Kossel 2020, possible auto level issue?

Hello everyone! Finally got my printer basically working for the most part. And now im running into an issue when printing. Here's the issue: when i click print everything works fine, bed and extruder heat up, printer hits endstops then comes down 30mm above bed. Then when it tries to start printing in comes down to about 1mm off the beds surface starts to move, then basically goes into the bed basically putting little marks in the bed and the motors just start freaking out. I noticed that the printer doesn't do the auto level with the servo motor and switch. I wired that up, so im not sure why it isnt doing it. I really dont have a clue how to resolve the problem. Can someone help me out with this issue!!?

Your gcode either needs to execute a G28 for manual level, or a G28/M401/G29/M402 (The M401/M402 deploy and retract the probe) for auto, and it is your task to insert these commands in the start gcode in your slicer - none of this is assumed, automatic, or initiated by the firmware.

I also *strongly* suggest that you do a good manual calibration no matter what . . . at the very least, it gives the auto a good starting point, and at best, you may find it works better than auto level . . .

- Tim

Quote
tadawson
Your gcode either needs to execute a G28 for manual level, or a G28/M401/G29/M402 (The M401/M402 deploy and retract the probe) for auto, and it is your task to insert these commands in the start gcode in your slicer - none of this is assumed, automatic, or initiated by the firmware.

I also *strongly* suggest that you do a good manual calibration no matter what . . . at the very least, it gives the auto a good starting point, and at best, you may find it works better than auto level . . .

- Tim

Tim,
The printer does execute the g28 command before print, I added the probe command and yet the same problem still occurs.. Want me to post a vid to show you exactly what is happening??

Have you calibrated the Z axis, and the printer in general? If you manually give it G1 Z0 F3000, where does the tip end up? Where it parks immediately prior to printing isn't at Z=0 . . . it's wherever your start gcode tells it to go, and once up to temp, it's going to move to the first layer height, which in your case, appears to be below the bed.

I'm not going to rewrite the calibration procedure, since it's in the docs, but when at Z=0, a single sheet of paper should have some resistance but still move between the tip and bed. You also need to be certain that Z=0 at all three towers . . .

For auto, you need to calibrate that as well, to ensure that the probe offsets are correct . . . otherwise, the autocal has no idea whee the nozzle is and can crash.

For what it's worth, I followed Folger's docs exactly, and had a working printer on the first try . . .

- Tim

Edited 1 time(s). Last edit at 07/04/2015 07:48PM by tadawson.

Yes I have calibrated the Z axis, using a sheet of paper and determining how low the nozzle should go. I also noticed when calibrating that when the extruder moves to the edge of the bed it goes down, which is what im assuming is the issue as to why the extruder is going into the bed. Im going to test what you asked did and ill get back to you.

Quote
tadawson
Have you calibrated the Z axis, and the printer in general? If you manually give it G1 Z0 F3000, where does the tip end up? Where it parks immediately prior to printing isn't at Z=0 . . . it's wherever your start gcode tells it to go, and once up to temp, it's going to move to the first layer height, which in your case, appears to be below the bed.

I'm not going to rewrite the calibration procedure, since it's in the docs, but when at Z=0, a single sheet of paper should have some resistance but still move between the tip and bed. You also need to be certain that Z=0 at all three towers . . .

For auto, you need to calibrate that as well, to ensure that the probe offsets are correct . . . otherwise, the autocal has no idea whee the nozzle is and can crash.

For what it's worth, I followed Folger's docs exactly, and had a working printer on the first try . . .

- Tim

Can you please give me the link to what you used to config your 3d printer, cause I cant seem to find the configuration portion for my printer.

Quote
tadawson
Have you calibrated the Z axis, and the printer in general? If you manually give it G1 Z0 F3000, where does the tip end up? Where it parks immediately prior to printing isn't at Z=0 . . . it's wherever your start gcode tells it to go, and once up to temp, it's going to move to the first layer height, which in your case, appears to be below the bed.

I'm not going to rewrite the calibration procedure, since it's in the docs, but when at Z=0, a single sheet of paper should have some resistance but still move between the tip and bed. You also need to be certain that Z=0 at all three towers . . .

