The solder bridges
Two jumpers on the edge of the board decide which of four protocols the sensor speaks, and a third, marked RANGE, changes how far it reaches. All three are open when it arrives, which is why an untouched TK50 works the way the HC-SR04 its listing is named after does.
The table, made operable
Pick a mode to see the pads, or work the pads to see what mode you made.
That table is printed on the back of the board, so you never need this page with the part in your hand. What the page adds is what is underneath it.
What the pads are actually wired to
Each pad pair has ground on one side. The other side goes through a 10 kΩ resistor to a pin of the RCWL-9616: pin 8 for J1, pin 10 for J2. Soldering the bridge connects that resistor to ground; leaving it open leaves the pin to whatever the chip does with it internally. The chip maker's own table says the same thing in its own words: 10 kΩ on pin 8 is I2C, on pin 10 UART, on both 1-Wire, on neither GPIO.
There is no copper between the pads of either jumper in the board's own manufacturing files, and the solder paste goes down as two separate spots, so nothing bridges them at the factory. That is how we know, from the board rather than from a claim, that a TK50 arrives in GPIO mode.
So "configuring" this sensor is grounding one pin, or the other, or both, or neither. Four states, four protocols.
Why it has to be solder
The two pins the chip reads the mode from are not spare pins. They are the two that connect to one of the transducers — the same copper that carries 40 kHz a few milliseconds later.
A pin cannot be a setting and a signal at the same time, so it has to be a setting first and a signal afterwards — the chip must read the straps before it starts measuring. That is a conclusion from the net list rather than a line in a datasheet, but it gives the one operational rule on this page, and the rule holds whether or not the reasoning does.
Power the board off before you solder, and on again afterwards. A bridge added to a running board changes nothing until the supply is cycled, and the resulting "I soldered it and nothing happened" is the most common report about this part.
Doing it
It is a small joint and an easy one. Tin the iron, touch both pads and the gap between them, add a little solder, lift. The gap is a chevron a fraction of a millimetre across and it wants to close — the difficulty is not bridging it, it is bridging only the pair you meant to.
Check the result with a magnifier or a continuity beep before wiring anything. Two seconds there saves the twenty minutes of assuming the sensor is dead.
The third jumper: RANGE
On the right-hand side of the back, beside the power LED, is a third jumper
printed RANGE. It has nothing to do with the protocol. It sits across a
2.2 kΩ resistor in series with one transducer, and it ships open, so the
resistor is in circuit.
The note on the board's schematic comes from the chip maker: with the resistor, a typical maximum of 3 to 5 m; bridged out, 5 to 6 m. Those figures are for the chip maker's own transducers, not these, and nobody has measured this board either way, so treat them as a direction rather than a number. The honest range was written with the jumper open.
Leave it open unless a build needs the reach and you can test it there. More reach is also more sensitivity to faint echoes from things you did not aim at.
What each one buys
In short: I2C puts the sensor on a bus it shares, UART puts it on the end of a cable, 1-Wire saves you a pin, and GPIO — doing nothing — is faster than all three. Which mode for which job is the version of that sentence with numbers in it.
When it does not work
J2 is the upper one and J1 the lower, which is the opposite of the order most people assume when working quickly. Both are labelled in white on the left edge beside the mode table. Bridging the wrong one does not break anything — it gives you UART where you wanted I2C, or the reverse. RANGE is the third jumper, on the right, and has nothing to do with the mode.
Power the board down and up again. Then look at the joint under a light: a bridge has to actually span the chevron gap between the two pads, and a blob that only wets one pad looks convincing and connects nothing. A multimeter on continuity across the pair settles it in seconds.
With solder wick and patience, yes. It is fiddly on pads this size and there is a real chance of lifting one, so treat it as a decision per board rather than per project. There are five boards in the box precisely so you do not have to reuse one.
No. Both pads open is GPIO mode, which is a complete working sensor and what the sketch in the first reading uses. The bridges exist for the cases where you want a bus or you are short of a pin, not as a setup step.
Because the chip has no spare pins to be told with. The two pins it reads the mode from are the same two that connect to one of the transducers, so they can only carry a setting before the sensor starts working, not during. A solder pad is the cheapest way to hold a setting that never changes.
Only if you need the reach and can test the result. It shorts out a 2.2 kΩ resistor in series with one transducer; the chip maker's note says that takes the typical maximum from 3 to 5 m to 5 to 6 m with its own transducers. Nobody has measured it on this board, and more reach also means more faint echoes from things you did not aim at.
What ECHO and TRIG are called in each mode, and the one that is a trap.
One pin, three names →Edit this page — content/books/ultrasonic-sensor/the-two-solder-bridges.mdx
Questions about this product
See what other owners have asked, and read their solutions.
Ultrasonic Distance Sensor
Loading discussions…
Discuss this article
Ask about this page. The answer stays here, on the page it belongs to, for whoever hits the same wall next.