Buildup above supported overhangs

I have recently printed some small parts, less than an inch in height. The parts have critical dimensions in x, y, and z axes. I have printed the parts oriented horizontal, vertical, and slanted at various angles. In X and Y, the dimensions are spot on, typically within .003"/ .076mm from nominal. However the Z is anywhere from .010"/.254mm to more than .030"/.762mm oversized in any feature that has an overhang, even if well supported with max density and including the first surface after the raft. It can be a hole and circular or just the beginning of a layer after a space. Angle does not matter. But the buildup is always on the side opposite the lightsource, and technically only on internal features for the build, including internal features caused by a support.

I am printing with the Form 4L due to the speed of the prints, however I tested on a Form 3/3+ (I think it is a 3+) and the problem is there too. I am currently using the High Temp (v2) resin, but the problem is also present in Rigid 4000 (which we use the most at our company) and Flame Retardant. I have not tested any other resin and do not have them available to use for this printer. I am trying to stick with this resin and printer since I used it as my baseline/control print.

I contacted support but they said that they can’t help me with fixing this and just said, “oh wow that is a bit excessive for that size of feature, you can try adjusting different custom settings but we can’t help you to fine tune your Z accuracy”.

It is likely that the reason for this build up is light leakage curing the liquid resin above the first layers of an overhang that is not meant to be cured. I have scoured the internet and have seen several posts about similar issues and their solution was to decrease the exposure for layer curing. I have not seen anything for the Form 4 yet, so I’m not sure what other people may have had success with this particular printer in terms of customized settings. It is very important for us to be able to print at better accuracy because many of our parts have tight tolerances. Some of our engineers have gotten fed up with the print quality and outsourced to Protolabs and Xometry with significantly better tolerances. Outsourcing has been very expensive for our company, and it is hard to justify our printers if we can’t even use our parts in our prototypes.

In the print settings editor in Preform, there are a number of exposure settings. There is one called overhang fill exposure, it is an array of exposures per layer. My interpretation of it is that each layer starting with the initial layer of an overhang, you can change the exposure and each next layer in the array is the one printed after. It was initially 81mj/cm2, but I did so far more than 7 prints with different settings lowering it. The last print I did had 5 layers set to 0mj/cm2, thinking that since the buildup is about .700mm and the layer thickness is .100mm, then remove several layers and see if offsets to the correct dimension. I have seen some people that had similar issues to this on other brands of printers and resins do just this, but they had added a script to their slicer to do it.

I printed the last print and it has exactly the same amount of buildup as the previous attempts, so I’m not sure if the setting even actually does anything. I have tried Z scaling and Z fine tuning with no success.

I made some test parts specifically to show this. I wasn’t able to get good pictures of them but here is an image of where the build up is: see highlighted areas.

Since the Form 4 is so different of a lightsource than the previous versions, has anyone played with the settings enough to dial in Z accuracy? Has anyone solved this problem on other printers with methods that may translate?
Thanks in advance!

Thanks for reaching out, @jagrata!

Our Support Ream and Engineers were discussing your situation. Because you are working with ultra-high precision, our default settings may not be best suited for your needs. We recommend leveraging the Print Settings Editor to tailor our settings to meet your requirements.

To address the concerns you’ve highlighted, you can adjust the Z correction factor to help compensate for small dimensional inaccuracies, and the irradiance setting to reduce light bleed at a minor expense to print speed.

Cheers,
Diana

Thanks for responding! Like I said, I had already contacted support and they also suggested using the print settings editor, and the Z correction factor. The Z correction factor did not seem to assist at all in this issue.

I’ve started experimenting with the exposure and irradiance.

For Formlabs, how is the the exposure energy density calculated?
I’ve seen irradiance x time before, but is that how it is calculated for the print editor settings in Preform or is there an additional transform?

In the Print Settings Editor, there is no exposure time setting like some other SLA printers.

There is supposed to be a correlation between irradiance and energy density, with exposure time being the main factor. However I don’t know how Formlabs determines exposure times.

I have changed the Irradiance from 10 to 2 mW/cm² and I have changed Fill Exposure from 22 to 12mJ/cm² and several combinations (about 8 prints with changed settings).

There is very little difference in anything other than the amount of time the job takes. I was able to shave off a few thousandths of the extra material by lowering the irradiance to 2mW/cm², but build time was about double.

The normal formula that I am aware of is Irradiance = Exposure Time x Energy Density. Based on how I changed the settings and how build time changed, this does not seem to be the case for Formlabs. For example, if I double the energy density, time changed only by 1 minute, but time changed significantly with changes to irradiance.

Is it possible to control exposure time (even indirectly) in Preform or does the program force a minimum amount of time based on irradiance and energy density for what Formlabs has decided?

Does anyone know the formula Formlabs utilizes between Irradiance, exposure time, and Energy Density?

Thanks