You want frame one to flash the light behind GPIO line A, frame two the light behind line B, and the pattern to repeat from there. Half of that is standard camera behavior: set Line Mode to Output with Line Source to Exposure Active and one chosen GPIO line pulses for every exposure. The other half is not. The standard feature set has no sequencer control for output lines, so the camera cannot walk the strobe across two GPIO lines on its own.
This guide puts the alternation where it works: a light controller or a PLC toggling two strobe channels per trigger, or a host swapping the strobe line after each image. You get the wiring, the ordered steps, a worked duty-cycle example at 20 triggers per second, and the oscilloscope test that proves the pattern holds.
What the camera can do on its own, and where it stops
Set Line Mode to Output and Line Source to Exposure Active, and the camera drives that GPIO line for every exposure. That is the whole of the standard strobe output. How the pulse lines up with the exposure window and the image timestamp is its own timing subject, worked through in camera strobe output timing: exposure pulse and image timestamp; for one area-scan camera's register-level walk-through, see this GPIO strobe trigger from ExposureActive example. You can set the same source on two lines and fire two lights together every frame. What you cannot do is fire them on alternate frames.
The GenICam feature naming convention behind those settings defines the camera sequencer around stepping acquisition settings such as exposure and gain. That is how a camera alternates two exposure times per trigger with no serial traffic, the mechanism camera sequencer control: alternating exposure and gain per frame covers. The convention defines no Sequencer Line Selector, Mode or Source that would bind an output line to a step. Search a conformant feature tree and the knob you need is not there, so the alternation has to come from outside the sensor.
The decision: where the alternation lives
Commit to one row before you buy anything.
| Your case | Where the alternation lives | What you need |
|---|---|---|
| Both lights should fire in every frame | Nowhere: one pulse drives both | One Exposure Active line, both light inputs wired to it |
| Two illumination geometries on alternate frames, PLC or controller in the cell | Light controller or PLC toggles two strobe channels per trigger | A 2-channel strobe controller fed by the camera pulse |
| Bench work, low trigger rate, host sees every image | Host swaps the strobe between the two GPIO lines after each image | No extra hardware, one frame of latency |
| The camera offers per-step line control of its own | In the camera sequencer | Read the feature list once; it is a vendor extension and rare |
In every hardware row the camera keeps one job: produce a clean Exposure Active pulse per frame. A light controller with two strobe channels then decides, trigger by trigger, which channel passes that pulse to a light. When a PLC already sits between camera and lights, it can do the toggling itself, and buffer relay between camera and strobe: when you need a PLC covers when that intermediate box earns its place.
Wiring: one camera pulse, two strobe channels
- On the camera, set Line Mode to Output and Line Source to Exposure Active on the line you have wired out. That pulse is the only per-frame signal the camera produces.
- Run the line to the controller's trigger input. Camera outputs are TTL-level or opto-isolated; the levels, pull-ups and series-resistor question are handled in wire a GPIO output to a camera trigger input: TTL levels and pull-ups.
- Break the camera connector out on an M8 8-pin power and I/O cable with 8 open cores, 7 m, drag-chain rated when your GPIO lines live inside a multi-pin jack.
- Type
- Power and I/O cables
- Length
- 6-8 m
- Wire light A to channel 1 and light B to channel 2, one channel per light. Both lights behind one channel is the fire-together case from the table, not alternation.
- Set both channels to strobe on the external trigger, then inhibit channel 2. Only the armed channel should pass a pulse.
- Drive the toggle from either side: a PLC output that flips on every trigger acknowledge, or one channel-select command per trigger from the host over serial. The LS-CTL-0487, a 2-channel 24 V digital LED light controller at 240 W, carries an RS-232 port for that.
- Power up and scope both channel outputs before you trust a single image.
Spare cores and longer runs come from the power and I/O cables range. Keep the strobe pulse line away from motor leads so the trigger edge stays clean.
Worked example: bright-field and dark-field rings at 20 triggers per second
Inputs, all of them yours to choose: the PLC triggers the camera 20 times per second; exposure is 1.0 ms, set to freeze the belt; the strobe pulse width equals the exposure at 1.0 ms; each ring draws 2 A while lit; light A is the bright-field ring on odd frames, light B the dark-field ring on even frames.
The rule in words: alternation halves each light's pulse rate, so each light's pulses sit two frame periods apart, and duty is pulse width over that spacing.
The arithmetic: T = 1 / 20 = 0.050 s = 50 ms. Spacing per light = 2 × 50 ms = 100 ms. Duty = 1.0 ms / 100 ms = 1%. Fired every frame instead, each light would run 1.0 ms / 50 ms = 2%. Average current per light = 2 A × 1% = 0.020 A = 20 mA, while each channel must still switch the full 2 A during its 1.0 ms pulse. The idle gap between exposures is 50 ms - 1.0 ms = 49 ms, the window every channel-select command or software swap has to land in.
The decision these numbers force: alternate in the controller. The 2 A pulse sizes the channel; the 20 mA average sizes the supply. At 1% duty heat is not the constraint, switching is. Overdrive headroom, if you want it, comes from the duty ceiling in the light's rating, which strobe machine vision light: overdrive, duty cycle and timing works through in detail.
Choosing the controller
Filter the light controller finder on what this job needs: at least two channels, digital control so a host or PLC can command one channel per trigger, and the strobe feature. The finder's feature filter carries both Strobe and a sequencing option. A fourth channel buys a third geometry later with no new box.
