Rebuilding Industrial 3D Printer

Hello all,

Noobie here. I'm looking for some advice and by the looks of this great forum, I think I've come to the right place.

I got my hands on an older 3d printer - Statasys Dimension 1200. The trick is that the guts (print head and controller) were sent back to the leasing company. However, all of the motors and hardware still remain. Before I haul it home, I thought I'd ask the experts here to see what kind of reverse engineering nightmare I'm looing at to get this thing running. Here are the tasks that I see to make this work:

• Buy an off the shelf extruder and bolt it (with a customized plate) to the existing carriage
• Buy an off the shelf control board
• Customize a RepRap profile to match the hardware

Here is what the inside of the machine looks like (minus the extruder head). The extruder is on an x-y plane and the build table moves in z-direction.



I'm unaware of how customizable the RepRap profiles are to make things work with the existing hardware. If someone could clarify that, I would appriciate it.

Also, does anyone have any recommendation on hardware (extruder and control boards)? I'd like the option of using support material if possible. It's what I'm used to when I used this machine when it worked.

Thanks for any help you can send my way.

If it uses stepper motors, then adding on one of the common RepRap controllers should work fine. The common firmware versions are very configurable and should be able to handle what looks to be a basic Cartesian mechanism. Your biggest headache will probably be sorting out the wiring harnesses.

... the mechanical setup is common for RepRaps, you only need to identify the wiring and needed voltage for the endstops and if the motor drivers are common STEP/DIR types - then any RAMPS or other Arduino with BOB would be suitable ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Also check out what the voltages and amperages are for the stepper motors. The A4988 stepper drivers are maxed out at 2A but need active cooling fans and cooling elements on them in order to supply those 2A. Above 2A it would be good to check out other stepper drivers. I've used 2.5A stepper motors successfully with the A4988 drivers but anything above will need a different approach because the motors will lose a lot of torque if underpowered.

---

http://www.marinusdebeer.nl/

... my understanding was, that only the controller is missing, so the motor drivers should be present ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
LoboCNC
If it uses stepper motors, then adding on one of the common RepRap controllers should work fine. The common firmware versions are very configurable and should be able to handle what looks to be a basic Cartesian mechanism. Your biggest headache will probably be sorting out the wiring harnesses.

Thanks LoboCNC.

I'll probably start with trying Sprinter

Quote
VDX
... the mechanical setup is common for RepRaps, you only need to identify the wiring and needed voltage for the endstops and if the motor drivers are common STEP/DIR types - then any RAMPS or other Arduino with BOB would be suitable ...

Thanks VDX. I'll probably start with with a RAMPS kit because I'm pretty familiar with Arduinos. Do you see anything wrong with this idea?

Something like this: SAINSMART

Quote
Ohmarinus
Also check out what the voltages and amperages are for the stepper motors. The A4988 stepper drivers are maxed out at 2A but need active cooling fans and cooling elements on them in order to supply those 2A. Above 2A it would be good to check out other stepper drivers. I've used 2.5A stepper motors successfully with the A4988 drivers but anything above will need a different approach because the motors will lose a lot of torque if underpowered.

Ohmarinus, that's a good point. I can just see one stepper motor right now:

Lin engineering
071912
5718M-24P-03RO
2.43A

Looks like it's a NEMA SIZE 23 | Step Size 1.8°

So I'm pushing the 2A like you did. I would assume it would complicate things to move to a non-standard driver that has more than 2A capacity. What would you do?

Quote
VDX
... my understanding was, that only the controller is missing, so the motor drivers should be present ...

I dont have the experience to reverse engineer motor drivers to be compatable with something.

If you're in luck they will have a step/dir/enable type interface which is very easy to work with together with common reprap hardware. Assuming the voltage levels are compatible, it could be as easy as hooking up wires to the correct terminals. Snap some pictures of the motor drivers and someone on here will probably be able to help you out. (It could be alot messier too with some type of propritary interface, so I don't want to get your hopes up too much, but it's definetely worth checking up)
/Andreas

Sounds like that amperage is good, however, I am not sure about the voltage?
Maybe someone else can clear that up?

---

http://www.marinusdebeer.nl/

... can you post a picture of the drivers and their terminals?

I've mounted different RepRap-style and industrial motor drivers to Gen3 and RAMPS by simply connecting the corresponding pins.

