Converting a Seiko D-TRAN XM 5000 Cartesian into a RepRap.

Most of you who linger in IRC have probably seen me brandishing this around lately.

My name is Christopher Thomas, and I am a student at Tarrant County College, Fort Worth, Texas currently majoring in Robotics and Automation. I have been given the privilege to use our Seiko D-Tran XM 5000 Cartesian Robot however I want. It started as just a simple PLC controlled project for my class, but I soon found that with a micro-controller and some simple materials, I could make it do so much more than Pick and Place...

Eventually I stumbled upon RepRap after looking for a micro-controller that is robust enough and developed enough to give me my desired results. (ie... Sanguino). As of today I have placed an order for 3-RepRap Motherboards and 6-Extruder Controllers (hopefully to be used in place of the Stepper Drivers, of which the PCBs were sold out.) I'm doing everything from parts to demonstrate to the classes that SMT is achievable with limited tools and is, at times, much easier and faster than doing the typical through-hole technique currently taught in schools.

Now, back to the Robot. Originally it comes with an SA3010 Servo Drive Controller that runs the DARLfour Language (referred as DARLV in the manual). Which is primarily based on BASIC programming. In addition to the SA3010 there is an additional I/O Drawer that houses additional motor drivers that can be used independently from the Servo Drive Controller. It also controls the full pneumatic system that powers the 5 Tool - Toolhead. This robot was donated to us by Motorola used in the mobile phone market as a pick and place device. It has since just sat there for the past 3 or 4 years. A couple years ago a few teams "got it running", but the last team blew out the 24V PSU, lost the manuals, and damaged a control panel.

What I'm hoping that I can accomplish in the next 3 weeks, is to have full 3 Axis movement controlled by either the Servo Controller or a modified RepRap/Sanguino board.

Now for some specs....

The X,Y,Z Axises are 800mm,600mm,140mm (200m total length), respectively.

The Resolution is .008mm for the X & Y, and .004mm for the Z.

The Speed is 1000mm/s for the X/Y, and 650mm/s for the Z.

Currently everything runs off of a homemade 24v Power supply built by a previous team, seeing as how I'm just starting out in the programming and building of this, its a challenge to overcome trying to adapt the RepRap electronics. Also, the Servo Driver Controller needs some repair on the front panel (I have to desolder buttons, key switches, all sorts of goodies that were destroyed.) I want to keep this thing as versatile as possible. Now if you will excuse me, exhaustion (and beer) is kicking in. I'm starting to lose my focus... Here are some pictures of my current robot, (and some others of another robot I'm trying to work on.)

My .Mac Robotics Gallery.

If you can get it running, you'll have the slickest repstrap anyone's seen, and you'll be able to build some pretty big objects. You might be able to build big pieces that other people are having cut from acrylic sheets.

Don't let the hardware repairs be an obstacle; a lot of guys are starting with no hardware at all.

Christopher,

That's one heck of a nice hardware starting point!
If the motors and their amps/controllers are still functional (and you can work out how to command them, you'll be over a mahor hump. IMHO, you top pripority should be fuses/breakers to prevent repeat electrical damage, and limit switches to help safeguard the mechanicsAre you doing this solo, or do you have team-mates? IMHO, debugging is often better done with at least on other person.

Good luck with your project,

Larry

---

Larry Pfeffer,

My blog about building repstrap Cerberus:
[repstrap-cerberus.blogspot.com]

@R.J. Bowman -

Thanks, yeah, one of my design goals and purposes was to be able to print the really large pieces in one go.

@Larry_Pfeffer -

Well, all of the axes have two forms of stop switches. hardware and software, built-in to prevent overall damaged, And there is a couple circuit breakers already on the Servo Driver Controller. However, I would like to attempt to bypass the SA3010 all together, if possible. Just to see. And do answer your question. I'm doing this solo for now. I've got some students near by in case I need some immediate assistance. Should have seen the look on one of the systems directors face today when I explained to him that it would only cost me about $300 to make a 3D Printer, compared to the $25K he threw down on his Dimensions 720BST. (granted the RepRap is still pretty limited on high resolution I'd think.) He said he shouldn't have a problem authorizing my purchase order for the PCBs and such, as long as I can present to him my goals. For retrofitting a $10K piece of equipment. (You can buy it on ebay right now for $2500).

Anyway, tomorrow should be fun, I'm going to attempt to power it up for the first time using its native hardware.

@Christopher:
What info do you have on the motors themselves? Stepper/Servo, Peak voltage, peak current, etc. The reprap hardware is set up for stepper motors with around 2 amps peak current. (Again, this depends on which generation of reprap/makerbot stepper boards you have.)

There are other options, such as using an arduino/sanguino/motherboard together with some higher power stepper drivers.

Please keep us posted,

-- Larry

PS Use a bit of caution when you say what you hope/plan to do (to your professors/administration.) Don't promise the moon, unless you already have it in your pocket.

---

Larry Pfeffer,

My blog about building repstrap Cerberus:
[repstrap-cerberus.blogspot.com]

Larry,

I'm still trying to find the technical manuals for the XM 5000. I'll take off the covers next week and look at them directly to see what I can derive from it. Do you think the newesst Extruder Controller would be suitable? Right now I know the I/O Drawer and everything operates on 24VDC. I can still get the robot running using its current servo controller, if I can accomplish that, they'll be satisfied, so its not too "risky."

Okay, just looked at the X & Y Motors. They're both 5VDC, So that's good. The control panel on the SA3010 is fried. So until I get a new one in, I'll have to find another way to interface the External I/O system without hacking at the cables. Talked to a supplier, they want me to pay $549 for the Engineering Manuals... I don't think so. Looks like I'm gonna be doing it the hard way.

That said, the I/O Drawer that the SA3010 and all pneumatic systems plug into is easily accessible and labeled moderately well. I just have to trace the hundreds of wires to each device card. With luck, all I'll have to do is run a few I/O wires to the controller board. If not I'll have to use the Extruder/Stepper Driver boards to try to control everything. (My original plan.)

I'll try to get pictures tomorrow.

Christopher,

To be able to advise re your machine's motors, I suggest:

Copy down any model number or related info printed on the motors

Take pictures of both the motors and whatever electronics they connect to.
Closeups of PC boards would be most helpful.

Count the motor leads, trace them, label where they went and disconnect

Measure lead-to-lead resistances

Caveat, if some of the leads have smaller diameter wires, those might be for and encoder or tach, so treat those carefully.


From that info, we can probably figure out what you'll need to make those motors go.

---

Larry Pfeffer,

My blog about building repstrap Cerberus:
[repstrap-cerberus.blogspot.com]

You more an likely have servo motors on this beast which might have controllers that simulate a stepper motor interface.

That would be the interface point to look for as directly controlling a servo motor is a much more difficult task.

Hi

I am working on seiko dtran xm5000 for my final project. I am currently using teach termial to communicate. I would like to know how can I use a computer to communicate with it. I mean the language ant etc. Kindly help me with this. My email id is vikasreddy9@gmail.com.
thanks