Woo Hoo!! New Toy!! Stratasys Dimension 3DP

Because they are 8 wire motors, they can wired in series or parallel. I have my motors wired series so that they draw less amps and used 24v instead of 12v to maintain good speed. Here's a reference: http://www.xylotex.com/Windings.pdf

 

The TB6560 is 3.5 amp controller is commonly used for CNC and is available at a variety of sources for cheap: CNC Router Single 1 Axis TB6560 3.5A Stepper Stepping Motor Driver Board 3D Printing, Arduino, Robotics | Sainsmart It may have problems with the pulse width of the controller so check it out first. I know that using this kind of drive through a breakout board (which may be the main problem) is not recommended with Marlin firmware. Of course, if you want to go for quality, get a Gecko drive: http://www.geckodrive.com/geckodrive-step-motor-drives/g251x.html

 

Anytime I feel a little too smart, I just come to this thread to let in a little reality. I have no idea what tug posted above means, but I give you this picture

 

What I've gathered through my research is that in theory, most any controller *should* be able to run a more powerful stepper driver such as the Geckos. There are three basic inputs typically required from the controller to drivers, they are step, direction, and enable - for each axis. There are more connections necessary for the driver to function, but they are power inputs, etc. That being said, getting those inputs can be somewhat difficult. One some boards, such as Smoothieboard, these signal pins are directly broken out on the board itself, easy peasy. On the Duet, they are available, but I have no idea which pins on the expansion header they are yet, the documentation isn't very good, and I've just started looking into it. With RAMPS, and other boards that can accept the individual Pololu driver boards, finding the proper signals should be pretty straightforward, it is documented on a number of the driver boards. One problem with the RAMPS, and probably many other boards, is the Marlin firmware sends such an incredibly short pulse to the driver, that many drivers cannot properly interpret it.

I really like the looks of this gecko driver board, has 4 stepper drivers on one board, but Is a bit spendy: http://www.geckodrive.com/geckodrive-step-motor-drives/g540.html

I also came across these that are supposed to handle 10A: Powerlolu Stepper Motor Driverup to 10A-100020

 

There is a possibility of this... I'm still trying to come to terms with it. My only major issue is using it to run more capable stepper drivers. I know it's possible, just not blatantly obvious as to how to do it.

Edit: It is indeed finally Delta capable.

 

The Geckos will run on just about anything that outputs step and direction signals. They don't even care about the short pulse from Marlin. I have Gecko G320x servo drivers on my CNC mill and I had them running off a Mega 2560/RAMPS combo - just had to bypass the parallel port breakout board. The step pulses on the Geckos only need to be 1-2 microseconds long and the direction pulses only need to be 200 nanoseconds. They are a completely different class than the cheap single chip drivers.

 

Thanks Mark, this is good to know!! I don't know if I can cut loose with almost $300 for drivers...... It's probably worth it though.

 

You do pay for the additional capability. If you need the capability though, the Geckos are the best bang for the buck. You can try some of their distributors for slightly better prices. The name has changed, but this is the company that I got my servos and servo drivers from: Gecko Drivers | Automation Technology Inc The best prices are when you are buying a set that includes stepper motors and drivers together.

 

On the 8 wire motors..... It looks like they can be used as either bipolar or unipolar, either of which can be wired as series or parallel. From what I've looked at, it seems my best plan of action is to go bipolar, series. Anyone have any thoughts on this?? I did find the wiring schematics from the manufacturer showing what color wires go where depending on the desired configuration.

 

I dated a series of bipolar ladies in my youth...not recommended. J/K

 

Had some time today to un-bury this thing so I could start trying to sort out what is what. Here are the observations from todays scouting mission:

The following have been verified to be 24v DC: Chamber lights, all fans (chamber heat circulation, power supply cooling, and a fan that blows air through a tube leading to where the extruders mount), solenoids that engage the drive rollers to load filament from the cartridges to the extruders. I have not verified the voltage of the drive motors that load the filament yet.

The current stepper motors are wired from the factory as unipolar, in series.

The endstops for each axis are normally closed prox type switches. These have one wire to each "finger", and open when interrupted by a piece of metal mounted on each axis moves between the two fingers. I will have to try to figure out what the firmware expects from these... I think it expects them to be normally open, but I haven't verified this yet.

There are also some prox type switches for overtravel on the opposite end of each axis. These are sort of peculiar. They are the same basic configuration as the home endstops, but there are two wires to each finger. I did not do any testing as of yet as to whether these are normally open or normally closed, or if the two wires that go to each finger have continuity between them as they exit the switch. I don't have any idea how these were incorporated into the original system. My control board only has 3 endstop inputs, so these may be left out of the system. I *think* I can rely on the firmware to limit axis travel??

I can not figure out what voltage the heating elements are, there are no markings on them anywhere, and the original leads were cut. Anyone have any suggestion on how to figure this out?? I thought about running 24v to them to see if they heat up or not....

