First Layer Help

[taggedzi]

Pardon my ignorance here, I tried to search the forums but I I wasn't sure what to call the problem I was getting. I searched around on google/youtube a bit but didn't find exactly what I was looking for so I could use a bit of help.

Info:
RMax v3 Firmware: 20170109
Connected via USB to Windows 10
Using: PLA Noz Temp: 210 Bed Temp 60
Stock Standard settings from latest Mater Controller 1.6, only changes:
Outer Surface perimeters 2 count or mm
Fill density: 25% pattern hexagon

Today I adjusted the Z-probe height in the eeprom, and found it didn't change anything at first I made small changes, and in frustration the number kept getting bigger and bigger until I tried +/- 10 and nothing change. So I set it back to 0, and tried to adjust the Z max length instead. I was having a problem that the nozzle was getting WAY to close to the glass and the first layers were barely coming out at all. But once I started to adjust the Z max length I started to notice some real changes. I backed the length up about .225 mm and found that worked pretty good. My first layer height is now (on average anywhere from 1mm to 5mm but mostly 1, 2, and 3s) about 2mm height when I use standard mode PLA settings.

Problem:
Problem only seems to happen when the print nozzle is starting to print after moving a distance. Once it starts printing, it continues to do so just fine, but if it picks up and moves, then starts to print again, when it starts to print again it is VERY thin for a while. Not thin like too close to the glass thin, thin like a thin bead out of the nozzle thin. Here is a print I just finished the first layer of to do some testing:

There are 2 places in the print that have real problems, the base of the V in the center of the wing (this is some small air plane), and the start of the tail, where it is under the left side of the wing. Both places are where it was printing perfectly fine then picked up and moved a good ways, and when it set back down started to print too thin. I did not let it finish the first outer layer for the tail, it started at the thinnest spot, and then suddenly got thicker at the base of the tail.

I think the rest of the layers look really good to me, but if you see anything that needs adjusting, I'll gladly take pointers.

I don't know what to call the problem, so I'm not sure what to search for. Gladly take any input on fixing.

Thanks much,
Matthew (TaggedZi)

[Xenocrates]

I'm betting it's a combination of your retract settings and your extra length on retraction setting. Try putting a small positive value in extra length on retract, and increase it until the problem goes away. If it's not doing something, it may be something else entirely. I'm sorry to say that since I've never had the problem myself I'm not concrete on the troubleshooting for it, but that seems like the best option to try.

[DeltaCon]

Today I adjusted the Z-probe height in the eeprom, and found it didn't change anything at first I made small changes, and in frustration the number kept getting bigger and bigger until I tried +/- 10 and nothing change. So I set it back to 0, and tried to adjust the Z max length instead. I was having a problem that the nozzle was getting WAY to close to the glass and the first layers were barely coming out at all. But once I started to adjust the Z max length I started to notice some real changes. I backed the length up about .225 mm and found that worked pretty good. My first layer height is now (on average anywhere from 1mm to 5mm but mostly 1, 2, and 3s) about 2mm height when I use standard mode PLA settings.

I really can't bake any chocolate from that!
(interested in one on one translations of Dutch proverbs and sayings? Goto dutchspeakwords.com, it's hilarious! At least when you're Dutch...)

You are talking about first layer heights of 1 to 5 mm what usually is somewhere between 0.2 or 0.3 mm. Z-Max Length is calculated by auto calibration, so it is not really useful to do manual changes there, they last only until the next calibration or power on. Z-Probe Height is used as a correction of the automatically calculated Z-Max Length if it needs tweaking (and it does). Z-Max Length is only used to correct a calibration, so changing it does not lead to any immediate effect of you do not re-calibrate.

So auto calibrate, take a thin piece of paper, jog to Z0 with the paper under the nozzle. You must be able to barely move the paper with slight force. If not, adjust the Z-Probe Height and recalibrate until it does. Now after an autocal you have a correct Z-Max Length without manual intervention, that is always consistent. Care that you do not have any Z-offsets in your slicer (you have just taken care of that in your calibration routine!) Check your first layer height in your slicer settings. Using a standard nozzle I believe that 0.2 or 0.3mm should do it. Now if the machine travels to Z0.3, it REALLY goes to Z0.3 and you should be able to get decent first layers.

