[Return to main site] 

Z axis compression... constant failures

Hi all,

I finally got up the nerve to start working with the Titan, but it has been nothing but frustration. I haven’t gotten even close to a print out.

Mind you, I need to pigment my resin as I’m using MJ G+ Black that came with the printer.

The main problem seems like it never gets higher in the Z axis. I’m running the Yoda sample at the basic parameters. The image below is an example of 25% of the print. This should have a 1cm height at this point. But it seems like the build plate (I’m using the small one) continues back to the same place in the vat. I’m at my wits end on this as I really don’t know what to try. Any help would be greatly appreciated. Thanks!

Can you post a screenshot of your print settings in the Kudo3D host software?

Does pressing the +/- 1mm button in the Kudo3D host software cause the motor to turn and raise/lower the build plate?

How many slices are in the slices folder that you’re printing from?

Thanks for the reply!

Motor control works with the software. +/- 1mm works as well as the other choices.

I’ve attached a print screen of my settings. They are the basic parameters. I’m running an xy of 44 right now. I wasn’t able to achieve a clear focus at 37, so I measured my clearest small grid which was 85 x 47. The slices folder holds 397 slices. It just seems that the build plate continues to go up and down to the same level at a period of time and I’m wondering why.

Honestly, I’m quite new to the printing side as compared to SLS 3d scanning. So any advice you can give is more than appreciated. If I should be looking at other resin as well as augmenting my parameters, please let me know.

Do you have some of the text from the System Output pane from when the printer was printing and exhibiting this behavior?

I don’t at this point. I will have to run the print again. There was no error messages during the print. However I did have two prints that just stopped mid print. I’ve attached the log for those.

I will rework the Yoda in CW and save a new folder of images. Then I’ll replicate the set up above and run it and take a screengrab if that helps.

I’ve order pigment and more resin from Makerjuice as well. I just want to get something (even a partial) out.

Let me know if you can think of anything else.

retrying the attachment…

Are those python errors in the text file in the zip also what you’re seeing in the System Output pane to the right of the settings area in the Kudo3D host software?

The errors in the log say the assertion failed due to invalid images… I’m not sure why that would cause the software to return to the same z-height all the time, perhaps an unrelated issue but the software has some weird bugs so maybe. Assuming the invalid image thing is the problem –

1st, close and restart the Kudo3D software every time you want to start a new print. Something may not be getting initialized correctly so you want to eliminate that as a possibility (I restart mine every time because it sometimes does weird things if I don’t).

Have you viewed each of the images in, say, the Windows photo viewer to see if it will show all of them? You could also right click the files to get the properties to see if the Width of the image as reported by Windows is valid (pain in the butt but just throwing ideas out there, maybe slice a thin object to reduce the # of files to look at).

Are your images on a flash (aka usb / thumb) drive or network drive where perhaps it’s inaccessible for a moment and the computer can’t load the image?

Your images have reasonable names? EG if you name your CW file Model1.cws the image files are called Model10000.png, Model10001.png etc which causes problems.

Your computer’s power settings are set to always on regardless of whether it is plugged in or not (so it isn’t trying to dim the display, turn off the HDD or USB ports etc after a few minutes of non-use).

Have you tried a different version of the Kudo3D software?

You said the build plate will move when you manually hit the buttons – is this when the plate is in the air, or sitting in the resin? The force to lift it when it is in the resin is larger than when it is in the air, so long shot here but you could potentially increase the stepper motor current by adjusting the little pot on the RAMPS board’s stepper controller. But, read the manual on that first before you go tweaking it to avoid any problems… just a thought.

Hmm maybe also try a slower down speed? Mine is 150mm/min, haven’t tried faster than that.

Given the viscosity on G+, I think it’s fairly unlikely that it’s a motor current issue. I run 200 mm/min even on much more viscous resins. If the stepper were skipping due to high load, there’d be a pretty noticeable sound that would be hard to ignore.

An image name parsing problem would manifest itself as the host software being unable to detect the current number of slices, but from the screenshot, it seems to be detecting the expected number of slices.

The log entries are undated, so it’s hard to tell whether they occurred during the print, but the log is small enough that it doesn’t seem to have occurred often enough to last through the duration of the print.

One thing that I do notice, however, is that you are using a layer thickness of 0.1mm (100 microns), using the baseline exposure time for 37 microns, and actually having 44 microns XY.

With G+, this may result in underexposure. Underexposure would not cause the build plate to lift and lower to the exact same location, but it could cause the print to not form above layer 50 and result in a remnant similar to the picture provided.

Because the build plate is only moving up a very small net amount per print, it can be difficult to tell whether the build plate is rising on any single layer. After 100 layers, however, it should be noticeably higher.

Regardless of the movement issues, I think you will need to either reduce your layer thickness (the recommended settings are for approximately equal XY and Z and you’ll need to re-slice) or increase your exposure times, probably by about double if you want to keep the current model.

Thanks for all the replies!

Just woke up to start again.

I recalibrated the printer to 37xy last night. I will start with the Kudo calibration model today. I’ll rework my slicing to match at 37, check my naming convention and the images in the folder.

I’ll post all the info. when I get that far. Thanks again for all the replies. It’s hard when this is honestly my first journey into 3d printing. I appreciate the help.

Greg

Well, after 5 attempts of the Kudo calibration piece at 37xy and 37z, all I have gotten is the attachment base. Mind you, this thing is only 15mm high. But I’ve tried overexposing, changing life height…etc.

Attached is a shot of the pane with the last crazy overexposure settings.

I’m really confused on several things…

I thought we are to set the z-zero firmly on the base of the resin vat. Then I read this in the latest build manual…

Click the “Z-Zero-10”, “Z-Zero-1”, and “Z-Zero-0.1” buttons to lower the build platform until you hear the stopping sound. When you hear the stopping sound, the z-axis is at its lowest position that is now set to be the z-zero position. There should be a visible gap between the build platform and the bottom of the resin container.

