The concept was to have them individually moveable and using only 1 nozzle at a time during a single print. But being able to switch to a different nozzle/color during that same print.
My mind turns to mud when trying to hold the idea of 4 independent krakens printing at the same time.
What are the challenges and benefits of that configuration during a single print?
If it helps reduce print time I am all for it.
Having been down the road of using separate hot ends/nozzles on my aged Mendel tricolour, and now using a Diamond mixing hot end, I feel qualified to have a stab at answering that. As well as the loss of X and Y travel which you've obviously thought about, I'd say there are two main issues that you'll be faced with. The first is getting all the nozzles at precisely the same height (within 0.1 mm or better). A single Kraken is pretty much a nightmare so 4 is going to be a real challenge. The second big issue is fighting ooze from the unused nozzles. The usual approach is to set a lowish standby temperature then switch between active and standby but you have to wait for the next nozzle to heat up. You'll need some sort of wipe mechanism but that shouldn't be too difficult.
Have you considered the idea of fitting two Diamond 5 colour hot ends? That would give you the possibility of using 10 colours with a lot less complexity. That may be too many extruders, so if you can settle for 6 colour, you could use two 3 colour Diamonds. Of course, the ability to mix filaments in theory gives you many more colours, but the mixed filament comes out like stripey toothpaste. This can be negated by using transparent filament but not every object lends itself to that. There are other issues with Diamond hot end so it's not as straightforward as it sounds. One that I have learnt recently is that PLA does not like to be kept at print temperature for half an hour or more without moving, but then you'd likely have the same issue with the Kraken unless you set a really low standby temperature, which will of course make the print time longer while it waits to heat up. At least with the Diamond, all the filaments are loaded so there is no waiting for one tool to heat up while another cools down, and of course the single nozzle means there are no Z alignment issue or X offsets to deal with.
There are 7 sets of step/dir/enable for external drivers on the expansion (5) and LCD_CONN (2) headers on the Duet. We have an expansion breakout board specifically designed to interface with large external drivers :
External drivers that do not need 5V logic will work connected directly to the Duet.
The information is available here on the wiki:
Swapping the X and Y motors, either the connections or in software, switches the coordinate system between left hand and right hand. In a right hand coordinate system (which is what slicers assume), the +Y direction is rotated 90 degrees anticlockwise from the +X direction as seen from above.
@davidbitton, you are probably not raising the carriage high enough or fast enough after the first probe. The probe height is the H parameter of M558, and the travel speed (non-probing speed) is the T parameter.
On my printer I use [c]M558 P5 X0 Y0 Z1 H5 F100 T7000[/c].
BLTouch will go into alarm if it does not have enough space to deploy the pin completely, and when RRF is probing, the pin is immediately redeployed while Z is raising, it does not wait until the carriage is at the next probing point (in contrast to Marlin which will retract during travel moves).
PS: I don't know what happened to Calvinx, he manually deleted all of his posts. For the BLTouch at least his info is still on the Wiki.
Yep, that was it sigh
I thought reversing motor direction on y should only be if you want to move y back and it goes forth or so. Another thing learned ^^
Yes you are sort of correct but David made a change to the firmware with 1.19 because, prior to that, Duet firmware was different to every other firmware in this respect. So, those of us with CoreXYs have had to reverse the motor direction to make it the same as it was on older firmware versions. (Not sure if I've made it clearer or just muddied the waters more). Anyway, just remember that if you revert back to an earlier firmware version, you may have to reverse the direction of the Y motor again.
I work in an aerospace tooling company, and we regularly build thermocouple extensions up to ~25', and many of the thermocouples themselves are already 10-15' long, so your printer shouldn't have an issue with wire length.
Regarding making them, as Edkirk was saying, just make sure you have a good connection between the two wires, but ONLY at the end, where you want to measure the temperature.
My work uses a specialized TIG welder to fuse the wires, but there are many methods for field expedient joints.
Whatever you do, make sure that you have the right type of thermocouple.
The boards are designed to work with Type-K thermocouples, so plugging in a Type-J, for instance, would throw the readings off completely.
One last thought: Also, if you are making, or buying, a complete new thermocouple, make sure to buy stranded wire if it will be moving. The cheaper solid wire does not like flexing, and will break in short order.
The DueX4 expansion pins that are available are listed at https://duet3d.com/wiki/Using_servos_and_controlling_unused_I/O_pins. What I can't remember is which ones (if any) are servo-capable. Any that are could drive the bltouch directly because it seems to be happy with a 3.3V signal.
Other than that, all of your suggestions sound suitable. For a level shifter, a small signal diode and pullup resistor to +5V would probably be sufficient.
Thanks dc42. I am still waiting for the ir sensor to arrive (as well as the duet display), so I am in manual mode for leveling the bed at the moment.
Thanks for explaining the posts that I read about using the E1 output. I may use that in the future, but for now I'm taking the path of least complexity until I get this printer set up.
You are right... didn't notice that expansion headers heater pins could be reserved for something else with M307...
I think I will use the M42 for my purpose... should be rather easy to implement on post processor...
So to summarize best course of action is:
1. First disable expansion header heaters with M307
2. Then you can use M42 for driving them.
3. For 5V Pwm it's easiest to use 3.3V -> 5V Logic level converter as interface https://www.sparkfun.com/products/12009
I suggest you use the 12V 30A PSU, with the voltage turned up a little if you need more heat. Use the trigorilla board to switch it if you have one already, but you don't have to. There are instructions for using a separate PSU for the bed heater on the wiki.
Looks like your connection to Duet3D was lost, please wait while we try to reconnect.