Matt's MP3DP Repeat Build

If you are unsure I have pics up as well on a few of my builds. You can just unplug it for now and flash your board.

I will correct that diagram ASAP.

Always verify the pins on the bltouch. Depending on version number the wires can be different. I had to rewire mine. Then if you mix in clones, who knows which version itā€™s a clone of.

1 Like

I got everything working! The BLTouch wiring color convention was all over the place. I had to rewire and everything works now. When I say ā€œeverything worksā€, I have only really tested movement up to this point. I also tested that bed heats up, which it does. I will do a test print at a later date.

Next question. I ran the Auto Z-Align function. That function probed the 3 points for like 10mins straight. What exactly is that function trying to accomplish?

1 Like

G34 is hardware bed leveling, true bed leveling. It tries to get those threes points planar with the nozzle.

Once it is planar it then runs G29, that is mesh bed leveling to work with any warp the bed might have, that fades out over 5-10mm so the first layers could possibly be really off but you will end up with perfectly flat layers eventually.

1 Like

I have a different printer, but same true bed leveling. Sometimes it gets it in one go, sometimes it take a few times. Iā€™ve also had it give up once. I havenā€™t had that issue since I went to a different probe. From what Iā€™ve seen if the bed is still heating while it probes, it can throw off the measurements since the bed is moving with the heat. You can also accidentally set the acceptable deviation too high and itā€™ll never get it because the probe canā€™t measure that fine. I have my tolerance set to 0.008000mm.

3 Likes

Watch your terminal, you will see all the probing points, the response and average between probes until it gets to you desired tolerance mark

1 Like

Someone tell me if this is going to be an issue. Is it common to have this much play?

Man, thatā€™s a lot of stuff cantilevered out there.

2 Likes

Seems a bit loose.

Thought so. I will see if i can tighten it up a bit later. Im not next to it at the moment but if remember correctly there is just the one M5 screw that tightens the core around the CF tube, correct? I will see if I have any more room available to tighten up. I will also double check the CF tube dimension, but I am confident that is correct as it was very difficult to fit into the left and right trucks, so I would think the tube wouldnā€™t need to be any bigger.

1 Like

From that clip I can still see a large gap to tighten it.

Mine has some play like that, but not that much. i concur that a little bit of tightening will clean that up.

I can say though that you could still get excellent results, just make sure that the fan shroud is still clearing the print at the corner. I ran mine for months before I realized that Iā€™d forgotten to install the captive nuts in the trucks, so the rail itself was kind of loose. I might never have noticed had I not had a problem with my BLTouch which cascaded into other problems.

2 Likes

I definitely had some room to tighten it up. It is much less wobbly now. I guess I was nervous to tighten too much before out of fear of cracking something. That problem is now solved.

My next issue is the extruder was making weird noises when trying to extrude some filament. I will need to take apart the print head/hemera to investigate. Iā€™m ready to start printingā€¦

  1. Why am I getting this error?
  2. Why does it set Z to almost an inch? It is clearly not an inch from the bed. If were to tell it to go to Z=0 then iā€™m sure it would go crashing into the bed.

Try not to fall asleep while the bed is rising.

Waaaaaaaaaaaait a minute. Are all of the repeat printers homing z like this?

You mean homing to the bottom? Mine does.

Actually when I hit the ā€œlevel bedā€ in the menu, I get something that looks very much similar, except that it doesnā€™t give a probing error, and goes on to complete the G34 sequence.

A couple things that I might check:

Create a Gcode file with a sequence that will take a little less time. My startup code that I implemented in PrusaSlicer is:

G28 ; home all
G0 Z30 F3000 ; Raise the bed to near probing distance quickly
G34 ; True bed levelling
G29 ; Bilinear mesh level

If you put a gcode file like that on your SD card it might tell you a few things and take a bit less time if itā€™s going to error out. Also, you might get more information in touchscreen mode using the terminal, or entering the commands individually from your gcode senderā€¦

Aside from that, Iā€™d check that the black and white wire for the BLTouch is set up with black to ground and white to the sensor pin. Also that your Z probe offset is reasonable.

Iā€™m running the 513RP firmware on my SKR Pro, but Iā€™ve tweaked the source and recompiled a few times, particularly for build volume dimensions. If your Z dimension is significantly off, it might not like the height where the Z probe stops.

Weird. My triple axis style printer uses the bltouch (well, euclid probe) to home z at the top.

I think Ryan did it this way so that the bed could start from being way off. still having the 34mm offset from probe to nozzle.

It works reasonably well. I have it so that itā€™s within a couple mm from being correct when homed (I know thatā€™s a long ways off in 3D printing terms) close enough that probing will fine-tune it.

I did some playing around with M666 to see if I could get it closer for the G34 leveling, but I wasnā€™t getting any good results from it. My G34 leveling is about 1.25mm variance across the bed surface, so it kind of bothers me. I want to try re-making the belt triggers to see if I can get it closer, but that hasnā€™t come to the top of the priority queue yet.

I was hoping to implement power loss recovery. So if you home from the bottom, you can resume. The beds cool off so fast though, the part would surely detach by the time you resumed.