3.5 hangprinter: Allow custom and prism-like setups #603
Conversation
jtimon
force-pushed
the
3.5-hangprinter-allow-prism
branch
2 times, most recently
from
582347e
to
7101a2c
Compare
jtimon
force-pushed
the
3.5-hangprinter-allow-prism
branch
2 times, most recently
from
1bd250d
to
7f30701
Compare
jtimon
changed the title
jtimon
changed the title
| // 0=None, 1=last-top, 2=all-top, 3-half-top, etc | ||
| uint32_t unsignedAnchorMode = (uint32_t)anchorMode; | ||
| gb.TryGetUIValue('T', unsignedAnchorMode, seen); | ||
| if (unsignedAnchorMode <= (uint32_t)HangprinterAnchorMode::AllOnTop) { |
There was a problem hiding this comment.
we could add an else part here to return an error if somebody tries to chose an option that's not supported, no?
I mean, right now it goes to the default which is the regular hangprinter, I think that's good enough.
There was a problem hiding this comment.
Something like this?
reprap.GetPlatform().Message(ErrorMessage, "M666 A?: Invalid anchor mode\n");
Nice to have, especially for ourselves during the time we will be testing different binaries and might forget which one is on the machine ;-)
|
Updated code, description and title. In my eyes it is ready (and the two PRs it depends too). I'm probably still missing some things or doing something wrong. Eager to fix whatever is left. |
Otherwise it gets too deep. I think there's an option in eclipse to forbid this by project in format, one can config depness iirc.
jtimon
force-pushed
the
3.5-hangprinter-allow-prism
branch
from
7f30701
to
d51a945
Compare
This is a clean generalization from 4 to N. Pseudo-pyramid because the vertices in the base of the pseudo-pyramid don't need to be on the same plane. There's no base of the pyramid as such. Better documented and cleaner thanks to @tobbelobb on github.
…etups - Change from bool to class and introduce None mode for custom (experimental) setups (all allowed) - refactor out IsInsidePyramidSides - refactor out IsInsidePrismSides - introduce HangprinterAnchorMode::LastOnTop for prism-like setups with all anchors on top - introduce a T parameter in M666 to configure the anchor mode
jtimon
force-pushed
the
3.5-hangprinter-allow-prism
branch
from
d51a945
to
fd9f260
Compare
| } | ||
|
|
||
| bool HangprinterKinematics::IsInsidePrismSides(float const coords[3], unsigned const discount_last) const noexcept |
There was a problem hiding this comment.
I like these function names
| ok = f->Write(scratchString.c_str()); | ||
| if (!ok) return false; | ||
|
|
||
| scratchString.printf(" U%.3f", (double)mechanicalAdvantage[0]); |
There was a problem hiding this comment.
Print integers after U.
" U%d", (int)mechanicalAdvantage[0] or " U%u", (unsigned)mechanicalAdvantage[0]
It's block-and-tally style gearing ratio, so can only be 1, 2, 3, 4, ...
| ok = f->Write(scratchString.c_str()); | ||
| if (!ok) return false; | ||
|
|
||
| scratchString.printf(" O%.3f", (double)linesPerSpool[0]); |
There was a problem hiding this comment.
Print integers after O. Lines per spool can also only be natural numbers 1, 2, 3, 4, ....
| ok = f->Write(scratchString.c_str()); | ||
| if (!ok) return false; | ||
|
|
||
| scratchString.printf(" L%.3f", (double)motorGearTeeth[0]); |
There was a problem hiding this comment.
Motor gear teeth is also a natural number
| ok = f->Write(scratchString.c_str()); | ||
| if (!ok) return false; | ||
|
|
||
| scratchString.printf(" H%.3f", (double)spoolGearTeeth[0]); |
| ok = f->Write(scratchString.c_str()); | ||
| if (!ok) return false; | ||
|
|
||
| scratchString.printf(" J%.3f", (double)fullStepsPerMotorRev[0]); |
|
Oh, I commented in the wrong PR again. Seems like the |
In addition to allow any pseudo-pyramid as in #598 which this PR depends on, this allows to setup hangprinters without a middle anchor.
This doesn't require any extra anchors. One could, for example, set each of the 4 anchors in one top corner of a room, resulting in a quadrangular prism printing volume. Or it could be used to move cameras in stadiums and whatnot.
For that we just set all the anchors on top.
It also allows a "None" anchor mode which always IsReacheable always returns true, which is interesting for people experimenting with potential future modes not contemplated yet, like halfTop-halfDown setups.
Dependencies:
Outdated
~~
Two options come to mind:
A) Allow users to manually select it by adding some parameter in M669. If we chose this option, any suggestion?
B) Calculate whether the last anchor is "in the middle" or not. Probably I should explain with code what I mean by "in the middle" rather than trying it in english to avoid misunderstandings.
~~