HMI to PLC serial cable maximum distance by interface

How far can an HMI be from its PLC on a serial cable?

Plan on about 10 ft of RS-232 when the panel runs from the PLC's 5 V auxiliary supply, 50 ft of RS-232 when the panel has its own power, and up to 1200 m on RS-422 or RS-485 at 9600 bps.

Those three numbers cover nearly every HMI-to-PLC link in a plant. The table pairs each with the physics that sets it, so you know which limit you are up against before you pull cable.

Two rules finish the decision. Where the panel's manual prints a distance for your exact model and port, the shorter number wins. Everything else in the table is a guideline that holds for ordinary industrial cable and typical driver silicon.

What sets the distance limit, the cable or the interface?

The interface sets the limit; the cable only decides how fast you reach it. RS-232 runs out of voltage margin and capacitance budget within tens of feet, while the differential RS-422 and RS-485 pairs carry usable signals for hundreds of meters.

RS-232 is single-ended: one conductor carries the signal against a ground shared by both ends. The driver swings to at least ±5 V and the receiver switches at ±3 V, so about 2 V of noise margin is all the link ever has. Cable capacitance loads the driver and slows the edges, and drop along the conductors eats the margin. RS-422 and RS-485 are differential: a twisted pair carries the difference between two conductors, noise that lands on both conductors cancels in the receiver, and the pair holds its shape for hundreds of meters. Pick the cable inside the interface's rule: capacitance in pF per foot for RS-232, conductor gauge in AWG wherever the cable also carries power.

How long can an RS-232 cable be when the panel has its own power?

Plan on 50 ft when the panel has its own power, because the interface standard budgets 2500 pF of load capacitance per driver and the assumed 50 pF per foot of ordinary cable spends that budget at 50 ft.

The budget divides by the cable datasheet's capacitance figure. A low-capacitance cable at 24 pF per foot reaches about 104 ft, since 2500 ÷ 24 = 104; a heavy cable at 70 pF per foot closes the run to about 36 ft, since 2500 ÷ 70 = 36. Lower baud rates help, because slower edges tolerate more capacitance, and practical RS-232 runs seldom pass 100 ft because that 2 V of margin leaves little room for loss and ground bounce. For stretching an existing run by a few feet, see extending an RS-232 HMI cable by 10 feet safely for the safe method.

At or under 50 ft, keep RS-232 and spend the difference on a better cable. Past 100 ft, stop stretching; the middle-device section below is the bridge.

Why is RS-232 limited to about 10 ft when the PLC powers the panel?

Because the cable carries the panel's power as well as its data, and voltage drop, not the signal, runs out first. A low-voltage panel that draws 0.5 A and holds a ±5 percent tolerance over a 24 AWG pair runs out of voltage at about 10 ft.

Power and signal get checked separately, and on a 5 V feed the power check fails first. The arithmetic is one line:

max length in ft = allowable drop in V ÷ (panel draw in A × loop resistance in Ω per ft)

A supply at that voltage with a ±5 percent panel tolerance allows 0.25 V of drop. 24 AWG copper runs about 25.7 Ω per 1000 ft, roughly 0.05 Ω per foot once supply and return both count. At 0.5 A, a typical figure for a small graphic panel with touchscreen and backlight, the line gives 0.25 ÷ (0.5 × 0.05) = 10 ft.

Draw doubles, footage halves; copper doubles, footage doubles. A panel with a 24 V input escapes the trap entirely: the same 2.5 W draws about 0.10 A at 24 V, and a ±10 percent tolerance allows 2.4 V of drop, which carries the power leg past 300 ft of the same 24 AWG. Then the interface, not the power, sets the limit.

Worked example: can a 120 ft RS-232 run power and feed the panel?

No, because the 120 ft run needs about 6000 pF against the 2500 pF driver budget, and the power leg drops about 3 V; this move ends the RS-232 arrangement, so the panel gets its own 24 V supply and the link converts to RS-422.

The scenario: a packaging line's HMI moves from the main enclosure to the far wall, the conduit route measures 120 ft, the link runs at 9600 bps over RS-232, and the panel currently draws its supply from the PLC's auxiliary terminals.

Signal check first: at the assumed 50 pF per foot, the run loads the driver with 120 × 50 = 6000 pF, which is 2.4 times the 2500 pF budget. RS-232 fails this run whatever the power does.

