma". It should be the top result.

If you would be willing to share some source code I could translate the VB6 to vb.net
Looking over my captured conversations I have decoded this much by hand. As you can see I am still trying to get the protocol together. If you can not send the source code, a secret decoder ring for this number jumble would be fantastic also.
(Puma was recorded in Decimal, Floppy in Hex. It helped me separate the two)

Puma:
d001 (SOH) <- LSI-11 asks for "VALI.I.560C"
d079 O
d086 V
d065 A
d076 L
d073 I
d046 .
d073 I
d046 .
d053 5
d054 6
d048 0
d067 C
d004 (EOT)

Floppy:
%10 (DLE) <- floppy sends some sort of ACK

Puma:
001 (SOH) <- LSI-11 sends what appears to be "READY"
082 R
004 (EOT)

Floppy:
%10 (DLE) <- floppy begins it's dump
%02 (STX) <- These three bytes are
%01 (SOH) <- always the same
%00
%08 Number of Bytes in packet?. Packets limited to 254?
%00
%EA ?
%F4 ?
%80 Somewhere in here I think there is a Checksum or other file information.
%01 ?
%98 ?
Every packet has 259 bytes including the (DLE STX SOH) header
The ones with less than 254 bytes are padded with zeros.
%00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 %00 < - - - Every packet ENDS with 2 zero bytes.


(First 2 packets are 8 bytes but one capture has a 4 byte 2nd packet.

Puma:
001
082
004
F:
%10
%02
%01
%00
%08 Again 8 bytes...
%00
%CA
%F4 ...and some more mystery numbers...
%06
%7F
%B4
----Trailing zeros left off---


(This next block appears to be the first packet of actual data
the 5th byte is FE (254) that happens to be the number
of data bytes that are in the packet after the 5th byte
Including the 2 trailing zeros.

Bytes 7 and 8 (00 A0 in this packet) increment by 248
with each packet. If That means that 248 bytes of data
Are transferred then 6 bytes are not part of the data.

If the trailing zeros are 2 of said total,
Byte locations 7 and 8 are another 2,
then Bytes 9 and 10 - - OR - -
bytes 251 and 252 must make up some sort of checksum.

I tried CRC-16, 2s complement and just rolling 8 bit. But
I have not found a match yet.

This is the entire packet so it should math-magic-ly solve to a checksum that is
contained in the packet... some-how...some-where...

Puma:
001
082
004
F:
%10
%02
%01
%00
%FE
%00
%00
%A0

%5F %00 %18 %00 %EE %9C %00 %00 %48 %2C %3A %52 %37 %03 %B4 %03 %C0 %03 %31 %04 %A1 %04 %80 %05 %DF %45 %01 %00 %4E %F5 %F7 %15 %82 %00 %DE %FF %DF %55 %01 %00 %4E %F5 %C5 %15 %9A %99 %40 %9F %0A %00 %C0 %0A %C0 %45 %E0 %FF %17 %20 %1A %00 %02 %05 %C0 %E5 %1A %00 %17 %74 %02 %00 %C0 %65 %74 %80 %0A %01 %42 %1C %0A %00 %43 %1D %0A %00 %8B %A4 %02 %02 %D3 %8B %FC %02 %04 %06 %40 %10 %01 %1C %04 %00 %F3 %02 %70 %11 %04 %00 %70 %11 %06 %00 %40 %1D %04 %00 %75 %10 %04 %00 %75 %10 %06 %00 %05 %10 %D8 %02 %77 %00 %00 %00 %DF %55 %01 %00 %4E %F5 %E6 %15 %0C %00 %6E %88 %56 %88 %C0 %15 %C8 %56 %C1 %15 %2C %80 %C3 %15 %12 %00 %50 %14 %C2 %7E %C0 %15 %EC %56 %C1 %15 %39 %00 %10 %0A %42 %7E %37 %0A %C8 %5F %08 %88 %37 %10 %48 %53 %C1 %15 %48 %63 %C1 %E5 %00 %37 %57 %74 %FA %FF %40 %E0 %17 %74 %FC %FF %80 %0A %80 %0C %37 %10 %30 %53 %68 %88 %37 %10 %2C %53 %08 %03 %C0 %E5 %00 %06 %17 %74 %FC %FF %80 %0A %80 %0C %37 %60 %18 %53 %F7 %09 %98 %84 %F7 %09 %64 %02 %F7 %27 %00 %00 %D4 %9B %0E %03 %DC %00 %00

(Here is the start of the next few packets
just to confirm that the 7 and 8 bits
count up by (F8) 248.

%10 %02 %01 %00 %FE %00 [%F8 %A0] %F7 %09
%10 %02 %01 %00 %FE %00 [%F0 %A1] %F7 %09
%10 %02 %01 %00 %FE %00 [%E8 %A2] %02 %03
%10 %02 %01 %00 %FE %00 [%E0 %A3] %37 %8A
%10 %02 %01 %00 %FE %00 [%D8 %A4] %4A %63
%10 %02 %01 %00 %FE %00 [%D0 %A5] %85 %15
%10 %02 %01 %00 %FE %00 [%C8 %A6] %F0 %0B
%10 %02 %01 %00 %FE %00 [%C0 %A7] %8B %20
%10 %02 %01 %00 %FE %00 [%B8 %A8] %06 %00
%10 %02 %01 %00 %FE %00 [%B0 %A9] %08 %00

Thanks
Mike@theatronics.com

 

[alainriaud]

I worked on the V_TRAJSIG ancestor before leaving Staubli.
I then worked on my own graphic user friendly version used in welding and grinding for big turbines repair and machining for the shoe industry. Absolutely same version can be connected to PUMA Val II robots, STAUBLI V+ robots and FANUC KAREL robots, of course with different performances
At that time it ran on Dos 6 based PC machines , more than 120 robots worldwide used it.
I started to port it to Windows, I reached 90% of the development, but seeing no market, I stopped and retired.
I'm open to reactivate this project if needed.
Alain Riaud
alain.riaud@ieee.org

 

[Puma760]

Alain, I helped to install a number of Units here in the UK in the shoe industrie, all have now been decommissioned. I did however buy a few complete systems and still have the DOS Systems Software and Dongles. No, there is no large market now, moden robotics trjectorys seem to have better OS sofware. And many have simulation packages. What was good about your software was that the path was constructed on the pc and the serial link was used to obtain the data. I have 1 Project that uses an old RX130 inverted on a 8 meter track, we use V_TRAJSIG for the TRACK and ROBOT to sync the MI6 & MI3 Cards together. A simple simulation package would benfit and a FTP package. IT can takes up 30 minutes to calculate a Path because of this 8 meter Track, and the data stroageis becoming a problem (SIO SCSI DRIVE 128mb). So i am currely looking a a KERMIT PC Driver, and, trying to find a 3D Path generator all at a very low cost. I will then tie this all together using VB.NET. Wish my Luck

 

[alainriaud]

Puma 760 : Are speaking about the RBSYSTEM software? If yes, we may have met.