For auto, you need to calibrate that as well, to ensure that the probe offsets are correct . . . otherwise, the autocal has no idea whee the nozzle is and can crash.

For what it's worth, I followed Folger's docs exactly, and had a working printer on the first try . . .

- Tim

Tim
When G1 Z0 F3000 is exectued the nozzle of the extruder goes all the way down to the bed and touches it. Im starting to believe the Z=0 may not be the same for all three tower. Do you know how to check this?

I jog the head as close to each tower as it will go with Z well above the plate (I actually setup buttons in Pronterface to allow recheckomg this easily), and then jog Z up/down doing the paper test, and then M114 to see where you are. You either need to adjust the mechanical endstops, or offset with M666 to get all three tower positions to match. Then, check the center . . . if not the same as at the towers, adjust DELTA_RADIUS until it is . . . and yes, this sucks, since it requires a recompile/download each time, and may shift the overall Z position. The key is to get all four points to match, and then reset Z=0 in the center, either with another compile/download, or an M206 offset in EPROM. It's a bit fiddly, and there is some interaction, so keep rechecking until everything stays thesame, and you should be good to go.

I can't find the specific info I started with, but it's here on RepRap . . . when I get back in my shop, I'll try to remember to post the link, but it's basically what I summarized above.

And if you are quite a bit off, recheck all delta values in the config, and verify your rod length . . . the values from Folger are fairly close if built to thier docs.

- Tim

Edited 3 time(s). Last edit at 07/05/2015 08:31PM by tadawson.

Dont want to sound majorly uneductated but once i do the paper test to each tower how do i tell the printer to set that. In the firmware what am I looking for to set it the numbers correctly?

As a rough setting, you move your endstops physically to get it as close as you can. Then use M666 Xn Ym Zk to set a homing offset for the tower in question (in M666, X, Y, and Z refer to the towers, not axis . . .so one parameter per tower . . .) to get it exact (these, IIRC, can only be negative values, and move the effector *down* - you can't go up, since it would try to drive through the endstop) followed by M500 to store that value in EEPROM (and M503 lets you see what's in the EEPROM).

So, physically adjust endstops, and then consider the lowest point to be your zero reference, and use M666 to get the others to match. Then check center, adjust offset, and then set overall zero either via compile or M206 Zn (for M206, X, Y, and Z *are8 the axis and not the towers . . .) and M500 to save . . . Once complete, the paper test should be the same for Z=0 at all four points, and you are good to go. (I tend to set my Z in the compile a tad high, so I can tweak it with M206 if needed, and not need to recompile.)

- Tim

Edited 1 time(s). Last edit at 07/05/2015 03:46PM by tadawson.

Just going on a whim here but it may not be a calibration issue.
Check your pronterface settings just under the port settings at the top make sure it says - XY: 3000 then Z:3000 this is the speed setting for the motors.
On default pronterface is set up with your z motor moving much slower than your xy motors; assuming that z is only for up/down, but since this is a delta printer your up/down is controlled by all three motors moving in unison so they all need to be set to the same speed.

On a Kossel, X, Y, and Z in Pronterface still refer to the actual directions moved, not the towers (that translation happens in Marlin) so having a different value for Z will do nothing more than to slow down vertical moves compared to X and Y - it will *NOT* cause a mismatch in tower speeds . . . . you would have to break that in Marlin.

- Tim

Edited 1 time(s). Last edit at 07/09/2015 01:30PM by tadawson.

Tim,

Still seem to have the same issue as noted before. I noticed when the auto level starts it only does half of the heat bed not the full thing, which could make sense why the hotend is crashing on the part of the bed that the auto level doesn't go over with. Would you know why this does this???

By the way i really do appreciate you helping me out! I really wanna pront something and you're making that a reality hopefully!!

We will help you with the autolevel, but it's hardly a requirement to print. Do a manual setup, print something (and see if any other issues show up), and come back to the autolevel. The first thing I would check on that, assuming your probe offsets are correct is the PROBABLE_AREA defines - it pretty much tells Marlin where autolevel can and cannot go.

