Stratasys 3D Printer Hacking

[Ross Bochnek]

We've been planning to revive our big Stratasys 3D Printer with a RAMPS board. The "AMPS" in RAMPS stands for; Arduino Mega Pololu Shield.  The "R" in RAMPS stands for RepRap- the DIY 3D printer we have, although the RAMPS board is suitable for controlling all kinds of other CNC applications.  The plan is to drive the native stepper motors already inside our Stratasys 3D printer with this system.

GOOD NEWS, EVERYONE!  We're getting a RAMPS board donation- more on that when it's in hand.  I hear it can also drive the support material print head.

This board combines with a Mega or clone, and 4 Pololu stepper motor driver carriers.  There seem to be several drivers available, and I don't know if it depends on the version of RAMPS board we are getting.  There is a link on the Lulzbot site and the RepRap site to these: https://www.pololu.com/product/1182  Each one of the required four seems to be about $6 each, but the price break at 5 pieces would give us an extra one for only ~$1 more (just in case).

I will try to secure this RAMPS board soon, and an Arduino Mega clone if we need one.  We can soon figure out which Pololu Stepper Motor Driver Carriers we would also need to order, and let's transplant a brain into this wonderful printer.  Let's keep this revitalization in mind when laying out the "Clean Area", and plan to include the Stratasys on our "OMFG It Kinda Works" list- hopefully before Summer 2016.

[Ross Bochnek]

This is basically the same Pololu Stepper Motor Driver Carrier as listed in my previous post, but are sold in denominations of 5: https://www.pololu.com/product/2975  It merely requires that we provide and solder on our own header pins from our existing cache, and will be fun and easy to do.

[Sam Harmon]

Pretty sure I have an Arduino Mega clone I'm willing to donate to this cause.

Also, you can find clones of those pololu drivers (or similar stepper drivers with the same pinout) on ebay for super cheap. I got a set of the DRV8825 ones for the Shapeoko a while ago.

On Dec 27, 2015 1:50 AM, "Ross Bochnek" <interac...@gmail.com> wrote:

This is basically the same Pololu Stepper Motor Driver Carrier as listed in my previous post, but are sold in denominations of 5: https://www.pololu.com/product/2975  It merely requires that we provide and solder on our own header pins from our existing cache, and will be fun and easy to do.

--
You received this message because you are subscribed to the Google Groups "Makers' Alliance" group.
To unsubscribe from this group and stop receiving emails from it, send an email to makersallianc...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ross Bochnek]

Sweet, Sam- thanks for the Mega donation and shopping leads!  Those appear to be about $2 each.  Do the drivers depend upon the RAMPS version?  I'll report on what version Jason is offering asap.

[Ross Bochnek]

If it's a RAMBo board, it may already have the Arduino Mega and motor drivers on-board.  I hope to find out by Tuesday.

Athough it would be nice to get some of the Stratasys's native control panels to function, they'd probably have to be hacked, so I'm guessing the basic plan is to control everything from a PC or SBC like a Beagle Bone Black.  

Are we going to try to standardize as much of our CNC software as possible throughout our shop, to make it easier for people to run the various machines? I realize there are several possibilities for generating G-Code, but the plan is to try to use LinuxCNC as the final GUI to control each stepper-based machine {Rep-Rap, MaxCNC Mill, Shapeoko 1, BigCNC}?

Or, will there be other GUIs to run some of the stepper-based CNC machines?  [unlike the Seiko Robot, which uses Raspberry Pi/Raspbian and Gertbot/Python because of its existing brushed DC motors].

[Sam Harmon]

> On Dec 27, 2015, at 3:04 AM, Ross Bochnek <interac...@gmail.com> wrote:
>
> If it's a RAMBo board, it may already have the Arduino Mega and motor drivers on-board. I hope to find out by Tuesday.
>
> Athough it would be nice to get some of the Stratasys's native control panels to function, they'd probably have to be hacked, so I'm guessing the basic plan is to control everything from a PC or SBC like a Beagle Bone Black.
>
> Are we going to try to standardize as much of our CNC software as possible throughout our shop, to make it easier for people to run the various machines? I realize there are several possibilities for generating G-Code, but the plan is to try to use LinuxCNC as the final GUI to control each stepper-based machine {Rep-Rap, MaxCNC Mill, Shapeoko 1, BigCNC}?

LinuxCNC is used as the linux distribution for the MaxNC, but the LinuxCNC GUI isn’t used to control it. The MaxNC is running the Grbl firmware on its Arduino, and the software used to connect is called Universal Gcode Sender.

I’m not sure I want to strictly standardize on specific software for all the CNC gear at the space- there’s a lot of interesting stuff going on in that arena, and it’s worth investigating a number of options. Right now the plan for the Shapeoko is to use Grbl on the microcontroller, but a BeagleBone Black as the computer (possibly using Chilipeppr as the frontend instead of UGS).

At some point we’ll probably upgrade the MaxNC to a TinyG or some 4-axis fork of Grbl in order to support that rotary axis we have.

Also, if something interesting shows up with working software of its own, I’d rather stick with that if possible instead of having to spend a ton of time reworking it.

Sam

[Ross Bochnek]

GratitudeBE to Jason for his in-process donation of a RAMBo 1.3L board!

http://reprap.org/wiki/Rambo_development

We don't need an additional Arduino or stepper drivers.  It can run a heated bed, end stops, and probably both print heads [we'll have to check how the support material head works, but if it involves a stepper, we should be good].