All of the industrial type inputs were optocoupled - mostly with 5V/TTL-input, so you simply connect them ... sometimes with 24V-inputs -- but here I'm simply replacing the input resistors for smaller types were sufficient too ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

VDX,

Thanks for your input. I am finally getting to bring this home today. I'm going to tear it down and do some investigation on the existing hardware and report back.

This something I am curious of. Can you post pictures of the inner mechanical parts? I would love to see the guts of that thing.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Quote
jaguarking11
This something I am curious of. Can you post pictures of the inner mechanical parts? I would love to see the guts of that thing.

Will do! I'm tearing it down tonight to get a good peak at what electronics are left. I'll take pics during the process with details of the mechanical guts too. The rigidity of the mechanicals is what I'm stoked about to start with as a base. I would really like to reuse the motor drivers if possible though too because the motors are the high end of the 2.5A limit of some chips.

There are polulu style drivers for ramps that supposedly can handle close to 3A. 1.5A per phase. They might be worth buying if the stock drivers don't work so well. I believe they are the DRV8825 drivers. Correct me if I am wrong though.

I doubt the stock electronics on that printer have segregated drivers. Reusing them may require a bunch of EE experience and a whole lot of circuit tracing and redesigning.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Quote
jaguarking11
There are polulu style drivers for ramps that supposedly can handle close to 3A. 1.5A per phase. They might be worth buying if the stock drivers don't work so well. I believe they are the DRV8825 drivers. Correct me if I am wrong though.

I doubt the stock electronics on that printer have segregated drivers. Reusing them may require a bunch of EE experience and a whole lot of circuit tracing and redesigning.

Good point. I will be asking for help with the electronics decision. It has a lot of cool features that I might or might not want to reuse. For instance, it has an autofeeder for the filament. You plug in a cartidge style into 1 of 2 slots and it pushes the filament up the tubing into the print head. then the print head pushes it through the hot end. The feeder is like a bowden feeder but only for loading.

It also uses a heated chamber instead of a heated bed. So calibrating that will be interesting.

what size is the bed? can you not just add a heat bed to it and then you will have both?

if you are using those motors you will want to use the DRV8825 drivers not the lower rated ones that most RAMPS users use, they are designed for 1.5 Amp before active cooling and 2 Amp with.

given the enclosed chamber, having the feeder sounds like a good idea, however would you be able to replace it entirely with a Bowden? I'd imagine this would make calibration easier.

have you thought about replacing the motors to work with something smaller? I would think it would make calibrating steps per mm easier if your using standard motors? however this would be more along the lines of making a repstrap out of your industrial printer than re-building it.

Here are some pics of the components. Sorry the pics are not correct orientation. I'll figure that out later on imgur

[imgur.com]

The biggest thing that I'm worried about is that the stepper motor for the extruder is actually a 24v dc motor with optical encoder. That scares me a little.

I like the extruder wheel design. You can always replace the DC motor with a nema17 or nema23.... As for the system running on 24v, there are some boards that handle that directly, and some you have to modify. There are guides to modifying ramps to run on 24v.....

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

I'd really like to keep the DC motor/encoder setup if I could. It is a strong motor (265 in-lbs) with a very fine resolution 1000 ppr. It's not the norm and it may cause some extensive troubleshooting, but reverse engineering this thing work is 90% of the fun for me. In my opinion, the only reason to get rid of the DC motor/encoder is because it's not "supported" by normal hardware/software/firmware. Would you guys agree?

I think I'm going to need the extra torque to be able to pull the filament a long way up through the tube

EDIT: Maybe I could add an arduino dedicated to the DC motor/ecoder that acts like a translator from stepper input to constrol the dc motor/encoder. (IF doing that on the RAMPS board is too complicated)

Edited 2 time(s). Last edit at 11/21/2014 10:21AM by TFinleyosu.

Quote
TFinleyosu
I'd really like to keep the DC motor/encoder setup if I could. It is a strong motor (265 in-lbs) with a very fine resolution 1000 ppr. It's not the norm and it may cause some extensive troubleshooting, but reverse engineering this thing work is 90% of the fun for me. In my opinion, the only reason to get rid of the DC motor/encoder is because it's not "supported" by normal hardware/software/firmware. Would you guys agree?