I note that I don't see any mention of what Marlin version you are on . . . be aware that Delta autolevel has been pretty screwed up at times, and it's not impossible that you have a broken version (I never could get it to work with what Folger shipped, and went to 1.0.3-dev instead of screwing with it . . . 1.2.0 might work, but I never tried it . . .)

- Tim

Edited 1 time(s). Last edit at 07/09/2015 01:37PM by tadawson.

Tim,
Im not exactly sure where to find the probable_area in config.H. and i did notice and at this point itd worth saying that in folger techs manual the config for Manual_Z_Home_Pos is 253.9, mine is 480.45 due to the fact that when i entered g1 z50 f4000 the extruder was no where near 50mm above the bed. More like 200mm. Could this be the possible issue in thatchanging this variable so greatly affected other variables?

There is no sane way that a height of 480.45 is correct on a Folger Kossel . . . if it varies from the 253.9 value, it should vary by a couple of mm at most. Check your steps/mm and microstepping jumpers . . . I suspect that you will find one of the two wrong, and since all other parameters in the Folger config are based on the proper motion, I suspect that has something to do with your issue. As I think I said before, if you build by the manual(s), the config is very,very, close as provided.

Are you sure you have the correct Marlin download? Folger has an image for both 32 tooth and 20 tooth pulleys, but have not shipped 32 tooth for quite some time. The error does not calculate out exactly with your numbers, but is in the range of what you are seeing . . . I know that they have updated the 20 tooth version pretty recently, but suspect the 32 tooth version may have been left for dead . . .

You still have not clarified which firmware you are one, but if you are not seeing 'PROBABLE_AREA' pretty much verbatim in the config, I suspect you are on something pretty old . . .

- Tim

Edited 1 time(s). Last edit at 07/09/2015 06:54PM by tadawson.

- deleted -

Edited 1 time(s). Last edit at 07/09/2015 06:49PM by tadawson.

That doesn't make sense then, because i followed the build dox word for word and use everything they told me to do. And im having these problems & now apparently i have an out of date firmware that supposedly the newest firmware is in the build dox yet im using the firmware in the build dox

Which version, 32 or 20 tooth? The docs often refer to 32 tooth pulleys, but were not updated to reflect the change . . Folger is not as diligent as they could be updating thier docs . . . And I looked, both were updated in April . . . not sure when you got your kit. Really the only difference in the two is config stuff . . . the actual Marlin software is not pri ter dependent.

And what Folger is shipping may be current to *them*, but is a couple of versions back as far as Marlin is concerned, so very little chance of getting fixes. I also noted that my print quality improved when I updated mine . . . . but that is *NOT* the source of your problem.

In your Configuration.h, what is the line for setting steps/mm?

- Tim

Edited 3 time(s). Last edit at 07/09/2015 09:43PM by tadawson.

I have 32 tooth & i bought the kit in late march so that may be the problem?? Idk

That's when I got mine, and they had switched to 20 tooth . . .Try the other firmware package, I think you will be pleased . . . And you can always count the teeth on one of your pulleys to be certain. The delta geometry will also likely be wrong, as you noted, causing severe high/low issues across the bed.

- Tim

Edited 1 time(s). Last edit at 07/09/2015 11:29PM by tadawson.

Tim!! Sorry for such an absence work got to me!
Fortunately your idea of installing the 20 tooth firmware worked!! It fixed the bed crashing issue!! Now unfortunately theres one more issue. It seems like the X axis carrige mount bottoms out when trying to print. Nothing else does but for some reason that tower does. Would you know the possible issue?

Bottoms out as in hits the stop? How horizontal are the rods when that happens? If pretty much horizontal, then you have hit the maximum diameter you can print with that rod length. If not, you likely have your linear rails mounted too high on the towers, and will need to shift them down.

if things are distorted, and X is moving too far (although I would think this would be obvious due to being so far off center when going to Z=0), then either steps/mm or microstepping may be wrong.

- Tim

Edited 1 time(s). Last edit at 07/21/2015 02:24AM by tadawson.

Quote
tadawson
Bottoms out as in hits the stop? How horizontal are the rods when that happens? If pretty much horizontal, then you have hit the maximum diameter you can print with that rod length. If not, you likely have your linear rails mounted too high on the towers, and will need to shift them down.

if things are distorted, and X is moving too far (although I would think this would be obvious due to being so far off center when going to Z=0), then either steps/mm or microstepping may be wrong.