 

Measure the resistance of the heating elements first and calculate how many amps they will pull using I=V/R. Then calculate power P=V*I. For the chamber heater I don't have a comparison, but the hot end heater generally uses about 40 watts.

 

Thanks Mark, that will give me a much better picture of what I am dealing with! I will try to do this during the weekend sometime.

I am still researching how to incorporate the Duet board... there just isn't a lot of CLEAR information out there. That is one certain advantage to the ramps board, there are so many in use that the support from current users is immense, especially on the firmware side where I am so challenged.

Biggest challenges I see to incorporating the Duet board so far:
1) Figure out how to interface bigger drivers, should be doable, but I'm not there yet.
2) Figure out how to alter the firmware to properly process the end stops. Mine are normally closed, the firmware expects normally open. As a last ditch effort I can replace the current endstops with microswitches, but I really don't want to do that - the factory ones are already there, and if I can use them, the install is that much cleaner.
3) Figure out how to incorporate the existing chamber heaters. This should be pretty easy. It looks like the factory had an overheat interrupt installed as well as a thermocouple for temperature feedback. I would like to keep both of these, the hardest part is interfacing the thermocouple, looks like bare minimum I would have to add a separate board to interface it, and do some firmware mods.... uggghhh... firmware mods!!!!!

I think that's about it for now........

 

The Duet is open hardware, so the schematics and board files are available at T3P3/Duet · GitHub I took a quick look at the schematic and noticed a couple of things. First, it doesn't look like the step and direction signals are broken out to one of the headers, so tapping them will be a bit of work. Second, it only has native support for two temperature inputs and two heater controls, so the heated chamber will take some more re-programming. It does look like extra PWM and GPIO pins are broken out on one of the headers, so at least connecting the heated chamber controls should be easy. In fact it might be easier to modify the pin configuration in the firmware to use the extra GPIO pins on the header for your step and direction signals.

I swear I saw a thermocouple interface board that did not require any firmware changes to work. It just hooked into one of the thermistor inputs. I'll look around and see if I can find it again.

Changing the firmware for the types of end stops and the thermistor values should be fairly easy. Although I admit that that was the part that concerned me most when I started on my journey to build a printer. I did eventually get comfortable making the changes and updating the firmware.

 

You should be able to use one of the following boards for the heated chamber temperature input, assuming that it's a K type thermocouple: Analog Output K-Type Thermocouple Amplifier - AD8495 Breakout ID: 1778 - $11.95 : Adafruit Industries, Unique & fun DIY electronics and kits I think you would have to remove the 1k voltage divider resistor on the board and then build a temperature to voltage table using the provided formula.

 

Thanks Mark!! I appreciate you taking a look at it. The more I research the different control boards and explore the printer hardware I have, the more I think I should have went with the Smoothieboard.

Pros to going to Smoothieboard:
1) Step, Direction, and Enable signals are already broken out for each stepper driver (no header, just through holes to solder wires into)
2) The documentation is heads and shoulders above the Duet, they have put a lot of work into typing up explanations on the board layout and how to hook up everything.
3) Support and improvements/ additions *seem* to be more consistent.
4) They also sell a thermocouple board, and I think I saw somewhere that it is documented how to interface it and modify the firmware for it - I could be crossing info from some other board on the interface and firmware part though.
5) Has inputs for both max and min endstops.
6) Has 4 thermistor inputs.
7) Can Drive up to 5 axis from the main board.

Cons:
1) I have to buy it.
2) No current support for plug and play touch screen display. If this were available, I would probably buy the board today. There is at least support and availability of plug and play LCD display.

Pros to staying with Duet:
1) I already have it, as well as the Duex4 expansion board.
2) PanelDue (touchscreen) available as nearly plug and play. Current firmware supports it. This is pretty important to me since this is not a desktop printer, and this would allow most all functions that would typically require a pc to be done at the printer.
3) I ALREADY HAVE IT.
4) Better web interface, not as important if the touch panel is added.
5) Currently with the boards I have there is support for 8 (I think) axis + extruders.

Cons:
1) Limited thermistor inputs, and there is a bug in the board which causes erratic reading on the temps for the 2nd hotend.
2) Only 3 endstops - not really a big deal.
3) Documentation is pretty poor, especially irritating since I am not real electronics friendly.
4) No breakouts for steppers, which leaves me wondering how they interface the additional 4 drivers on the expansion board through the expansion header....... those signals have to be there somewhere..
5) Would *probably* be more difficult to incorporate thermocouplers.

All this being said, I guess I have a decision to make before I go much farther. One of the main reasons I went with the Duet/ Duex4 combo was because I thought I would have to drive 2 steppers for each extruder due to the way the machine was set up (in my mind) but on farther examination this isn't the case.