Two Free UV Printer Tests: Z Height Range and Printing on Spheres

Two questions every UV printer owner runs into. How much Z height variation can your machine actually handle before the print goes bad, and how large a design can you fit on a sphere without losing quality? These free files answer both, on your machine, in an afternoon.

What you need before you start

These are two separate tests and they do not need the same equipment. Test 2 you can run today with a ball out of the closet. Test 1 needs a physical object made first, and that part is 3D printing, not UV printing.

Test 1: how much Z height variation can you work with?

Your printer focuses at one height. Call that zero. Anything lower than that is printing at a distance, and ink droplets that travel further land less accurately and spread more on the way down. The question is how much lower you can go before it shows.

The block is 50 x 100mm with ten steps in it, each one half a millimeter lower than the last. The graphic prints the exact same artwork on every single step. Same lines, same registration hairline, same color patches, same microtext.

That matters, because it means any difference you see between one step and the next is not the artwork. It is purely the height.

Step one: 3D print the block

This is the part that happens on a 3D printer. Download stair-test-50x100.stl, slice it, print it.

• Use 0.25mm or 0.1mm layer height. Not the usual 0.2mm. This is the one setting that matters. The steps drop by 0.5mm, and 0.5 does not divide evenly by 0.2, so a 0.2mm layer height rounds your steps off and the block no longer measures what it says on the label.
• Print it in something light and matte. PLA is fine. A dark or glossy block makes the fine printed lines hard to read.
• Flat on the bed, no supports needed. The steps face up and there is nothing overhanging.
• Infill and wall count do not matter. Nothing here is structural.

Step two: UV print the graphic onto it

Now you are back on the UV printer. The file is uv-stair-test-graphic-50x100.png.

• Focus on the tallest step, the 5mm one at the end of the block. Every number on the print is relative to wherever you focused, so if you focus somewhere else the labels lie.
• Drop the graphic on at 100%. It is built at exactly 50 x 100mm at 1270 dpi. Do not let the software scale to fit.
• CMYK only. No white layer, no varnish, single pass. Anything extra muddies what you are trying to see.

Reading it

Start with the thin vertical hairline. It sits at the exact same spot on every tile, so sighting straight down the block from the end shows you lateral drift. If the line marches sideways as the steps get lower, that is your droplets landing off target from the extra flight distance, and it is the single most useful thing on the card.

Then the two line-pair wedges. Find the finest pair that still reads as two separate lines. Do that on the top step first, because that is your machine at its best, and every lower step gets measured against that instead of against a number I made up.

Then the solid patches and the tint ramp. Ink spread shows up here as patches that creep past their edges and a ramp where the light steps stop separating.

Why this matters on real jobs

I ran into this before I ever built the test block. I printed a leather key fob where the strap wraps over the key ring and buckles back on itself. The buckled part is the high point, so that is what the machine focused on. The face I actually wanted to print on sat a little over three millimeters lower, and the print came out poor.

That is the whole lesson. Before you commit a part, work out what the machine is going to focus on and how far your print surface sits below it. If the answer is more than your window, shim it, re-fixture it, or print it in a jig that brings the surface up.

Test 2: how large a design fits on a sphere?

Same problem, different shape, and no 3D printing needed for this one. On a ball, the surface starts falling away from the focus plane the moment you leave the top. So there is a circle around the high point where the print is clean, and outside it the quality drops off.

This target finds that circle. It is 35mm across, printed at 1270 dpi so it drops on at true size. The center dot is 1mm across, and every ring after it steps the diameter up by 2mm, labeled at the top and bottom in even numbers all the way out to 34mm.

The ring colors cycle every five rings, so the same color comes back around every 10mm of diameter. That gives you something to count by instead of squinting at numbers. And the numbers themselves are printed at 0.67mm cap height, which means they double as a legibility check. The ring where you can no longer read the number is roughly the ring where fine detail gave up.

Printing it

• Hold the ball still. A jig, a socket, a printed cradle, anything that stops it rolling. A ball that moves mid-print tells you nothing.
• Get the target centered on the high point. This is the hard part and it is worth taking your time over. If the target is off center, the rings on one side fail earlier than the rings on the other and the test is spoiled.
• Focus on the top of the ball. Same rule as the Z height test.
• Read off the largest ring that still prints as a clean, continuous circle. The number on that ring is your working diameter for that size of ball.

You can predict it too

Once you know your Z window from Test 1, the geometry gives you a starting estimate. For a ball of radius R and a usable drop of d, the printable width is:

diameter = 2 × √(2Rd − d²)

A golf ball is about 21.35mm in radius. With a 2mm window that works out to roughly 18mm of printable diameter. A ping pong ball at 20mm radius gives you about 17mm.

Treat that as the ceiling, not the answer. Geometry does not know about your alignment accuracy, your ink, or how well you centered the thing. The printed target is what tells you the truth, and in my experience the real number comes in under the calculated one.

A few questions I get

Do I need a 3D printer for this?

For Test 1, yes. The step block is an STL and it has to exist as a physical object before you can print on it. Test 2 needs no 3D printing at all, just a ball and your UV printer.

I do not have a 3D printer. Can I still run Test 1?

Email me and I will print a block and mail it to you for a small fee covering material and postage. A local library, makerspace or a friend with a printer works too. The STL is free either way.

Will these work on my printer?

Yes. Neither test is brand specific. Every UV printer has a focus plane and the same physics applies to all of them. The files are plain PNGs and a plain STL.

Why does the layer height matter so much on the block?

Because the whole test depends on each step being exactly half a millimeter lower than the last. At 0.2mm layers the slicer cannot land on 0.5mm boundaries, so your steps come out at whatever it rounds to and every number you read off the print is wrong by an amount you cannot see.

Why identical artwork on every step instead of a gradient of sizes?

Because a test where two things change at once tells you nothing about either. Same artwork, one variable, clean answer.

How often should I run these?

Once per machine is enough, unless you change something that moves the head height. These are not maintenance tests, they are numbers you write on the wall and design around.

Can I share these files?

Please do. Print them, hand them to your team, use them on client work. Just leave the Samcraft mark on them so the next person knows where to get their own.

Where to go next

Grab the files