Variable Exposure Questions


How does variable exposure work to even out the exposure intensity? Does it make the smallest area full power and decrease larger areas or does it add photoexposure time onto the smaller areas and shut off light to the central areas?

I wish to print a lattice structure inside a larger cylinder. The lattice I print requires a greater photoexposure time to print successfully than the cylinder. If I try to print the combined object at the settings of the cylinder, the lattice isnt fully formed. If I try to print the combined object at the settings of the lattice, the cylinder jams and breaks.

Would variable exposure help? If so, should I set the layer exposure to that of the lattice or that of the cylinder?

Another option I have considered is simply exporting both objects separately from print studio, unzipping the file, intercalating the layers manually with the requisite parameters so that each layer of the combined object results from two separate photoexposures with no build motion between them.


The idea behind variable exposure was to limit the amount of light to large cross sectional areas of each slice.
My experience with Variable exposure is limited, and I’m not sure the variable exposure setting even works currently. The last information I had was that while it was a clickable option, it didn’t actually do anything.

What I ended up doing was post processing the images in Photoshop using the batch feature. I was able to create a custom ACTION that selected all the white values from the loaded image and apply a CUSTOM STYLE that generated a gradient from light to grey.

Once you have processed the images you need to combine the image stack into a ZIP file along with the print settings file. Upload this ZIP file to the ember using the IP address method.

hope this helps…


Another possible solutions would be to process each section individually - providing you can do this - and post process the cylinder images with a lower greyscale value, then merge the two images. Then you only have to print one layer.


Using your suggestion, you can inter-stitch the image layers but instead of not separating, you give the second exposure layer a 5 micron layer height and a long dwell time to let the resin evacuate more…I only say this because I’m not sure you can eliminate the slide separation PER LAYER. I thinks thats one option that’s all or nothing.

Either way - a lot of work in post. If variable exposure works it would be the easiest solution by far.


Thanks Scott

I’m not confident on the conversion of exposure time of “white” light to an equivalent exposure scale at an appropriate grey value. I think I’ll try your second suggestion and just do the alternating 5 micron layer. I’ll update to let other uses know if this worked. It could be a useful thing to script for multimaterial reasons as well.


I know there was a post (or instructable) that discussed the grey value in relation to exposure value, I just can’t pull it up off hand. Sems it might have been on the topic of sub-pixel resolution.

As for the multi material - I have contemplated this myself. Early on I saw an application where the Ember was used to embed RF antennas using a multi-material setup. Essentially, they created half the device with a pocket, dropped the RF antenna in, then finished it off with a different color. I’m sure it was more complicated that that, but you get the idea.



Your description of the “variable exposure” option is correct for slices created at However, when that option is selected in PrintStudio, it just applies a gamma correction to the gray values at the edges of antialiased slices. That correction is no longer needed when using Ember firmware versions 2.2 and later, because Ember’s output is now a linear function of the gray value. That means that for a fixed exposure time, the amount of energy received at a voxel is directly proportional to the gray value of the corresponding pixel. So for example, a 1 s exposure to a 50% gray pixel has about the same effect as a 0.5 s exposure to a white pixel.


@greener1 So slicing with should give the correct result for @T.j_Wallin?
I’ve not had a lot of experience with slicing on the webpage - since there is no option to move, rotate, or otherwise adjust the model before slicing. It means that I have to do all this beforehand in PS, so why go to the extra step? Since it’s no longer needed as it’s originally intended in PS - why not make it function like the webpage slicer?


Slicing with and “variable exposure” selected will darken the central regions of slices, so that may get @T.j_Wallin what he wants.
AFAIK, there are no plans for further development of PrintStudio. The latest and greatest slicing options for Ember are now available in Netfabb.


Thanks for the confirmation.

As for Netfabb - I have read through the BASIC manual find it very disheartening that this is the way of the future. It seems overly complex for 98% of the tasks I want to use if for. Also, the fact that we once had a free software that does 98% of the job, and now we are now being directed to a paid software for any future updates makes me not want to try it (PS - Pattern Mode is not free, not sure about variable exposure). I will try it out just to be fair, but my opinion is already skewed…

I hope Autodesk realizes that there are so many programmers out there creating free software for 3D printing that it will not be able to corner this market with a paid program. I will likely stick with the free PS for now and continue to evaluate all the other free options - including writing my own code as required.


The free version of Netfabb provides all of the Ember functionality that PrintStudio did, and goes it one better in that it provides anti-aliasing in the Z direction as well (PrintStudio only handled it in X & Y). As mentioned above, PrintStudio’s “variable exposure” option only provided correction for a problem that has now been resolved in Ember firmware, so there was no need to include that option in any version of Netfabb.

Print Studio no longer supported?

I will have to try the anti aliasing in the Z to see if it’s of any real benefit at this point.

As for the Variable exposure - I feel its important that I am clear: I want the functionality of the variable exposure, where the outline of the slice gets the most light and the inside areas receive a grey scale gradient. This has uses for very large cross sectional areas, or where there is a large variation in thicknesses across the slice.