Your final calibration (always before you start printing) should be done at least with a heated bed, preferably also with a heated nozzle. However that is not advised by SeeMeCNC because of the way the probe works. I advise to heat up to working temps. Kill the heaters and immediately start calibration.

[taggedzi]

Ok, I understand I shot a lot out there and probably didn't help much.

I have reset everything back to "stock", and tried a couple of tests. I think my problem is due to thermal expansion.

IF I autocalibrate and print on a COLD plate, my first layer comes out nearly perfect, I have tried it 3 times now.

IF I autocalibrate on a COLD plate, and heat the plate, then print I have all kinds of problems, the EDGES tend to print better, but the center is so close it impacts in some places.

I'm going to try your advice of heating everything up, turning OFF the heaters and auto calibrating. maybe this will fix the problem. I just know I've beat my head against a wall trying to figure this one out.

Thanks,
Matthew (TaggedZi)

[taggedzi]

Hey there,

Figured I'd check back in. I found a solution that works for me. I am now getting AMAZING first layers.

I downloaded the firmware from github and modified the Z_PROBE_HEIGHT in the Configuration.h file. I set it to -0.35. (6 places in the file I think) Then installed the firmware in the RMax.

I run the auto calibrate on a COLD bed with a COLD nozzle. Then I heat up and run with pretty much stock settings in matter controller, or cura. This works wonderfully for me. Why I need to do this? I'm not exactly sure. IF I'm running on a cold bed I don't need too.

When I set the Z_Probe_height in Matter Controller OR Cura it does not seem to listen to the settings at all. I don't know why that is, I've tried putting in ridiculous numbers (+/-20)... an no difference. But, when I change it in the firmware... it works.

Don't know if that helps anyone or not. But it helped me.

Matthew (Taggedzi)

[geneb]

...or you could change the Z Probe height in the EEPROM using MatterControl...

g.

[taggedzi]

Hey geneb,

I was changing the setting in Matter Controller on the EEPROM settings screen. I found the changes I made there seemed to have no effect on the z probe height (the first layer, where that would show up). That is where I tried quite a LOT of experimenting with z probe height settings with no difference in outcome until I just started getting crazy with it +/- 10 (no decimal there), even that showed no difference at all in my prints. I don't know if I am using it wrong, or ??? but I would connect to the printer, go to Options -> Configure (next to EEProm) and find the Z Probe Height and send. Initially it was set to zero (on the firmware I started with). The latest has it at -0.1. I tried using auto calibrate both before and/or after doing this and found it didn't matter what I set the number to, it never changed the first layer height. Other changes I made in the EEProm did show up and impact prints, the only one that did NOT seem to take was the Z probe height.

When I updated to the latest firmware and found that it made a slight difference in my layer height and I got a little hopeful. So I looked at the diff in github, and saw they changed the Z Probe Height in the firmware. So... I took a wild leap from there and would adjust it in the firmware and upload. All the changes I made there showed up in the printer as expected. After about 10 test runs (a lot of time and my calipers) I finally settled on -0.350 for my machine with a hot bed. That setting got me what I think is a perfect first layer, edges perfectly blended, no bulges, not too squished, not too spread out good adhesion to the bed, and to each other.

Again, I don't know if I'm using Matter Controller wrong. Perhaps I am the next time I connect my pc to the the printer I'll get a screen shot of what I was doing and perhaps I can be guided in the right direction. Unfortunately I can't open the EEProm config screen with the printer not connected and I have moved it to the other room printing right now.

I have a strong background in engineering, machining, welding and programming. But I know absolutely nothing about 3d printing, and less about Delta printers. So I'll gladly take correction, or resources that can educate me better.

Matthew (TaggedZi)

[geneb]

Did you re-run G29 after changing the z-probe height each time?

The parameter in the EEPROM is the same as the one #defined in Configuration.h - if you change it there, you really only change it in the one block that pertains to your printer, not all of them.

g.

[DeltaCon]

I was changing the setting in Matter Controller on the EEPROM settings screen. I found the changes I made there seemed to have no effect on the z probe height (the first layer, where that would show up).