Power check second: the panel needs 4.75 V minimum, so 0.25 V of drop is allowed. Take 0.5 A as the draw, an example input; read the real figure from the panel's manual, and at 0.25 A every footage above doubles. The loop is about 0.05 Ω per foot, so the drop is 120 × 0.05 × 0.5 = 3 V and the panel sees about 2 V. Upsize to 18 AWG, loop about 0.013 Ω per foot: 120 × 0.013 × 0.5 = 0.8 V of drop, leaving 4.2 V, still short of 4.75 V. The power leg fails at any reasonable gauge.

The numbers that close the job: give the panel its own 24 V supply, the common shop value. The same 2.5 W draws about 0.10 A, the drop is 120 × 0.05 × 0.10 = 0.6 V, and the panel sees 23.4 V, inside the ±10 percent band. Convert the link with an RS-232 to RS-422 converter at each end: 120 ft is about 37 m, a rounding error against the kilometer-plus guideline at this rate.

Result: nothing of the old arrangement survives except the panel's mounting holes. The run passes with hundreds of meters of headroom.

How far can RS-422 or RS-485 go, and what does baud rate cost?

Plan on 1200 m, about 4000 ft, at 9600 bps, and drop the guideline step for step with rate: 600 m at 19200 bps, 300 m at 38400 bps, 200 m at 57600 bps. The standards themselves print no hard maximum length.

That figure comes from the guideline's cable assumption, which allows 66 percent signal loss; below 90 kbps that works out to 4000 ft of the assumed cable. Where you land inside the guideline depends on your cable, connectors and junction count, and on keeping the pair twisted all the way.

For a run measured in thousands of feet, the walkthrough of connecting a remote HMI screen 1,500 feet from the PLC carries the same arithmetic up the plant. Past a few hundred meters, terminate the pair at the far end in a resistance matched to the cable's characteristic impedance; RS-422 termination resistor value for a long HMI cable gives the values and where they sit.

What do you add in the middle when RS-232 must stretch?

Add a line repeater, which gives each segment its own fresh driver and its own 2500 pF budget, or convert the run to RS-422 with a converter at each end, which turns the same copper into a differential link good for the full kilometer-scale guideline.

A repeater re-drives the signal mid-run, so two 50 ft segments make a 100 ft RS-232 run with margin restored at the halfway point. An isolating repeater also breaks the ground path, which matters where the two ends sit on different ground potential. A converter pair is the stronger fix: bidirectional RS-232 to RS-422 converters at both ends turn the same route into a differential link, and the wiring and distance planning for that conversion is its own job, covered in RS-232 to RS-422 cable distance for an HMI panel link.

Pick by length. At or under 50 ft, keep RS-232. From 50 to 100 ft, a repeater is the cheaper bridge. Beyond 100 ft, or across buildings, convert.

Recommendation: keep plain RS-232 for runs at or under 50 ft with the panel on its own supply, and about 10 ft when the PLC's auxiliary terminals feed the panel; convert the link to RS-422 or RS-485 for anything longer, and size it from the rate table above.

How do you check a serial link before the line depends on it?

Test in stages: power the panel on its final supply first, run a short known-good cable second, and only then install the full run, because a panel that browns out on a long power leg fails exactly like a too-long signal cable.

Lock out and tag the panel and PLC supplies before you land any conductors; that step comes before everything else. Then separate power faults from signal faults:

1. Power the panel on its final supply and measure the voltage at its terminals with the link running.
2. Bench the link on a short cable and confirm the project reads and writes.
3. Install the full run, set both ends to the same interface and rate, and read known PLC values.
4. Watch one production cycle: values track the PLC with no retries.

A panel that resets, dims or drops words while the values still track points at the supply; measure it at the panel terminals under load and compare the reading with the manual. When the symptom is a timeout with a dark com light, the diagnosis path is its own subject, covered in HMI to PLC serial timeout with a dead com light.

What should you confirm before ordering the cable?

Confirm three numbers before you order: the panel's own printed distance limit if its manual has one, the cable capacitance in pF/ft for any RS-232 run, and the baud rate, because the differential-pair guideline falls step for step as the rate rises (see the table above).

Then work the list:

• Panel manual, serial page: the printed distance limit for the port you will use, and the auxiliary supply's rated output in A if you planned to power the panel from the PLC.
• Cable datasheet: capacitance in pF per foot for RS-232 runs, conductor gauge in AWG wherever the cable carries power.
• Converter or repeater datasheet: the interfaces at each end and the rate it passes at your baud.
• Rate choice: 9600 bps buys the full length guideline; 38400 bps cuts it to 300 m.

Decided: about 10 ft for a powered RS-232 panel fed from the controller, 50 ft for separately powered RS-232, and the kilometer-scale differential guideline at the low rate. The converter pair is what buys the long run.

Questions