RE: Can't print below x=5 ish - print truncated
when the printer is at x+250 how close the the right travel limit is it in reality
So, at 250, you are printing on the edge of the removable build plate. I didn't take a picture, but to fit my 245mm wide print, I push it right up to the limit and removed the skirt.
Edit: Physically, you might have 5-10 mm of travel left on the right side.
RE: Can't print below x=5 ish - print truncated
that sounds like your bed is offset 5 mm to the left - you should be able to reach from 0 to 250 and 250 should be 1-2mm from the right physical limits
RE: Can't print below x=5 ish - print truncated
@david-a66
So I doubled checked my printer and moved the print head via move axis after doing auto home. The range was 5.2 - 250mm and at 250mm, the nozzle was on top of the print bed 250 line. Also there was about 3-5mm of space for the cable bundle and carriage to move to the right.
RE: Can't print below x=5 ish - print truncated
that sounds like your bed is offset 5 mm to the left - you should be able to reach from 0 to 250 and 250 should be 1-2mm from the right physical limits
That doesn't address the fact that the firmware won't move the head to less than 4.8mm. No matter what you do physically to the build , no matter how incompetent you are or even if you use the wrong parts, if the problem was with physical assembly the print head would at least try to get to 0mm, and it does not. The firmware sets the printhead at 4.8mm when it hits the endstop. The firmware limits the printhead range to 4.8mm - 250mm.
Thanks billcampbell for pointing out you don't even need to be able to type Gcode - if you use the front panel to do calibration - auto home, and then settings - move axis - move x, the range the printer offers to move the head is only 4.8 to 250m.
RE: Can't print below x=5 ish - print truncated
Hello all,
My 2.5s (upgraded from 2s) also has this issue; my minimum X is 4.1. This is just a little (perhaps 1 mm?) left of the x-endstop-switch limit, so the x-endstop-switch is open in this x-position.
Since it sounds like a) everyone has reported slightly different minimum-X values, and b) everyone's maximum-X is right on the 250 mm line on the bed, I wonder if this is related to initial XY-calibration? During initial calibration the printer spends quite a lot of effort precisely locating several spots on the print bed, so that's how it might know where the 250 mm line is. If the endstop switch then was found to trigger at some positive X-value, perhaps the calibration routine then remembers that value as the left edge and adds a little margin, just to prevent problems (better to lose about 5 mm of print space than regularly, or even worse intermittently, ramming the left side, I suppose).
If so, I suspect this is a consequence of grafting the Mk3S extruder into the Mk2 chassis: there wasn't quite enough room (especially clearance between the cable bundle and the frame) to do that cleanly in all examples of the Mk2 (which may well exhibit a mm or two of variance here and there because the gummi bears had arguments while assembling them), so rather than change the Mk3S extruder or add a different extruder just for the Mk2S, they came up with this workaround/tradeoff...
- vince
RE: Can't print below x=5 ish - print truncated
Maybe Prusa does not even know. I don’t think they do a lot of testing on the mk2.5, and It’s easy to miss.
RE: Can't print below x=5 ish - print truncated
Interesting. Am about to upgrade my MK2 to MK2.5s. But I just played with the Home/Move X axis settings on the Mk2, to see what would happen.
When I do a Home, the Y and Z axes are set to 0, but the X axis is set to 12mm, so the print head moves well in from the left side of the bed. Then I can move the X axis using the Move to go from 0 ( where the endstop hits the X microswitch ) to 250, right at the very edge of the bed. So yes, the Mk2 firmware allows a full 250 mm travel, even though the Home position is offset 12mm from the switch.
Not sure if this throws any light on what you see, but it is a bit of a surprise that the X axis does not home to 0. At least I now have some reference to compare with what I see after the upgrade.
Come to think of it, the microswitches must have built-in software offsets ( and sign reversals too ) since the Y axis switch is actually at the other end of the Y travel. Maybe somewhere in the Mk2.5 firmware, the X axis offset is wrongly set, so the distance between the microswitch trigger point and the nominal left edge of the bed is in the wrong place.
RE: Can't print below x=5 ish - print truncated
Come to think of it, the microswitches must have built-in software offsets ( and sign reversals too ) since the Y axis switch is actually at the other end of the Y travel. Maybe somewhere in the Mk2.5 firmware, the X axis offset is wrongly set, so the distance between the microswitch trigger point and the nominal left edge of the bed is in the wrong place.
Is it? I'm pretty sure my Y-axis microswitch has always been at the back of the printer, and triggers when the print-bed gets fully back, i.e. Y=0.
RE: Can't print below x=5 ish - print truncated
You are correct. S betill interesting that the x axis home point is displaced 12mm from the microswitch trigger point.
RE: Can't print below x=5 ish - print truncated
I see what you're saying, regarding instructing the print to go to x=0.