Yes, that's what I said. It does not change anything UNTIL a new calibration is run. Z-Probe-Height is ONLY effecting the calculations of the new Z-Max-Height AFTER a calibration. So you will see no difference until after a new calibration.

[taggedzi]

I could swear I did. But it was late, and i honestly can't remember for sure. So, today when i get home I'll hook it back up to the pc, flash it to the stock eeprom, set the z probe height, auto calibrate, and test. I'm willing to give it a shot again maybe i did forget to auto calibrate. Sad thing is i can't remember i tried so many things.

I'll let you know, my findings either way. Thanks for the help and patience.

Matthew

[taggedzi]

Well. Results are in:

I must have not run the autocalibrate after updating the changes to the EEProm. My apologies, I do remember running autocalibrate a lot that evening, but I must not have done so because that evening. Tonight when I uploaded the "stock" firmware from github (master branch, pull from tonight) and autocalibrated, it took the settings as I entered them.

I'll duck my head in shame for a bit... and get back to printing. Thanks for the help and hopefully patience.
Matthew (TaggedZi)

[taggedzi]

Well, i thought i was done...

Happy i had a perfect first layer for a test cube, i moved the printer back to its location. I auto calibrated twice... and printed something much larger. I am getting varying layer heights across the bed again. All of them are to low. I adjusted height via the menu, but it is not even across the plate. i checked the obvious points i could think of and found nothing to fix.

I checked all the screws holding the bed and plate. They seemed snug and evenly applied.
I checked the belts, they were tight, but not snappy.
I checked the towers they were in tight and did not seem to be moving.
I ran autocalibration a few times and double checked that the nozzle and bed were clean.

Any ideas? This is the same problem i started this thread with. I am not sure what changed to make it work/stop other than moving. I guess I'll just tighten everything more... and see how that goes.

[geneb]

If you've got a steel ruler handy, sit your build plate on a countertop and use the edge of the ruler to see if you've got any weird spots on the glass.

If you've got a raspberry pi handy, you might want to try the delta-cal wizard for it.

g.

[DeltaCon]

Did you run the calibration with heated (but shut off) bed and hotend?
The bed can raise a bit when heated, so you should calibrate at the temp you want it during print.
I know SeeMe tells you to calibrate cold, but it is really better to calibrate hot. It just that the heaters must be shut off before probing.

[taggedzi]

@geneb: I found 1 side had a very slight concavity to it, the other was without problem. So I marked the good side and it seemed to help some. I used a steel straight edge, and then switched to a piece of finished/highly polished) slate marble. I got it wet, and slid the glass on it from the side, covered with water. The concavity stood out very clearly (as a bubble that could not be "pushed" out), When the side with concavity was down it generated quite the vacuum when I pushed on it trying to remove it took a bit of consideration. The flat side, while it generated some vacuum did not "suck" down onto the plate when pressed. Also I do not have access to a raspberry pi at the moment, it is on my to-get list though.

@DeltaCon: I ran some tests and took some pics just to show. The first layer with auto calibration COLD, the first layer with auto calibration HOT, and the first layer with manual tweaks to the HOT auto calibration.

AutoCalibration with a COLD bed and nozzle. Printing with hot bed and hot nozzle.

AutoCalibration with a HOT bed, and hot nozzle (not actively heated at the time of calibration). Printing on hot bed with nozzle obviously hot...

Manual modification to the Z-Height via the navigation menu after HOT auto calibration.

These are items with SMALL foot prints, so far I have not had issues with small foot print items changing layer height across the bed. When my current print is done. (prob an hour or two). I will attempt a larger foot print item and see how the first layer goes down over a large area. I THINK it may be fixed due to a minor imperfection in the glass. I have manually tweaked it JUST a bit more beyond the pictures, but I think I'm dialing it in.

I'll keep trying to get it straightened out. I know I'll get there. I just would like to try to get my consistency up, and not have to tweak so much between changes. I have done a couple nice prints once I got it calibrated. (last time around).

