Sturdy options
Posted by plankton
|
Sturdy options December 14, 2014 04:32PM |
So I built an i3 (box frame), and it's printing great (within it's limitations), and I learnt a lot from the process, however I have this recurring problem...
Whenever I think of an improvement or upgrade for some aspect of the printer, I invariably end up saying to myself, "well that's how the Mendel90 already does it"... it's very frustrating.
I always knew the i3 would be a stepping stone to my next printer, I just never expected to be considering a new one so soon, but I think come the new year I'm going to use the i3 to RepStrap a Mendel90 Sturdy. So now I have some build choices to make, which I hope people can help me with.
X axis:
I'll initially be printing with a J-head, but may want the printer to work with an E3D v6 in future, and dual heads.
I might also want to try a Z probe.
Is there any benefit in specifying a slightly longer X axis than default, to accommodate those options, or does the current X carriage have sufficient room for expansion?
Looking in "sturdy_config.scad" I can see "X_travel = 214" is the default, that implies to me there is not a lot of wiggle room if I ever needed a longer X carriage to accommodate those options?
Frame:
I plan to make the buttresses about 30 mm deeper (and adjust the base accordingly) to accommodate a slightly different sized PSU to the one Nophead uses.
I may use good quality ply instead of MDF, and will glue and screw it with countersunk screws instead of using the "B&Q style connectors".
Y axis:
I want to turn this through 180 degrees so the motor and enstop will be at the back, shorter cabling, and I can't see any downsides.
Electronics:
Still researching this, but for cost and customisability I think I may stick with the RAMPS I already know. The biggest downside with RAMPS (in my opinion) is the poor way that connections are made to the board, but I have an idea to run all the PCB header pins on RAMPS off to an adjacent board with more suitable connectors on it. This way the connections to the PCB headers on RAMPS don't have heavy cable looms pulling them off, and the cable loom connections to my daughter board would be more secure, and easier to access.
Finally, a question about the way Nophead hacks his ATX PSU, I notice he uses a resistor to provide a load to the 3v sense wire, however the Wiki suggests you can just connect the sense wire to 3.3v (which is what I've done on my i3). I admit I don't know enough about the way modern PC PSU's work to know if Nophead is just being cautious, or is the resistor still important?
Comments on any of these choices welcome.
Whenever I think of an improvement or upgrade for some aspect of the printer, I invariably end up saying to myself, "well that's how the Mendel90 already does it"... it's very frustrating.
I always knew the i3 would be a stepping stone to my next printer, I just never expected to be considering a new one so soon, but I think come the new year I'm going to use the i3 to RepStrap a Mendel90 Sturdy. So now I have some build choices to make, which I hope people can help me with.
X axis:
I'll initially be printing with a J-head, but may want the printer to work with an E3D v6 in future, and dual heads.
I might also want to try a Z probe.
Is there any benefit in specifying a slightly longer X axis than default, to accommodate those options, or does the current X carriage have sufficient room for expansion?
Looking in "sturdy_config.scad" I can see "X_travel = 214" is the default, that implies to me there is not a lot of wiggle room if I ever needed a longer X carriage to accommodate those options?
Frame:
I plan to make the buttresses about 30 mm deeper (and adjust the base accordingly) to accommodate a slightly different sized PSU to the one Nophead uses.
I may use good quality ply instead of MDF, and will glue and screw it with countersunk screws instead of using the "B&Q style connectors".
Y axis:
I want to turn this through 180 degrees so the motor and enstop will be at the back, shorter cabling, and I can't see any downsides.
Electronics:
Still researching this, but for cost and customisability I think I may stick with the RAMPS I already know. The biggest downside with RAMPS (in my opinion) is the poor way that connections are made to the board, but I have an idea to run all the PCB header pins on RAMPS off to an adjacent board with more suitable connectors on it. This way the connections to the PCB headers on RAMPS don't have heavy cable looms pulling them off, and the cable loom connections to my daughter board would be more secure, and easier to access.
