LCD viewing angle and HMI mounting height: test before you drill

Why does an LCD lose contrast when you view it from an angle?

An LCD modulates light through a liquid crystal layer whose contrast between black and white holds only within a cone of angles, and the datasheet's viewing angle marks where that contrast has already fallen to a level the vendor deems acceptable. Straight on, you get the panel's full contrast. As your eye moves off the normal, blacks lift toward gray and colors drift; at the printed angle the image is already at the edge of what the vendor calls readable.

That acceptable contrast level is a vendor choice, so the same printed number can sit on two panels whose off-angle appearance differs. A panel can sit comfortably inside its printed angle and still look poor in the flesh. So the datasheet number builds the shortlist, and the energized test makes the decision.

Is a given LCD stronger from above or from below?

Each panel has one strong vertical direction, and you cannot predict it from screen size or price: some industrial panels read best from below, most desktop monitors read best from slightly above, and the datasheet names the difference. Datasheets state it two ways: degrees up, down, left and right, and sometimes a clock position. A 6 o'clock panel reads best from below its center; a 12 o'clock panel reads best from above. Desktop monitors are usually built to be read from slightly above, which is why one hung high on a wall grays out when you look up at it.

Industrial panels go either way. Price tracks the width of the angle more reliably than the direction, and on small screens the technology matters: STN and TFT viewing angle on small industrial screens compares the two, with TFT generally publishing the wider cone at the same size. Read your own panel's up and down numbers off its datasheet, and mount so the operator's eyes sit on the strong side of the normal.

What does each bad-view symptom point to?

Match the symptom to its cause before you touch the mounting: angle problems follow eye position, glare follows the lights, and a haze present from every direction points at the backlight, a film, or the glass. Work down this list in order.

• Washes out only above or below the panel center. Text grays or color-inverts when your eyes move off center height and returns when you stand level with it. Vertical viewing-angle limit. Move the center toward eye height or tilt the face toward the eyes.
• Drops off when you step to one side. Readability falls left or right and returns at the center. Horizontal angle limit, common when a portrait install turns the panel's narrow vertical axis sideways. If the content is rotated in software instead of remounting the panel, the panel timing has to follow, and fixing a portrait LCD rotation with weston.ini and panel timing covers that job.
• Changes with the lights. The screen reads well until a luminaire or daylight hits it, and you can see the light source in the glass. Glare. Fix the light path or the panel aim; screen sunlight washout checks and fixes runs those checks for daylight and lamps.
• Dim or milky from every direction, straight on included. Backlight setting, shipping film left on, or grime. Clean the glass, peel the film, and set the documented brightness and contrast before you condemn the mounting position.
• One operator struggles, everyone else reads it fine. Eye-height mismatch. Measure that person's standing eye height and run the arithmetic below for it.

What height keeps a standing operator inside the viewing angle?

Put the panel center at the operator's eye height, about 1.6 m for an average standing adult, and keep it within about 0.15 m of that height at a 0.8 m working distance, which holds the sightline inside roughly 11 degrees. The formula is one line: sightline angle = arctan(height offset ÷ working distance). Measure the working distance where operators actually stand; 0.8 m is a common standing position at a machine face, and the example below uses it. Closer is harsher: at 0.5 m, the same 0.10 m offset gives arctan(0.10 ÷ 0.5) = 11.3 degrees.

Text size is nearly never the limit at this distance. The planning figure of 30 ft of viewing distance per inch of letter height (FHWA) needs about 2.2 mm capitals at 0.8 m, and the safer fast-reading figure of 20 ft per inch needs about 3.3 mm. Normal HMI text clears both. When digits that size cannot be read at arm's length, the fault is angle or glare, and the screen size and viewing distance guide covers the sizing half of the question.

Recommendation: Center the panel at the standing operator's eye height, about 1.6 m, whenever reach and head clearance allow it, and hold the center within 0.15 m of that height at a 0.8 m working distance. When the machine forces the center higher, tilt the face toward the eyes by about 7 degrees per 0.1 m of offset, then pass the energized view test before you drill.

How do you run the view test before you drill?

Power the panel, load the densest page it will ever show, and have a person hold or stand it at each taped candidate height while every regular operator reads the smallest text from the real working position. The test takes twenty minutes and it is the only part of this decision that cannot be done on paper.

