Finding the best Z-stop on a TAZ 5

(This is my first post, so please be gentle.)

I’m (very) new to 3D printing and after some struggles with a blocked head, some extrusion issues, adherence issues, temperature irregularities, and a bad power supply (quickly and kindly replaced by the folks at LulzBot), I wanted to put some time into getting a good z-stop.

I’m hoping to get some feedback on whether or not this is a good process, or if there might be something better. (I searched, but didn’t find any really good current posts.)

I started with the standard “paper between the extruder and bed,” but that seemed a bit “fuzzy” to me and I had trouble with comparing the example lines (really too close, too close, just right, too far, way too far) to what I had.

I built a quick little 2cm x 2cm x 1mm solid box in Blender, exported it as an STL (
untitled-2cm-2cm-1mm.stl (1.06 KB)
), opened in Cura, created the GCODE (
untitled-2cm-2cm-1mm ABS high.gcode (24.5 KB)
), and printed it with Pronterface.

I ran it a couple times to make sure I was getting a consistent print before changing anything.
I used ABS sourced from LulzBot with an extruder temp of 260 (EDIT: CORRECTION: 240), bed temp of 85, and quality setting in Cura of High (which appears to translate to 50mm/min).

Between each pass, I extruded 7mm @ 100mm/min to make sure the extruder was good and primed and I wiped the bed off with alcohol on a cotton round (usually used for removing makeup).

Here’s the first pass (with the “paper-measured z-stop”):

The Z-stop looked a bit too high, so I adjusted it about a 1/4 turn lower:

This seemed better - fewer ridges and a little flatter.

I thought I could better, so another 1/4 turn lower:

This one I didn’t like. I don’t have a picture of the first layer, but it seemed the extruder was dragging the first layer around, and the part curled off of the bed a bit.

I raised the Z-stop about an 1/4 [EDIT: CORRECTION: 1/8) turn:

This one seems about right to me - fairly flat. With the pad of your finger it actually feels smooth, but if you run your fingernail over it, you can feel the ridges.

I’m guessing the artifacts you can see on the top are from speed, but I’m not sure.

Anyway… feedback would be appreciated, and if this is a good procedure, then hopefully it may help someone else out.

Hello, and welcome. First of all I am new also. I have had my printer for just over a month. But I have done extensive reading here on this forum and others. I have also done a lot of practicing and have been getting progressively better. So let me help you with what I have learned so far.

First from the looks of things your z end stop is set too low. If you haven’t already, read mhackney’s guides that are stickied at the top of the forum. He includes a practice piece to print that is one layer thick that is 75mm around and 2mm thick. This is a really good piece to start with as it helps you recognize when you are getting a good first layer. I mostly print this piece every day, before I start to print anything else. This helps me warm up my printer and lets me know if anything is off. When I am really nailing the first layer, the back of it is completely smooth and the top looks almost like a sheet of paper. When I see this it lets me know I’m ready to print. Read the guides and try the test prints.

Secondly, although I haven’t printed in abs, your temps seem like they may be a little on the high side from what I have seen recommended. I would try to lower the temp to the lowest temp that you can get a consistent extrusion with, and start from there. Also you can raise the bed temp to 110. I have only used hips and I have discovered that this temp varies from roll to roll and color to color. My black hips printed perfectly at 230. The white hips would only print at 240. So it’s best to do some experimentation and find out what temp works best for you.

One of the biggest things that helped me was getting a dial indicator to level the bed. There is a thread in this forum about it. That improved my print quality by a lot. Getting the bed level first and then just dialing in the z stop height worked so much better for me than doing it in reverse. Like I said, your z end stop looks a little low. Whenever I printed out mhackney’s test part with the nozzle too low, I could always tell. It looks just like you said, like the nozzle is dragging through the part. I raise my in tiny increments and then retest and check the quality until I am happy. Too close and you will get the drag lines, too far and the lines will look like they aren’t sticking to each other. So make adjustments in very small increments.

Hope that helps. I’m sure the more experienced guys will chime in soon. Good luck with your printing.

I actually like the 2nd version you posted. The 3rd and 4th look a little too squashed. Instead of visually trying to determine the z-stop, I usually reference the adhesion to the bed. As long as the part adheres well to the bed, you’ve probably have a good z-stop.

The bed calibration gcode on the SD card is a great tool also. The pattern covers the whole bed and is only a few strands. When you remove, you can tell which part of the bed isn’t sticking well. Since its g-code from the factory, its a good benchmark for speed also…

The z-stop knob has always felt a bit “sloppy”. With the little play of the threads and the spring, its hard yo get really precise. Get it close and optionally use the z-offset in a slicing software to fine-tune the start height. For prints with a lot of surface area, I adjust by .1 to allow easier removal from the PEI… your never going to get .1mm from the end stop knob.