It can use between 12 and 35 VDC.

We'll have a range of Firmware options too.

This is not only an incredible windfall, but it seems to be the perfect board for Frankensteining the Stratasys to life with lots of features!

[Ross Bochnek]

It was great to literally dust off the ol' FDM tonight and swarm around it.  We opened panels and probed around for stepper motor and heater specs, and traced some wiring.  We uncovered one stepper motor label; indicating that the RAMBo 1.3L should provide sufficient current.  We also found screw terminal blocks for the stepper, so if the same motors are used for other axes, we should be able to determine their coil pairs without extensive testing.

Here's a photo of the stepper from our machine, taken tonight:

I filtered what I thought to be the model number through the interwebz, which suggests we have a: VEXTA MODEL PK268-01A 2-PHASE 1.8 DEG PER STEP

http://www.ebay.com/itm/VEXTA-MODEL-PK268-01A-2-PHASE-1-8-DEG-PER-STEP-STEPPER-MOTOR-STEPPING-/221730081593

Here are some images of the terminal block wiring from our machine:

The support material seems to be P400 ABS thermoplastic.  I think it prints like the main material, but is easy to chip away from the model after printing.  MSDS from Stratasys: http://usglobalimages.stratasys.com/Main/Files/MSDS/P400-ABS/SDS_US_ABS_P400_English_US.pdf?v=635869915907181104

Alternative: http://www.argylematerials.com/argyle-abs-p400-refill-compatible-with-dimension-768/

It appears to be expensive stuff.  Argyle Materials suggests that one reason for the cost is that the spool cartridge contains an EEPROM that identifies the material type to the machine.  If this is the case, it seems that we might be wise to wind replacement filaments for this printer onto the spool cartridges that came with the machine; although the fact that we're replacing the brain AND its firmware in the machine may prove this moot.  We may have an opportunity to enter material information into our system some other way.  Here's an image of the P400 support material's packaging:

[Ross Bochnek]

Update: Ian got the proper software, installed it on a laptop, and we reassembled the printer.  

After trying to establish a serial connection between the computer and the printer, we had a thought about the new controller hardware we installed into the machine.

We compared the labels on a pair of ROM chips between the old and the new controller, and realized they're different.  We now plan to swap the old IC chips into the new controller board, which should have the firmware we need to communicate between the computer software and the various heaters and motors in the printer.

[Ross Bochnek]

If we can get this plan to work, we can:

save the RAMBo 1.3L  board and our newly salvaged 5/12/24 VDC PSU for the big CNC router project!

[Ian ‍]

We made a small semblance of progress tonight.

Using the Compaq laptop (Running windows 7) we managed to achieve communication with the Modeler (What the software calls the Printer) And were able to do several things:

Firstly, Computer was able to handshake with the Modeler and attempted to calibrate, however it failed.. More on this later. 

Sam was able to get the modeling software (Insight v4.2) to acknowledge the Modeler's existence

Sam then downloaded a simple cube model from thingiverse and sliced it with Insight

When sam tried to print the model, The machine attempted to operate (Woohoo) But the grease on the shafts, rollers, gears, and corkscrews was so old and gummy. it restricted almost all movement and we elected to power it down before any harm was done. 

I sprayed everything liberally with WD-40 and PB-Blaster in an attempt to loosen stuck parts, Hopefully by Saturday (at the earliest) The parts will be moving free enough to operate. 

In conclusion... IT WORKS!!!!!...........Kinda

[Robert McTrusty]

WOOT

Could be the grease was frozen

[Sam Harmon]

Last night’s revelations:
-getting it to start in the “motors energized” state is random at best. Once in a while I can get it there by fiddling with certain connectors (but not always), but there’s no rhyme or reason to it beyond that.

-Once it’s in the “motors energized” state, I can’t do too much via the front panel (reset will work in that state, but that’s about it). Once, It did light up the LEDs on the up/down arrow keys, but still no dice.

-the little sub-board on the back assembly (with the surface-mount PIC) is actually a Parallax BASIC Stamp (the 1997 edition of this one in particular:https://www.parallax.com/product/bs2-ic), for whatever that’s worth. Not entirely sure what it’s doing, but that’s what it is.


Sam

[Sam Harmon]

Today’s progress:
-Learned a new error code: “Env Temp Error”, which is not in the printed documentation, but a search online indicates it’s related to the envelope temperature changing rapidly (makes sense). It also required rebooting the unit in order to clear it.
-Discovered the IDC crimp connector on the cable that runs between the FDM’s display/keypad and the back pcb was failing. Have it held together a bit better with some hot glue until a replacement 26-pin connector can be ordered.
-At one point the x-axis motor did seem to be moving the extruder carriage slowly and erratically to the left.
-Other than that, no real progress

Things to look at next:
-test continuity on the 6-pin to 9-pin connector between the front panel and the motor driver board
-test continuity on the 20-pin IDC connector between the front panel and the motor driver board
-test continuity on the 10-pin IDC connector between the front panel and the back PCB (the blue-connectored one with the DB9 interconnect in the middle)
-take apart the front panel panel boards to see how they work (so far the only button that’s worked is the reset button)
-Does anyone have an RS232 tester? (i.e. lights up LEDs depending on what’s on the line) Or should I get this rather deluxe version: http://www.amazon.com/RS232-DB25-BREAK-OUT-BOX-TESTER/dp/B005H46XS2 - with something like this we can see if the Insight software is actually doing something useful.


Sam