However the fact that my printer can't physically reach the edge of the build plate isn't going to be solved by firmware, and from what I could gather from your message your printer has the same issue?
Now on a regular printer there's no other way to know if you're at zero than the endstops, but prusa has the xyz calibration and checks for the location of the metal on the bed, I would expect the firmware to use those as reference points to calibrate origin. (pure speculation on my part, have not looked at the code yet).
So it can know where those points are, set origin based on those and then hit the endstop and not allow past that.
I agree it could in principle do that during the initial calibration. It's still the case that the printer won't move the head to a value less than about 5mm, and on mine could not get to 0mm even if the endstop was made 5mm shorter (because cables would hit frame), so whether it's hard-coded into the firmware at 245.2mm, or it does detect that it has less than 250mm travel during calibration, I believe the 2.5S does not have the full 250mm print width, and I don't believe my printhead is somehow assembled 5mm wider than it should be - I generally work to rather finer tolerances than that. On an assembly that's about 50mm wide you need to be pretty cack-handed to make it 5mm wider than it's supposed to be.
So I decided to test this hypothesis, i.e. that the printer detects the bed position during calibration and thus knows how far right it can go. Unfortunately, in doing so I broke my PINDA mounting so the sensor is now not securely attached, which was not part of the plan.
Anyway:
I packed my PINDA to be about 1mm to the right, figuring that as far as the printer would know that's the same as moving the bed about 1mm to the left (since it needs to move the head less to the right to get to the far side of the bed, but the x-carriage endstop still triggers in the same place on the left). I then re-ran the whole setup and calibration wizard from scratch, as if it was a brand-new machine.
After that, my range of head movement is now 5.6 to 250mm.
I then put the PINDA back where it was (actually, bodged it back roughly where it was, because I've snapped the mount), and my range of movement is 4.4 to 250mm.
So, I conclude that the printer DOES detect where the bed actually is during calibration, and adjusts its end limits accordingly. However, it does it by always making the right hand side 250mm, and limiting the left side to whatever it needs to - in my case (now) 4.4mm.
I don't think I could realise the full 250mm with my printer. My bed is asymmetric in the frame, and there is space to move it about 4mm to the right, but I don't think I could get the full 5mm without the cable bundle clashing. Also, I can't see how I'd achieve that 4mm movement - there aren't any oversize holes with that range of movement in the bed assembly - I suppose the threaded rods could displace in the frame slots, but the rest of the connections have much less than 1mm tolerance.
So, since I've only wanted to print one thing more than 240mm, I'll just try and remember that my object limit is 245mm and if I go above 240mm I need to omit the skirt and fine-tune the placement of the object on the bed.
Now I need to try and reprint an 'extruder body' part on a printer with a dodgy PINDA and rebuild my printer...
RE: Can't print below x=5 ish - print truncated
Its worth reading the comments about the endstop assembly step 65 in the manual:
RE: Can't print below x=5 ish - print truncated
A few more observations.
The critical distance seems to be from the Z centreline of the PINDA probe to the X face of the microswitch endstop. These two dimensions should be the same for both the MK2 and MK25s versions, as far as I can see. ( unless Prusa moved the PINDA targets in the new heatbed - something I havent checked )
On the Mk2, its possible to measure that distance from the STL files for the extruder body and the X carriage, using one of the mounting holes for reference between the 2 parts. Comes to 49 mm, if I read it right. That also agrees with the physical hardware I have.
On the MK2.5 you can do the same, but you need to add in also the X thickness of the endstop_block. Comes to 52.5 mm according to me. That also agrees with my measurements on the parts - I actually got 53mm, but its hard to measure exactly.
So there is a 3.5 mm difference between the two common dimensions on the different models.
Maybe this is the cause of your problems.
I thought also a bit more about Jiri Syrovako's comments I posted above, but I cant possibly see how it could work without the new endstop. Its absolutely crucial that the X microswitch is triggered during any Home command. I cant see how you could get that position off the probe and heatbed target without doing a lot of searching around during a Home - just like is done in the XYZ calibration refinement process.
RE: Can't print below x=5 ish - print truncated
@ajisis
to the best of my knowledge, the left hand pinda spots are in the same place but the other spots have been moved on the mk2.5 prusa with the 12 volt mk52 heatbed!
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: Can't print below x=5 ish - print truncated
I think maybe Jiri has spotted the same issue but hasn't realised that failing to get closer than 4mm to the zero is correct for a 2.5S.
That is, I think the new extruder assembly does intentionally stop with the nozzle 4mm to the right (partly due to the extra block, but it doesn't matter whether it's an extra block bolted on or if it was integral to some other part in the assembly). If the 2.5S firmware corrects for that you'd get what Jiri is reporting if you tried a calibrate without updating the firmware - the PINDA would miss the spots and would go too far right (because it thinks the head hits the microswitch at x=0, so would want to move the head across 31mm to find the first PINDA spot, but if the switch is triggered with the nozzle at x=4, you only need to move 27mm to find the spot).
