OK. That's what I thought. I do wonder, though, if mine is working as intended, since the bed doesn't move down as the print head moves to each of the 9 points. Instead, the head just gets dragged along the glass. I've read where others see a 4 mm gap between head and glass during transit.
Can you post a picture of your nozzle and ir sensor? Did you follow the guide to put it 2-3mm higher than the nozzle on a bit of belt?
Hi, @Pierce. Yes, I did the belt spacer thing as part of commissioning. I found, though, that I had to move the IR sensor a couple of times after - usually when the heads would finish the auto home sequence too far from the bed. In those instances, I nudged the sensor a bit higher.
This could be your fault already. After Autohome, the nozzles are supposed to be lifted to 4mm, as also shown on the display (z = 4).
Did you go through the information and troubleshooting page for the IR sensor? https://miscsolutions.wordpress.com/mini-height-sensor-board/ In rare cases it may require adjustment, though if there's something wrong with it obviously you should contact support.
Yeah, I think that's at least a symptom of the underlying problem. When I autohome, the nozzles move to the appropriate position on the bed, but the bed does not drop 4 mm. Same issue when doing the mesh leveling. So my nozzles are almost always in close approximation to - or direct contact with - the glass. When I start a new print, however, the bed does drop a bit while the head goes to the purging dock, then rises up to meet the head when the print begins. Or, at least it used to until this past weekend. Now the bed stays where it is, and the nozzles clip the edge of it when they come out of the purging dock.
Aside from homing is your bed moving well when you move axis ? Also what was the offset when you set up the Z home (so when you autohome what is the Z height when it is done)
If the bed does not move in Z during homing, there must be a fault. Try to send M119 at different heights to see if the sensor is triggered all the time or if it's triggering at all.
The bed will move during homing. Typically, it rises to touch the nozzles, then backs down very slightly. This, I think is normal behavior. However, sometimes it will keep rising, pressing up on the nozzles. What it doesn't do, that I have seen others describe, is move down 4 mm after homing and/or move down 4 mm as the head travels from point to point during the mesh leveling procedure.
Assumed there is nothing broken and the cabling is ok and you're always waiting for octoprint to come up, this could only be the sensor mounted too high or mounted in an angle. Did you modify the firmware?
Nope, I'm using the original firmware. If the IR sensor is out of position, I'd expect the behavior to be more consistent - i.e., if it's too high, then the bed would always smash into the nozzles. Instead it's just unpredictable. I've disconnected Octoprint completely, and will run through the commissioning process again when I have some time to get back to the machine. In the meantime, my (previously) much maligned delta is working like a champ, so I'm not completely out of the game.
You've had a pretty good go at this. Support will be back in the office tomorrow: it might be time to check with their faults database.
Just wondering if there is any way to check the consistency of the IR measurements, i.e. another way other than keep raising the bed.and hitt nozzles. I like quite a few others I suspect are using the original firmware and printing from a single nozzle and it works. If it isn't the hairspray causing inconsistent height measurements then, a faulty IR sensor? (even loose inside the housing?) faulty wiring? bad connection on the Rumba board? etc. What you need is a way to move something like a mirror a set distance under the IR sensor and see at what distance it triggers and see if this is consistent. Plus working through the whole system, sensor in housing, wiring, connection to Rumba would not be a bad thing.
The 4 mm move only occurs during the mesh leveling process. As each point is completed, the bed will move down before it travels to the next point. You then move the knob to position the head and click the knob. It will them move down and move on to the next point and so on. When you home, the head will travel to the near left corner move down and then move up. The rest of the process will follow whatever starting script comes next. For me it is the dump bucket, filament purge and then moves to start the print. If you auto home or move the sensor or bang the head or bed, you need to redo the mesh process from the start. With mesh leveling, the z-stop/ sensor is only used to as a z-stop and will have no function in the leveling process. If everything is working, you will not need to re-level again until you move the sensor by hand or the bed, etc.
Another thing occurred to me, You know this theory that the IR sensor triggers different depending on whether it hits the white lines on the heated plate or not ? So that's during autohome...mine hasn't done that once so maybe it is just luck of the draw if you happen to set the X and Y endstops at the exact point where the IR sensor finds a white line at the homing position (IR sensor over centre of bed). So I wondered, people getting this intermittent homing...if you gave the X and Y end stop screws each a quarter turn (in any direction) to slightly change home and therefore the IR sensors homing position on the bed then maybe that clears it. I didn't go all OCD on getting the nozzle exactly over X0.00 Y0.00 as the instructions say...mine is within a few MM of home according to the bed grid.
I'm not sure if this helps but my starting code for simplify3d is included to see if it helps. I was having levelling issues and I found when I started using the blue tape on the galss bed, there were fewer faults. however I still had to drop the bed an extra 15mm prior to homing for it to level perfectly, so I added some extra lines in the Start script. Also not sure if the G29 is really needed but I added it to be sure its being used. I have a dual, but this script is for my profile when im only running ABS through the Main extruder (Being the right side extruder)..... ;BigBox Printer Start Script Begin M218 T1 X34.8 Y-0.8 //Dual Nozzle Offset (T1 to T0 Relative)// ;Homing M117 Homing. G1 Z15 F400 //Lower Bed 15mm// G28 //Home Bed// G29 //This command loads the stored IR bed levels// ;Dock M42 P6 S0 //Turn off LED Lights to 0% (Rumba P6 Position)// T0 G1 X70 Y200 F3000 G1 X70 Y240 F3000 T1 //Select Tool 1 (Right)// G1 E25 F400 //Prime the Nozzle// T0 //Select Tool 0 (Left)// G1 E27 F400 //Prime the Nozzle// M117 Print Starting. M42 P6 S255 //Turn on LED Lights to 100% (Rumba P6 Position)// ;BigBox Printer Start Script End
I've seen some discussion about what the IR sensor actually "sees" when it fires - whether it's the surface of the glass or the PCB beneath it. I couldn't find a definitive answer. If it's the surface, would it make any sense to make a "target" on the surface of the glass (white paint or something) in the area that the sensor sees when homing? Probably not, since someone smarter than I am would have already done it...
You could apply some UHU directly under the sensor area. This should opaque the glass enough that it will pick up the glue rather than what is underneath. It will also not significantly affect the distance to the nozzle.