Finally, a question about the way Nophead hacks his ATX PSU, I notice he uses a resistor to provide a load to the 3v sense wire, however the Wiki suggests you can just connect the sense wire to 3.3v (which is what I've done on my i3). I admit I don't know enough about the way modern PC PSU's work to know if Nophead is just being cautious, or is the resistor still important?
Comments on any of these choices welcome.
|
Re: Sturdy options December 15, 2014 11:10AM |
@nophead
Thanks for the feedback, I can see what you mean now about the Y axis, although with the extra 30mm there I may still get away with it (not that it's a big deal, I'll just see what works at build time).
That's very interesting about the 3.3v load, I hadn't considered it might affect the 12v rail, think I'll investigate that when I come to build the Sturdy, presumably the theory is that a load on 3.3v keeps the 12v rail more stable under load (I know that on my current setup the 12v drops to 11 something under various loads).
I may just end up adding 40mm or so to the X axis anyway, just to give a little more room for any future experiments with dual heads, or detachable Z probes.
Yes, I did consider unsoldering the existing headers on a RAMPS, to fit replacements, but unsoldering a connected row of headers is possibly beyond my soldering skills (unless the oven technique would work). Another option may be to buy a bare PCB and fit my own connections, I've seen bare PCB's available for the 1.3 version, though I need to have a better understanding of the differences between the 1.4 and 1.3 versions first (but that is probably a question for the electronics forum).
@minimales
I agree about the Z probe, I think it's a cool idea, but as you say, once your bed is calibrated correctly it becomes excess weight. I'm just trying to keep all my options open at this stage, build a platform that can cope with any future experiments I can envisage. I can't find the link right now, but I've seen Z probes that can be detached from the X carriage (by electromagnet) and just get parked at one end of the X axis when not in use.
My instinctive response to Bowden extruders has always been "why add additional complexity to your feed system, it's just more points of failure", but once you start using multiple extruders those negatives start to be outweighed by the advantages of removing all that additional weight from the X carriage. What we really need is a lightweight gearbox, so we can use one NEMA17 and a small servo to selectively feed multiple filaments, but until I design that I may have to familiarise myself with Bowdens when I want to try multiple hotends.
Interesting dual extruder implementation, but it wont scale very well past 2 extruders, and I'd have my doubts that the 2 carriages would stay perfectly in synch (however good his endstops).
Thanks for the feedback, I can see what you mean now about the Y axis, although with the extra 30mm there I may still get away with it (not that it's a big deal, I'll just see what works at build time).
That's very interesting about the 3.3v load, I hadn't considered it might affect the 12v rail, think I'll investigate that when I come to build the Sturdy, presumably the theory is that a load on 3.3v keeps the 12v rail more stable under load (I know that on my current setup the 12v drops to 11 something under various loads).
I may just end up adding 40mm or so to the X axis anyway, just to give a little more room for any future experiments with dual heads, or detachable Z probes.
Yes, I did consider unsoldering the existing headers on a RAMPS, to fit replacements, but unsoldering a connected row of headers is possibly beyond my soldering skills (unless the oven technique would work). Another option may be to buy a bare PCB and fit my own connections, I've seen bare PCB's available for the 1.3 version, though I need to have a better understanding of the differences between the 1.4 and 1.3 versions first (but that is probably a question for the electronics forum).
@minimales
I agree about the Z probe, I think it's a cool idea, but as you say, once your bed is calibrated correctly it becomes excess weight. I'm just trying to keep all my options open at this stage, build a platform that can cope with any future experiments I can envisage. I can't find the link right now, but I've seen Z probes that can be detached from the X carriage (by electromagnet) and just get parked at one end of the X axis when not in use.
My instinctive response to Bowden extruders has always been "why add additional complexity to your feed system, it's just more points of failure", but once you start using multiple extruders those negatives start to be outweighed by the advantages of removing all that additional weight from the X carriage. What we really need is a lightweight gearbox, so we can use one NEMA17 and a small servo to selectively feed multiple filaments, but until I design that I may have to familiarise myself with Bowdens when I want to try multiple hotends.
Interesting dual extruder implementation, but it wont scale very well past 2 extruders, and I'd have my doubts that the 2 carriages would stay perfectly in synch (however good his endstops).
Sorry, only registered users may post in this forum.