Repeated print failure analysis

My replacement printer is working quite well. At least, I do get relatively successful prints now.

One thing I’ve noticed with a print failure today that it seems to always occur at the same point.

We all know that good supports make or break a print. As you can see here, the print failed very badly so I had to cancel it.

I seem to have issues with this build platform. It’s a refurbished unit and the base is woefully smooth. As usual, adhension is poorer away from the hinge side.

What’s interesting is that the supports all failed at the same level.

What is with this level???

Is this where the slow peel stops and the printer starts to do fast peel? It seem to have taken a shock that caused simultaneous failure of supports at that level. Looking at some more successful prints, the mark is also there.

Interestingly, when I measure the height from the bottom to that line, it’s exactly 5mm. Which is the same as this??

This feels like a software bug. If not, can the slow peel be gradually transitioned to faster peeling? This is not the first time I’ve seen such failure.

By the way, this is the part in preform. Please help. Support ticket submitted.

This has happened to me. One of the 3 models was already completed, and shortly after the other two sliced. I did NOT cancel this print, this is what it looked like on the platform, the rest of the model drawn onto the bottom of the tank. and looked like a cool cracker when I peeled it off. This support ticket closed without resolution.

Just to prove my point, I did another print with the same file but with the height above base changed to 4mm.

My build platform is meticulously cleaned each print. Mirror is pristine, PDMS scrapped each time and resin in tank combed to remove ‘floaters’.

These are the result.

@JoshK, I can’t tell that it’s the same problem as mine? Where are the supports breaking off in this print?

No, the problem did not manifest at the supports.

What’s kind of interesting, is that when I printed @JoshK’s calibration form1 file, the cubes exhibited a slight x & y size reduction exactly at 5mm z hieght. The size reduction was even on all sides, so that the bottom 5mm of the 1cm cube was about .010" larger than the top 5mm. Wondering if this is related. These cubes get built right on the platform without any supports.

@ChristopherBarr Thanks for the reply, guys. I’m definitely interested in testing @JoshK 's calibration file soon.

Anyway I did a 3rd print of the same file. I lowered the build platform using fine tuning and got much better bed adhension but it didn’t solve the support failure issue. It failed the exact same way.

So, I tried printing again using my grey resin. Same file. And it’s PERFECT.

But it can’t be a problem with my resin too because before the slew of failures, I printed this in clear which is my tallest print that is almost perfect.

Reply from support says,

To ensure that the print sticks well to build platform we double expose the first few layers of the print. Depending on the print, but on average around 4.5mm, the settings switch to one pass rather then two. It is not unusual for this transition make a visible make which is why we have it set in the software to occur before the model starts.

Which makes perfect sense.

But the recommended remedy is to clean the mirrors which don’t make sense since I can print in grey perfectly. My thought is that the laser is tuned to be the weakest among the different color resins (to control light leakage) for clear resin. The combination of that plus tear strength of the tilt caused the repeatable failure.

What I think should be a fix should be a software fix that progressively switch from double exposure to single.

I really don’t want to break open the printer to clean the mirrors and it’s a REALLY new printer too.


Last I tried to print clear parts, it was either breaking at the the support when the double pass stops or just ends up as a gooey puddle on the tank. It was hopeless.

I saw this post by @Monger_Designs that mentioned about using different resin setting and I thought I would give it a try.

I’m still using the same clear resin tank and sieved resin. Nothing fancy and set the type of resin to BLACK in preform. I wasn’t hopeful since clear parts have notoriously high failure rate…


Surface finish is still good and I finally can print in clear reliably!

I’m glad it worked out for you. But I think you should contact Formlabs about this, because technically since the clear setting doesn’t work for the clear resin for you, it’s possible that your laser lost power somehow or has degraded. The problem may get worse over time if that’s the case.

Yeah I did get in touch with support as mentioned in this thread. I think some mirror cleaning might be in order but I’m still waiting for some PEC-PAD so this can be a good reference for others having trouble with clear resins too.

Judging from what I read in the forum, there are quite a few.