I think I'm going to need the extra torque to be able to pull the filament a long way up through the tube

I have heard that some drivers out there turn the DC motor/encoder into a stepper compatible solution. It may work. Honestly it is an ideal solution if it did work. However finding how it works may prove to be a challenge. There has been research done in the possibility of using a DC motor and encoder setup. However it seems there is not much headway for it. It seems pretty difficult to attain accuracy.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Quote
jaguarking11

Quote
TFinleyosu
I'd really like to keep the DC motor/encoder setup if I could. It is a strong motor (265 in-lbs) with a very fine resolution 1000 ppr. It's not the norm and it may cause some extensive troubleshooting, but reverse engineering this thing work is 90% of the fun for me. In my opinion, the only reason to get rid of the DC motor/encoder is because it's not "supported" by normal hardware/software/firmware. Would you guys agree?

I think I'm going to need the extra torque to be able to pull the filament a long way up through the tube

I have heard that some drivers out there turn the DC motor/encoder into a stepper compatible solution. It may work. Honestly it is an ideal solution if it did work. However finding how it works may prove to be a challenge. There has been research done in the possibility of using a DC motor and encoder setup. However it seems there is not much headway for it. It seems pretty difficult to attain accuracy.

The accuracy is a benefit IMO. I'd be able to easily fine tune the feed rate with the feedback from the encoder. I've worked with motor/encoder before and it was excelled for positional/speed accuracy. If Stratasys decided to go with this setup in a $50,000 machine, I think there is a good reason.

EDIT: I'm nervous that the board I go with is not fast enough to deal with the interrupts of the encoder and deal with everything else. That's why I'm thinking of having a separate board to dedicate to this sub-system.

Edited 1 time(s). Last edit at 11/21/2014 10:31AM by TFinleyosu.

Quote
TFinleyosu

Quote
jaguarking11

Quote
TFinleyosu
I'd really like to keep the DC motor/encoder setup if I could. It is a strong motor (265 in-lbs) with a very fine resolution 1000 ppr. It's not the norm and it may cause some extensive troubleshooting, but reverse engineering this thing work is 90% of the fun for me. In my opinion, the only reason to get rid of the DC motor/encoder is because it's not "supported" by normal hardware/software/firmware. Would you guys agree?

I think I'm going to need the extra torque to be able to pull the filament a long way up through the tube

I have heard that some drivers out there turn the DC motor/encoder into a stepper compatible solution. It may work. Honestly it is an ideal solution if it did work. However finding how it works may prove to be a challenge. There has been research done in the possibility of using a DC motor and encoder setup. However it seems there is not much headway for it. It seems pretty difficult to attain accuracy.

The accuracy is a benefit IMO. I'd be able to easily fine tune the feed rate with the feedback from the encoder. I've worked with motor/encoder before and it was excelled for positional/speed accuracy. If Stratasys decided to go with this setup in a $50,000 machine, I think there is a good reason.

EDIT: I'm nervous that the board I go with is not fast enough to deal with the interrupts of the encoder and deal with everything else. That's why I'm thinking of having a separate board to dedicate to this sub-system.

If your willing to write some code then I would go with a 32bit solution right off the bat. Azteeg x5 comes to mind. Right now its the only board out there with any major use and gaining traction. Although I would prefer a rams/due combo, although slower I like the modularity a ramps on microcontroller offers. The price is also nice. An off brand rams modified to run on due is about 30usd, add an off brand due and for about 55usd you got yourself a 32bit solution. The problem is the firmware with the ramps/due combo. azteeg x5 is more supported as of right now, although they are talking about porting Marlin on due and optimising it for the due instead of a straight up port that would offer no real benefit from the mega.....

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Can someone help me confirm that this is just an in-line filter for the 110V input (bottom board)? There is a switch in there too.



Edited 1 time(s). Last edit at 11/21/2014 02:52PM by TFinleyosu.

Follow up...

^^ That was a filter with a circuit breaker, which I am using.

I bought a 12V 30amp power supply that I connect to this circuit breaker, which works well.

I am using the Azteeg X3 with the DRV8825 drivers and decided to power the stepper motors at 12V instead of worrying about trying to go 24v. I just got them running with pronterface and finally figured out the calibration. The steps per mm were pretty high: X:126 Y:267.5 Z:1266, which I like for resolution. The motors are nice and smooth using pronterface. I have not figured out the max speed/acceleration.

Anyways, I thought I'd follow up on the thread for those who helped point me the right direction so long ago. Having a kid really effects your free time...