The spots on a 12V Mk52 bed are in a different place from those on a Mk42 bed (as Joan says), but that isn't the issue here because 2.5 to 2.5S upgrade doesn't change the bed - that's in the 2 to 2.5 upgrade. However, by my measurements, going by the markings (which are in the same position relative to the fixing bolts on both print beds), the left spot is at x=35, right spot is at x=243 on the Mk52, but used to be at x=238 on the Mk42. There are no middle spots on the Mk52 either, and the bed levelling probes at different points anyway.
I agree the homing cannot be using the PINDA spots - not only does it not do a PINDA dance when it homes, they are covered by the steel plate when it homes, and an inductive probe is going to struggle to find them through a steel plate, I'd expect.
I still wish Prusa had been up-front about reduced print width. I would probably have done the upgrade anyway (because I regularly had trouble with the previous filament sensor, and had disabled it). I don't buy the 'maybe they don't know' argument because there's too much that needs to have changed (including where the PINDA starts searching during calibration) for it to be an accident.
RE: Can't print below x=5 ish - print truncated
Just finished my upgrade from Mk2 to Mk25S. Heat bed is now Mk52, and I used the standard X endstop on the microswitch. Firmware version 3.8.1. XYZ calibration failed at point 3, but lowering the PINDA probe about 0.5 mm fixed that. Then calibration completed with no errors. From the way the bed is searched by the PINDA probe for the target points, I doubt this has anything to do with the X axis positioning, only with things like the axis skew.
After an Autohome, the X axis reports the position as +12.0mm. Then using the Settings I can move the X axis back to 4.3 mm, where it triggers the microswitch, and stops maybe 4 mm in from the left edge of the bed,and up to 250mm right. So indeed the range is restricted by 4.3 mm. Since the microswitch is hit only just before the X axis cable bundle hits the frame. I cant see that you could go any further left anyway. Guess its just a limitation we have to put up with. Maybe one could gain another mm or so by shortening the endstop, but I think I will live with it.
There is now a readout in the Support menu for XYZ cal details, which says the [0,0] offset is -4.28 mm in X, but that has no bearing on the X movement range that I can see.
RE: Can't print below x=5 ish - print truncated
From the way the bed is searched by the PINDA probe for the target points, I doubt this has anything to do with the X axis positioning, only with things like the axis skew.
I think it does rely on the PINDA for X positioning during calibration, because different printers have a different offset (as you can find by reading back through the thread - everyone has four-ish mm, but not the same value), and because if you move the PINDA relative to the carriage you'll get a different offset when you calibrate.
I hadn't noticed the support menu had that information. Mine is now showing -4.48mm.
RE: Can't print below x=5 ish - print truncated
@ian-s3
So if you edit the pinda mount and print a new part you would theoretically be able fo fix this? 🙂
R
RE: Can't print below x=5 ish - print truncated
@ian-s3
So if you edit the pinda mount and print a new part you would theoretically be able fo fix this? 🙂
R
No, because the printer thinks it knows where the nozzle is with respect to the PINDA. So if you increase teh PINDA-nozzle distance by 4mm then yes you'll get the nozzle 4mm closer to the zero position, but it will stop 4mm further from the far edge. You'll still have 246ish mm of nozzle movement, just in a different place on the print-bed. Plus, of course, you need to overcome whatever clash it is that caused the offset to be introduced in the first place.
Also, as alluded up the thread, if you change the offset between the nozzle and the surface that triggers the endstop microswitch by more than a little amount, then the PINDA apparently doesn't find the sensing zones at all (because it starts in the wrong place, but the amount you've changed that offset).
If you actually feel you need to fix it, I think you need to change the firmware (as well as build the printhead tighter, or even modify it, to eliminate the clash the offset is fixing).
RE: Can't print below x=5 ish - print truncated
I was able to see similar behavior. My MK2.5 can start at x=2.6 and my MK3s starts at x=2.7
Looks like it's hardware related. Depending on the possible end-stop the printer adjust the print plate size.
Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram
RE: Can't print below x=5 ish - print truncated
@ian-s3
That would be a nice update/bugfix from Prusa I think.
If I mount this part on 2.5 upgrading to 2.5S, it actually blocks the print head too far from the left ~4mm. In the effect, the XYZ calibration fails (bed calibration point was not found - on the 2nd or 3rd point - on the right side - it is also visible, that PINDA then makes it too far to the right out of bed). When trying Auto home, it puts the PINDA 4 mm to the right from the mid of the circle, where it must be precisely located.
See the help: https://help.prusa3d.com/article/zQeBMm2...
Any ideas? Remove this part completely?
Jiri Syrovatko - 10/16/2019
- Well, without waiting, I removed this part in Step 65 and the printer immediately calibrated OK afterwards.
Jiri Syrovatko - 10/16/2019