Get the dial gauge, its helped me greatly with leveling the bed also. I came from a printer which had an autoprobe and a lot of attention was placed on the “squash” of the first layer. It needs to be oval for best adhesion (think of the patch of tire on the pavement, too little bad traction). But too much an you’ll lose initial layer detail and the first layer of the project will be “mushroomed”.

OK… I’m rambling now. :slight_smile:. You’re on the right track.

One last tidbit… If the top layer is rough or raised it could mean to much filament being extruded. Adjust through the flow / extrusion multiplier in the slicing software. The other method is to spot check the diameter of your filament… Often times it won’t be 3mm.

If too much filament being extrudes, the nozzle will start hitting your print and possibly knocking it off the bed.

kcchen is right the second one looks best for the bottom. And the top is over extruding a bit. Set the extrusion multiplier to 0.9 or 0.85 and see how it looks.

BaltimoreBully - thanks so much for the response.
I did read mhackney’s guides, but I’m reading them again.

For the temps - you’re right, 260 was a typo, I was actually working with 240 based on the TAZ Slic3r Profiles (I edited the original post just in case someone like me comes back to read it later.) I’m still a bit confused about where to start since the Filament Guide suggests an extrusion temp of 225-230, the product page suggests 230, and the quick start guide suggests 240 (see page 6, step 3). I’m also really gun-shy of the lower temps since I started with HIPS and had nothing but issues - it managed to block up my extruder to the point where I was sweating disassembling it to try to get the HIPS back out…but that’s another story (though, I think it was all related to the idler bolt tension).
As for the bed temp, I saw the slic3r profiles page suggests 110 (from my TAZ 5), but I also found a post suggesting that LulzBot wanted to reprint back to 85. After looking at the date, though, I’m pretty sure this is for the TAZ 1-4 and not for the 5, so I’ll up to 110 and see how that goes. Oh, and the quick start guide also says to use 85.

I did see the dial indicator level somewhere. I’ll poke around again on that one.

kcchen_00 - All but one (the 3rd) of those parts adhered just fine, so I wasn’t sure if that was a valid gauge of “close enough.”
I did run through the bed calibration gcode a few times, but other than it sticking fine and looking about the same all over (I can’t really tell any difference), I figured it was more intended for indicating that “hey, your bed is way out of level” or “hey, your z-stop is way too high (or too low).” That’s why I opted to try something else.

I’ll look for the z-offset, that sounds like a fantastic idea.

I actually tried this using the Estep calibration, but I got wildly different numbers every time I tried it. (by “wildly” I mean that extruding 100mm @ 100mm/min, I got numbers between 95 and 105) So when I tried to update the TAZ on the LCD to the updated Estep value (currently, and originally at 850 BTW), it was off even farther. I went back to the factory settings (850).

I’ll fiddle with the slicing software next - nice idea.


First, I backed off the head temp to 230 and just extruded a few times to make sure it wasn’t going to plug up. That seems to be working, so I’ll work at 230 for a while.

I cranked the z-stop back up to where y’all suggested it looked best and I tried mhackney’s .6mm calibration cylinder (from here). I couldn’t get Cura to create any gcode that would print, so I switched over to slic3r directly and printing with Pronterface. (I’ll come back to Cura later.) It was adhering just fine, but was too high. I started playing with the extrusion multiplier as tmorris9 suggested. I got it down to 0.7 and thought to myself that was too far for its intended purpose. So I went back to the extruder calibration and tried it a little more thoroughly and systematically. (meaning: 3 times each before any adjustments) I ended up with a an Esteps/mm of 914. Sounds huge to me, but seems to be working well and consistently.

Then I went back to the cylinder. I was still getting heights of around 0.79mm, though the skirt was measuring a relatively nice 0.35-0.4mm (skipping the starting and stopping points). I figure that means it’s doing pretty well.
But I was troubled by mhackney’s statement that it should be about 0.6mm high and I was getting close to 0.79mm consistently (even after backing off the extrusion multiplier a couple times).
I went back to the math. If the first layer is 0.4mm and each successive layer is also 0.4mm, then a 2-layer object can’t really be anything other than about 0.8mm (0.4 + 0.4 = 0.8). I figure that means I’m on the right track.

Now to move on to the other models and see how they behave.

Oh, and I’ve still got the bed set at 85 and adhering well so I’ll try that for a bit, but monitor the adhesion.

Thanks, again, for the advice, folks! And do let me know if I’m off the mark somewhere…

Edit: I’m a little slow… I adjusted the slic3r settings so the first layer is 0.35mm and the subsequent layers are 0.1mm. Now, the skit is 0.35-0.4mm and the cylinder is exactly 0.6mm +/- 0.01mm. :smiley:

Nice job with the calibration. So just be careful, you’re not too far on the cool side. It may create weak bond between the new layer and previous layer… which could contribute to delamination / splitting higher up in the print.

Just think each filament color / bramd could have its own tweaks. :slight_smile: something to look forward to.

But you’re definitely on the right track.