The LS-CTL-0487 covers the two-light case: 2 channels, 24 V, 240 W, RS-232. The LS-CTL-0070 is the grow-into choice, a digital control box with 4 channels, 200 W, RS232 and a strobe feature.
| Specification | LS-CTL-0487 | LS-CTL-0070 |
|---|---|---|
| Price | $385.00 | $549.00 |
All 2 share: Output voltage 24 V. Values as printed in each product's data sheet; open a part for its drawing and full specifications.
Where several lights share one enclosure, the LS-CTL-0322 brings 4 channels, 16-24 V and 8 A total output.
- Output
- Constant voltage 24 V
- Channels
- 4 channels
- Control
- Digital
The software swap, step by step
Choose this at the bench, or in production only while the idle gap comfortably beats your host's reaction time. Measure the callback-to-write time once. A 2 ms write inside the 49 ms gap of the example leaves 24x margin; the same 2 ms inside the 4 ms gap of a 200-trigger-per-second line (5 ms period - 1.0 ms exposure) is gambling on jitter.
- Confirm in the feature tree that both GPIO lines accept Output mode and both list Exposure Active as a source. Some cameras tie certain sources to certain lines.
- Arm line A only, with line B's source cleared.
- In the image-event callback, after the image arrives, move the source: clear Exposure Active on the line that just fired and set it on the line that fires next.
- Accept one frame of latency. A line setting written now governs the next trigger, the same one-frame behavior as set exposure every frame: latency and the one-frame delay.
- Count parity from the image counter itself, so a resync after a dropped trigger is automatic.
Two related behaviors shape this route. Cameras that alternate exposure per trigger without serial writes do it in the sequencer, which two exposure times one camera: alternate per trigger without serial covers, and that is exactly the block with no line control, which is why the strobe cannot ride along. And if the camera is out of physical lines, use camera trigger lines as digital outputs when space is short covers pressing trigger pins into output duty before you buy hardware.
Checks before you call it done
- Scope both controller outputs across at least four triggers: line A on frames one and three, line B on two and four. Oscilloscope check: lighting pulse timing and camera exposure overlap shows where to probe.
- Each pulse sits inside its exposure window. A light that is visibly on while the image is black is a timing miss, not a wiring fault, and strobe light on but camera image is black: fix the timing sorts the two.
- Expect one dark or partly lit frame after power-up while settings settle. First camera frame dark: exposure settling or light timing tells you which you are seeing.
- Drop a trigger on purpose and confirm parity recovers on the next image.
- Run at production rate for a minute. A swap that barely fits the idle gap fails late, not on the first scope shot.
When alternation is the wrong answer
- You only want more light: put both lights behind the one Exposure Active line and fire them together. Duty per light doubles to 2% at the example rate, and there is no toggle to fail.
- You need both geometries in one image: alternation delivers two views in two frames. To blend them into a single exposure, sequence both channels inside that exposure from the controller instead.
- The rate leaves no idle gap for a host write: put the toggle in hardware from day one.
- Your camera does expose per-step line control under a vendor name: read its feature list once, because that extension removes every wire on this page.
Wire it once, scope it once, then lock the camera's line settings and let the controller carry the alternation.
The configurator matches a camera, lens, light, light controller and cables to each other, step by step, and lists only parts that fit what you already picked.
Free expert support comes with every order, so if you want the channel plan looked at before you cut cable, use the contact page.
Kérdések
Why can't the camera sequencer move the strobe to the other GPIO line?
Because the sequencer only steps acquisition settings per trigger, as described above; a Sequencer Line Selector, Mode or Source that would assign an output line to each step is not part of the standard, so a conformant camera offers no such knob. A few makers add per-step output control as an extension; if your camera's feature list shows none, the alternation belongs outside the camera.
Is swapping Line Source in software fast enough?
It is, while the host reacts inside the idle gap between exposures. At 20 triggers per second with 1.0 ms exposures the gap is 49 ms, and a swap issued from the image-event callback lands with margin. The write governs the next trigger, so plan for one frame of latency when matching images to lights. Measure your callback-to-write time once: when it stops being a small fraction of the gap, move the job into the controller.
Can I drive both lights from one GPIO line instead?
Yes, when they must fire in the same frame: wire both driver inputs to the one Exposure Active pulse and both flash together. That is not alternation. Each light then fires every frame, doubling per-light duty from 1% to 2% at the example rate. Alternation pays only when the two lights must not fire together, such as bright-field and dark-field views on successive frames, or two stations sharing one camera.
How does alternation change the strobe duty cycle?
Each light fires every second trigger, so its pulse spacing doubles and its duty halves. At 20 triggers per second with 1.0 ms strobes, each light runs 1.0 ms per 100 ms, a 1% duty, against 2% if both fired every frame. Average current halves with it, 20 mA against 40 mA in the example, while each channel still has to switch the full 2 A pulse.
How do I confirm the two strobes really alternate?
Scope both controller outputs, or both GPIO lines, and trigger at least four times: line A should pulse on frames one and three, line B on two and four. Then drop one trigger on purpose and confirm the parity recovers on the next image. If a light is visibly on while its image is black, the pulse is missing the exposure window, so chase the timing overlap first and the wiring second.