Edited 1 time(s). Last edit at 06/24/2015 06:13PM by TFinleyosu.

hi, i'm very interested in your topic as i have pretty much the same thing to do (as school work) ^^

we have to make a Stratasys Dimension 1200 work back again using an arduino (we have all the components thought, just the head cover that has been broken before)

so i have some questions :p (you will be credited for it)

what did you succeeded to control with the arduino? (stepper motors, lights, fans, radiators, filament motors, ...)
what did you had to add to make the printer work? (if it work now)
what did you changed in the firmware to adapt the ramp card to this printer? (stepper motors pitch, temperature sensors, ...)

more random questions (if you have an idea, it would be cool to understand, otherwise it's not important for what we do) :
why the Z motor isn't right on the axis of the screw? (there is no speed reduction and the space under the screw is empty)
what the 2 pins per cartridges of filament are made for? (there is a tiny PCB in the cartridges with 2 contacts as well, but we couldn't find how to open it)
what the sensor at the back of the head is made for? (you don't have it, but there is the 2 sensor at the front of the head for X axis ends, but there is also this one with a spring-loaded piece under that rise a wall in the sensor when pushed by the top side floor of the printing enclosure)

it would be nice if we could help each other making this printer work
here are the pics with all components if it can help you (names are in french thought) : [goo.gl]

PS : could you send us a picture of how is wired the front switch, because it got unplugged when we removed the front panel but we don't know where which pin was plugged ^^

Edited 1 time(s). Last edit at 09/24/2015 12:33PM by Nagby.

With the arduino and a stepper motor board (SainSmart CNC Router Single 1 Axis 3.5A TB6560 Stepper Stepping Motor Driver Board) I was able to test movement of each stepper motor and get the steps per mm. I then went in a separate direction from the arduino with the Azteeg X3 that gave me full control of all components. I then uploaded Marlin firmware to the Azteeg, calibrated it and it works great. I didn't have to change much in the firmware because I used an off the shelf heated bed instead of the heated chamber. I also used a boden extruder with E3D_V6 hot end wit a custom mount on the printer carriage. I really didn't use much of the original hardware other than the mechanical components. This is probably not much help.

If you want to use the heaters in the heated chamber you'll have to customize the firmware or create yourself a solid state relay board to turn the AC heaters on and off from the RAMPS / arduino board.

If you want to stay on the arduino board, you'll need to buy a RAMPS shield or reverse engineer it to create one yourself.

I assume they were just used to belts for motion transfer rather than using a error-compensating coupler.

The printer cartridges PCB their way of making sure you pay $$$$ for their filament. There are ways to re-program their PCB, but I would avoid trying to recreate this feature. It will allow you to use cheaper filament and just refill their cartridges.

I believe the sensor is to confirm that the switch between build material and support material was engaged properly.

Let me know what else you need help with. I think our tasks are slightly different. I just wanted to get a 3D printer working from this shell and mechanical components and not recreate all of the options that this machine had.

that is exactly our task ^^
our school have this printer since ~2008 and it had a software problem but everything physical is working
and a projet from last year was about a reprap printer
so we have to make the old expensiv printer work with cheap and open source reprap components (arduino and ramps at least)

but if we have time we could try to get more features (but we have to know what to buy pretty early)

+ our formation isn't about code or electronics, it's mainly mechanical, so you know where we lack skills

so, as i said, first thing we need is to find what we have to buy (for early november)
and then beginning of next year we will actually build the thing

we will also have to understand as much as possible in the printer mechanism and about 3D printing in general to present it later (that is work we will do ourselves of course ^^)

about extra features, easiest one to add i guess would be to use the light rails in the printing enclosure
then, a level higher would be using the original head with double filament and use the cleaning brush (i don't know when it is used by the printer normally working, maybe only when changing filament and end of print)
and last would be the heating system, but because that would make the printer more "dangerous" and loud it might be better to stay with a heating bed


thanks a lot for your quick reply

A hard thing you are going to have to figure out is how to control the extruder motor. It is a DC motor with encoder. All of the firmware options are stepper motor control. It's going to be hard to write your own if you are not a cs person. So find someone who can help you write that control into your firmware.

I'd buy a cheap ramps board and some stepper drivers capable of 2 amps. The install marlin forward, calibrate your stepper motors and then confirm with moving the head around using the GUI in pronterface. That should be step 1. Then step 2 is the DC motor for extruder control.