- Tim

Tim,
Im talking about hitting the actual botton of the linear rail, it like gets jammed. The rod goes completly horizontal when this happens. So your saying my micro stepping could be wrong or my z is offcenter? How could i adjust these issues??

If you hit the bottom stop, then the question becomes (as I said above), are you trying to print something larger than your printer is capable of? This printer, build as per the docs, has a max radius of about 76mm . . . . limited, as you have seen by the rods being horizontal. If you print a modest square or cube, is it square or distorted? And for that matter what are you trying to print? This may well not be a problem in the printer at all . . .

And no, I am not suggesting that Z is off center, only that if one tower was set at the wrong stepping or steps/mm, it would be seen as an offset as the head moved in Z.

- Tim

Edited 1 time(s). Last edit at 07/21/2015 01:36PM by tadawson.

Quote
tadawson
If you hit the bottom stop, then the question becomes (as I said above), are you trying to print something larger than your printer is capable of? This printer, build as per the docs, has a max radius of about 76mm . . . . limited, as you have seen by the rods being horizontal. If you print a modest square or cube, is it square or distorted? And for that matter what are you trying to print? This may well not be a problem in the printer at all . . .

And no, I am not suggesting that Z is off center, only that if one tower was set at the wrong stepping or steps/mm, it would be seen as an offset as the head moved in Z.

- Tim

Well unfortunately i cant even print right now die to the problem. I haven't been able to get an extrusion because of this lol. When i move the extruder using G1 X76 then to G1 Y76 one of the towers i believe it to be the X tower but im not exactly sure; bottoms out. But the what i assume to be Y tower will bottom out but then the 2 other towers still move and continue executing the commands. Lol its throwing me off

X=76, Y=76, doing the geometry gives a radius of 107, which *should* bottom things out - it's beyond the maximum radius of this printer. X=76, Y=0 is a radius of 76, X=0, Y=76 is a radius of 76, and X=53.7, Y=53.7 is a radius of 76 . . .

It still sounds like there is no problem with the printer . . . . and keep in mind that the print area is a circle, not a square . .

- Tim

Edited 1 time(s). Last edit at 07/21/2015 06:37PM by tadawson.

Quote
tadawson
X=76, Y=76, doing the geometry gives a radius of 107, which *should* bottom things out - it's beyond the maximum radius of this printer. X=76, Y=0 is a radius of 76, X=0, Y=76 is a radius of 76, and X=53.7, Y=53.7 is a radius of 76 . . .

It still sounds like there is no problem with the printer . . . . and keep in mind that the print area is a circle, not a square . .

- Tim

Well i no longer believe theres a printer issue. I just need to fix the bottoming out issue. I believe its definitely firmware. Ill take short vids to show you what i mean!

What you have never mentioned is exactly what you are doing (or trying to do) that causes this to happen?

Not knowing the cause, makes it very difficult to be of much help, although the simplest is simply trying to print thjngs that are too big. I suppose that if you errantly set up your slicer with 0,0 on a corner and not in the center (as it should be), it would offset everything and likely do this as well. But again, not kjowing what you are doing, I can only guess . . .

And there is a setting in Marlin (well, the current stuff anyhow) that sets maximum radius . . .

- Tim

Quote
tadawson
What you have never mentioned is exactly what you are doing (or trying to do) that causes this to happen?

Not knowing the cause, makes it very difficult to be of much help, although the simplest is simply trying to print thjngs that are too big. I suppose that if you errantly set up your slicer with 0,0 on a corner and not in the center (as it should be), it would offset everything and likely do this as well. But again, not kjowing what you are doing, I can only guess . . .

And there is a setting in Marlin (well, the current stuff anyhow) that sets maximum radius . . .

- Tim

Tim,
Sorry for such a long absence, i got in a car crash and i was in and out of the hospital not really able to function.. so i basically abandoned printrun and tried repeaterhost and everythimg seems to work fine!! Besides the fact that my printer wont extrude.. the printer looks like its doing a dry run but i need it to extrude!! Any insight on this issue?