1. Tape candidate center heights on the machine face, for example 1.45 m, 1.6 m and 1.75 m.
2. Power the panel and load its densest real page, alarms and smallest text included. A blank or menu page hides the fault.
3. Stand the panel at each candidate, held by a helper or clamped to a strut, at final mounting depth if it has bezel keys. A wrong mounting gap can kill keypad input, and the bezel keypad mounting gap check shows that test.
4. Have every regular operator read the smallest text aloud from the real working position, standing the way they work. Note which heights pass for both the shortest and the tallest person.
5. Toggle the room lighting and block suspected reflections once per height. If readability follows the lights, treat it as glare and leave the height alone.
6. Run the lean test: lean across the machine the way operators do and confirm the candidate sits outside the path your head sweeps. Head clearance outranks the last few degrees of viewing angle.
7. Drill at the height that passed every check, then repeat step 4 on the mounted panel before you call it done.

How does the mounting decision work out in an example?

Example: a panel must clear the lean-over path, so its center goes to 1.75 m; operators' eyes sit at 1.6 m and 1.5 m, the working distance measures 0.8 m, and the tilt that fixes the geometry is about 14 degrees. Inputs first: the lean path forces the center to 1.75 m. Operator A has eyes at 1.60 m, operator B at 1.50 m. The working distance measures 0.80 m. The panel datasheet prints 65 degrees up and 55 degrees down, which are illustrative inputs; use your own sheet.

• Operator A sightline: 1.75 − 1.60 = 0.15 m above the eyes. 0.15 ÷ 0.80 = 0.1875, and arctan(0.1875) ≈ 10.6 degrees up.
• Operator B sightline: 1.75 − 1.50 = 0.25 m. 0.25 ÷ 0.80 = 0.3125, and arctan(0.3125) ≈ 17.5 degrees up.
• Against the sheet: 17.5 of the printed 65 degrees up is 27 percent of the limit, so both operators sit inside the printed angle on paper.
• View test at 1.75 m, untilted: operator B reports graying digits. The margin on paper ran into the contrast rolloff that the printed angle already contains.
• Fix with tilt: aim the panel normal between the two sightlines, (10.6 + 17.5) ÷ 2 ≈ 14 degrees of down-tilt. After the tilt, each sightline sits within 3.5 degrees of the normal.
• Text check: the smallest digits measure 5 mm, against the 2.2 mm the planning figure needs at 0.8 m, so size passes and angle was the whole fault.

Decision: drill at 1.75 m with a 14 degree down-tilt, but only after both operators read the alarm page aloud at the tilted position and the lean test clears.

When should you tilt the panel instead of moving it?

Tilt when the machine forces the center off eye height: give the panel a down-tilt equal to the sightline angle, about 7 degrees for every 0.1 m the center sits above the operator's eyes at a 0.8 m working distance. The table above pairs offset with tilt. A tilt equal to the sightline angle aims the normal straight at the eyes, which is the strongest position the panel has. Panels forced below eye level take the mirror fix: tilt the face up toward standing eyes. For a seated operator, whose eyes sit near 1.2 m, center a console panel there and the arithmetic is unchanged.

Two checks govern how far you lean on tilt. A tilted panel reflects a different slice of the ceiling, so recheck both eye positions for luminaires after any change. And tilt moves the bottom edge out over the machine face, so re-run the lean and reach checks at the tilted attitude before you drill.

Does turning up the brightness fix a washed-out view?

No: brightness raises the light behind the glass while the washed-out black level stays washed out, so the digits stay gray and the reflections get stronger, and the repair is position, tilt, or glare control. Set the controls correctly first, because a mis-set panel imitates every other fault; a brightness, contrast and time adjustment walk-through for a touch HMI shows the kind of menu path that setting lives in. Once set, more brightness adds light to the washed-out black as well as the white, and it strengthens whatever the glass reflects. The off-angle contrast rolloff stays. The repairs that hold are height, tilt, glare control and, at purchase time, a panel with wider printed angles.

What do you recheck after the panel is mounted?

Repeat the read-aloud test on the mounted panel with the densest real pages, including alarm pages and the smallest text, from every operator's eye position, and recheck reflections with the room lights in their working state. Close it out against this list.

• Both extreme operators read the smallest text and the alarm page aloud, from working stances, in working light.
• Touch and bezel keys respond at the final mounting depth.
• No luminaire, window or bright machine surface sits in the glass from either eye position.
• The lean test still clears, and the controls sit inside a comfortable reach.
• The chosen height and tilt go in the machine file, so the next relocation starts from a tested number instead of a guess.

If the machine keeps its HMI and you also want a status board the crew reads from across the aisle, that job belongs to a separate screen, and the same sightline arithmetic applies at the longer distance; shop-floor signage displays cover that use.

Questions