09 Feb Detail on 110/54 Mesh
You Can Increase the Amount of Detail with Some Knowledge
I receive a great number of questions, some of which need a more thorough explanation. This is one of those questions. If you need a question answered, there is a link at the bottom of this article to ask your question and receive a well-thought-out answer, free of bias and with factual knowledge.by Bill Hood, Bill Hood Consulting
Question: When I’m exposing on a white 100-mesh/inch, I can’t get 100-percent detail especially on thinner text fonts, (ex. Times New Roman). But when I’m exposing on a yellow 355 mesh/inch, I can perfectly expose finer texts and even a period after a sentence. Is my exposure unit bad? It is a fluorescent LED 9-watt, with 6 tubes. I use Saati Textil PHU (Blue) emulsion. Is this a good choice of emulsion? I have asked this question in a rather large Facebook group and there is no consensus of agreement on why I am having trouble holding detail.
Answer: It is not your mesh count, it is the mesh color, your exposure lamp, and your expectations. A multisource exposure unit cannot hold small details and while we are at it, the resolution (details) are captured by the emulsion, not solely the mesh.
Let’s discuss fonts first. Serif fonts, like Times New Roman, do not work well on low mesh counts, because the thin lines can become hidden behind the thicker threads. They only work when the small details of the font are considerably larger than the thread size, and even then the fine serifs may not print well. It is better to go with a sans serif font that has thicker lines on both the horizontal and vertical standards.
In the illustration below, the mesh is 110 threads-per-inch with an 80-micron thread diameter and is to scale. The fonts are both 10 points and placed behind the mesh threads to simulate the ink that would be passed through the mesh openings. Ink will flow slightly under the thread, but the fine lines would have to fall in the mesh opening in order to flow.
Note that font Arial Regular 10-point prints well on all three rows. The Times Roman in Row 1 shows that the detail on the crossbar of the letter “e” is lost as is most of the top of the letter “o”. Now, look at the Times Roman in Row 2. Note that the top of the letter “e” prints well, but the top of the letter “a” and letter “o” disappears. And Row 3 shows much the same loss. As you can see, you can achieve more detail by going with a sans serif font that is bold enough to not become hidden behind the threads of the mesh.
When you choose 355 threads-per-inch mesh with a 40-micron thread more of the font is visible and prints much better.
Each fluorescent lamp emits light in a conical shape and the further away the stencil is from the fluorescent lamp the worse the undercutting becomes, as shown in the illustration below. The black area of the film positive is what is supposed to be printed and the light goes through the white areas to cross-harden the emulsion. As you can see, the lamps on either side of the center lamp are projecting light under the black image, and undercutting the image and thus it becomes smaller. And, on thin lines and small dots, they disappear from the stencil altogether.
With a single point light source, shown to the right, there is no undercutting (unless the exposure is unusually long) because the light is only coming from one central position and is far enough away from the image to maintain the width of the black area of the image on the film positive.
The Smallest Detail
June, there is a limit to the detail of an image that can be held in screenprinting, i.e., the smallest detail that can be printed. For decades, the smallest line or dot was thought to be equal to the distance across two threads and one mesh opening. However, this theory falls apart when the dot falls in the center of a mesh opening as can be seen as 1a in the illustration below. And there will always be dots that fall in the center of a mesh opening, presenting an irregular pattern to the image.
The illustration is to scale and the three smaller dots are equal to two threads and one mesh opening. As stated the 1a dot is centered on a mesh opening and the edges of the circle are clipped so that the circle appears as a square that is much smaller when printed. Of course, there are many other possible positions as can be seen in the illustration. Circle 2a is centered on a knuckle, i.e., where two threads overlap one another, and 3a shows what happens when the edges of the circle aligned with a thread.
However, my research has shown that a minimum dot size of two threads and two mesh openings as illustrated in 1b, 2b, and 3b below, is an improvement. When the circle 1b is centered on a mesh opening it still appears as a circle when printed, as well as in positions 2b and 3b.
When you print on 355 threads per inch mesh, the thread diameter is much smaller and closer to one another allowing it better support the smaller details in the stencil. And, the use of two threads and 2 mesh openings follows form with an improvement.
A 100 thread per inch mesh usually has a mesh opening of 200 microns and a thread size of 54 microns. A 355 thread per inch mesh usually has a mesh opening 68 microns and a thread diameter of 34 microns.
100/54 mesh: ((2 x 54) + (2 x 200)) = 508 microns minimum dot size, or 0.0508-cm (0.020-inch)
355/34 mesh: ((2 x 34) = (2 x 68)) = 204 microns minimum dot size, or 0.0204-cm (0.008-inch)
508 / 204 = 2.46 difference
So you see the 355 mesh can hold a dot that is 2.49 times smaller than what a 100 mesh could hold.
Mesh is either dyed or undyed. The dyed mesh comes in yellow, gold, amber, and red. If the mesh is undyed, then it is a semi-transparent polyester filament, that appears cloudy. There is no white mesh in screenprinting.
The undyed mesh cannot hold as much detail as a dyed mesh. As light passes through the undyed mesh threads, there is a certain amount of light scatter (halation) as well as some refraction that will take place. And, the light falls off according to the inverse square law as it travels beyond its point of penetration.
Depending on the intensity of the light source, the amount of halation can be profound. A 5k, 8k, or 10k lamp will halate much more than a 1k or fluorescent lamp. However, the lower the energy, the more likely the scatter is refraction.
While the exposure time for dyed mesh may be up to four times that of undyed mesh, by using the dyed mesh you will eliminate much of the unnecessary light scatter or halation gaining more resolution and a higher quality print with more detail.