Stopping sound? Visible gap? Does that mean not to rest the plate fully?

The amount of lift changing per levels is confusing. If initial layers aren’t releasing, is this overexposing or the need for more lift to separate?

I’m sorry if this comes of really stupid, I’m just trying to get my head around this. I’ve ordered new resin and pigmentation in hope of helping. If this is part of the problem, I can’t get over how Kudo sent us not only a inferior vat but also less than capable resin…especially if they knew this printer would need further pigmentation to print.

Sorry for my venting. I’m just frustrated. Thank you all for the help so far. There really is nowhere else to turn.

Greg

There should be no gap between the build plate and the vat floor at Z=0. If there’s a gap, the attachment layer won’t attach to the build plate and the print won’t form at all. This would also result in a flat remnant like you depicted in the beginning.

The Kudo3D build plate leveling procedure has you bottom out the Z stage motor, and then has you loosen the screws holding the arm to the Z-stage to drop the build plate and have gravity align the build plate with the surface of the vat. If you miss the step where you’re loosening the screws and dropping the plate, your build plate will be too far away from the vat floor to have the print attach properly.

When printing the 15mm part, did you visibly see that the build plate was around 15mm above its starting position when curing the last couple of layers.?

Sorry for the late reply. I didn’t get a notification and forgot to check.

Today, I’m going to go back and recheck my build plate calibration and try another run or two at the calibration piece.

Honestly, I never let it finish completely as it was obviously not working midway through. I’m going to go back and try base config with adjustments based on the printing guide, then I’ll try doubling my exposure to see how that affects it.

I’ll send some pics of what I have from these later in the day.

I have some more MJ G+ coming with pigments. For normal resin, would you recommend another brand? Funtodo?

Ok, time to see what’s in store for the day. Thanks.

I don’t really have a particular recommended resin. MJ G+ is fine if you like the material properties (less viscous, cures a little more brittle, but very hard).

FTD IB is a little more viscous, cures less brittle but softer. I haven’t yet tried FTD Standard, which is more price comparable to G+ and is softer still.

If you can afford it, you would probably want to print in 3DM ABS all the time because it holds much better detail (with no pigmentation added) and is both harder and stronger. Alas, it is both more expensive by the liter and requires more diligent cleanup.

Once you can actually produce a part, you’ll have a better idea of what the material properties are. It’s probably premature, at this point, to stock up on resins until you get some parts out.

Well another day of no success. I rechecked everything… levels, build plate, calibration…

The first time was a complete mess. The z showed that it reached 15mm, but it honestly hardly moved. I know this because when it was done, it pulled hard from the vat and went up 100mm. But when I went down slowly towards Zero for a second try, my Zero was off by almost 12mm.

First pic of the overexposing config and second pic the result.

The second try did get the base correct then it seemed the entire shape was squished into itself.

Third pic of lesser config and fourth the result.

I don’t know what variables to consider. My setup is now as exact as I could get it. My leveling was slightly off, but it was minimal. I guess I’m looking for a sweet spot for this resin in exposure and why I’m not getting any build height. Seems that it gets to a point where either the layers just squish into themselves, or create a mess and stick until a high lift at the end happens.

I can’t help but look back at people discussing adding pigmentation into the resin and hoping this will help. Getting a V2 vat would help, but I can’t see putting out over $100 (as it seems you have to buy them in pairs) for this machine until it does at least a partial success.

NOTE: I reached out to Jonathan and he’s sending my results to his “techies” as well.

Is this the Calibration part or the Yoda? Can you show a pic of how it’s sliced in CW perhaps? I have never pigmented my MJ-SF and though I’ve found the exposure times that Kudo3D recommends really good, I"ve adjusted them slightly. I’d imagine the MJ-G is close to the -SF in viscosity but probably not the same. I would try doing the Yoda at .1 or .05 mm and not worry about pigmentation for now. I have a batch of FTD-IB I’ve mixed up with Tap Plastic pigment but haven’t used it yet as my computer is sick right now, however stay with it with the stock resin and you’ll succeed. I just posted a thread in Showcase your prints of a torso I did. I titled it SM Print and Learn because I’m learning more I realize from my failures than my print successes. It’s nice to have a good print here and there too. But I wanted to mention that is printed at xy100µm and z50µm with MJ-SF non pigmented red.
The raft is exposed at 60sec with an up of 10mm/m then supports are 30sec at 15mm/m and mostly the print part is 14sec with an up of 25mm/m. Compare the Yoda Kudo3D settings but I think if you set something similar and make sure everything is actually faily level and operating as it should; you’ll get your Yoda print. The Calibration will be good too but this res is to low for the end fine details. Hope this helps point you in a base direction.

If the build plate isn’t rising the correct amount, neither pigmentation nor a new vat is going to help.

My suggestion in this case would be to run prints in an empty vat with no resin so that you can watch the behavior (or even better, video the behavior) and see exactly what the build is doing in Z.

Sorry, my post is irrelevant as I didn’t fully realize your Z linear stage may be malfunctioning.

I don’t know if it’s malfunctioning. The reason it was off was due (in my mind) to not separating from the vat and therefore the height was pulled off. I will do a test without resin as jkao suggested and make sure.

I have more MJ resin coming as well as pigmentation.

Also, I used the smaller build plate and I don’t know if this had more/less bearing on this. Jonathan stated that I should try a z of 35 instead of 37 when printing 37xy.

Unfortunately, I’ll be visiting NJ for two weeks starting Wed. and my resin will not arrive until after I leave. So just some non-resin testing probably til then.

Thanks for the advice!