As usual thanks guys for the input and I'll keep working at it.
Matthew (TaggedZi)

[taggedzi]

Sorry for all the "add on's" but I have tried printing something with a larger foot print at my son's request. My first print with the same settings as before, with NOTHING changed (I hadn't even moved the glass bed.) came out awful. I tried to manually adjust a bit, but that did not seem to help. So I auto calibrated twice, and ran again. When I saw the prints were coming out bad, I tried to adjust the Z height. It helped but I'm getting uneven results. One end of the bed is too far OFF the platform, and the other is squished down on to it. Here is a pic to show:

Auto calibrated Hot bed, hot nozzle. No elements in/on or near the printer were physically moved out of place. One end is squished (z tower) the other end is to high above the plate (x/y towers).

This is the type of thing I was getting before, and it was driving me crazy. I'm going to try flipping the glass over and make sure I didn't mark the wrong side of the glass when I found the concavity... but that is all I can think of.

Matthew (TaggedZi)

[taggedzi]

I have taken a step back, taken my time and re-measured and checked a lot. I cannot for the life of me get a good first layer, all the way around. I can get it good in one place, but not all the way through the plate.

Yesterday, I attempted to auto calibrate a few times. I tried the calibration wizard, and got 1 good first layer. But as soon as I printed something else, it was off again. So I completely disassembled the top of the unit and checked for anything lose, or out of place. Taking great care to make sure everything looked the way I thought it was in the instructions. I verified the extruded aluminum was touching the top of the brackets and was in very tight. I verified the screws connecting the Top assembly to the brackets holding the aluminum extrusions was in nice and tight.

Today I removed the hot end from the towers, cleaned the nozzle, and verified everything on there seemed to be right and connected it back up the way it should be. While I had it down I took the liberty to check the carriages. There was some play in them (in and out radially at the tops but it was minimal 1-2mm max... I don't know how much there should be if any) Basically the tops of them wiggled outward if there was any pressure applied outwardly, but they went right back into place if the pressure stopped. The adjustment screws on the top were about level to the top when calibrated (2 of them were even with the top assembly the other was down about 1mm, don't know if that makes a difference).

If it is still printing poorly tomorrow I'll disassemble the base and try there...

Here is a video of one of my calibration prints, sorry for the rotation half way through the phone seemed to think I wanted to switch orientations.
https://goo.gl/photos/XUjbraMD4Kp8Cf5QA

Is this just my glass bowed? Is there something not setup right on the machine? When I auto calibrate it keeps bottoming out in the middle. I adjust it up. But I get mixed results. Some times OK, some times awful. If I adjust it up it is perfect in the middle for "small" prints, but if I print anything large it bottoms out in the middle and doesn't touch at the ends. This happens even when I flip the glass over.

I was "enthusiastic" when I first got the printer because I wanted to see what it could do, and I have idea's I was/am hoping to work on. But I'm more than willing to take my time and do it right. But I certainly need some help here. I've been reading calibration guides online, I've been tempted to try to do a manual calibration to see if I can do it to either figure out what is off in my machine, or maybe just fix it out right? I'm open to ideas.

Matthew (TaggedZi)

[geneb]

Matt, I'd check the carriages to make sure the screws are fully tightened. There's a gap of about ~1mm between the carriage halves when it's fully tightened. I just checked mine for motion and there's a tiny bit, but requires a non-trivial amount of force to move.

g.

[taggedzi]

@geneb

Thanks so much. I checked the carriages, 2 of the the screws were "tight" hard to turn, but not fully seated on the X-axis carriage. I tightened them all the way down, and it has helped significantly. I autocalibrated, then used the calibration wizard on the site and it has got SO much better. I'm was getting a bit of "feed" inconsistency too, that was causing a problem. I opened up the extruder and found that there were a good deal of shavings up there, I blew it out, and turned down the feed rate just a touch on my prints. And so far... (fingers crossed) the first layers are worlds better. Still not perfect, which I figure I'll always have to "shoot" for, but I'm thrilled with the results now.

I also did adjust the z-probe height to -0.18 and now calibrate cold, and that seems to get me almost perfect first layer height.

Thanks SO much, again.
Matthew (TaggedZi)

[geneb]

Glad you got it sorted. Keep in mind that ground filament in the extruder may be indicative of an improperly seated bowden tube in the hot end - make sure that sucker is fully seated, otherwise you're going to have filament starvation and jamming issues.

g.