A hand-held load cell indicator is a battery-powered instrument that goes to the measurement instead of making the measurement come to a bench. It powers a strain gauge sensor, reads the very small voltage that comes back, and displays what that voltage means in pounds, inch-pounds or psi — in your hand, with no mains outlet and no data acquisition system involved.
The thing it reads is a load cell: a machined body with foil strain gauges bonded to it, which deforms under load so that the gauges change resistance and a bridge circuit turns that change into a few thousandths of a volt. Converting those millivolts into a number somebody can act on is the indicator's entire job.
And it is harder than it sounds. A load cell's whole output at full capacity is smaller than the voltage across a nerve. Reading it accurately on a bench, plugged into the mains, is an ordinary engineering problem. Reading it accurately from a battery, in your hand, on a factory floor, for sixty-five hours, is not.
The specification sheet explains how it is done — but only if you read it as a system rather than a list. Three of the numbers on it look like ordinary marketing until you divide them into each other, and then they turn out to be one design decision described three times.
Start with the sampling rate. The converter runs at 15,360 samples per second on 60 Hz mains and 12,800 on 50 Hz. Those look like two arbitrary figures. Divide each by its mains frequency and both give exactly 256. The instrument is not simply fast; it is locked to the power line, taking precisely 256 conversions in every mains cycle.
Now look at the display rate. The screen updates 3.75 times a second on 60 Hz and 3.125 times on 50 Hz — which reads, at first glance, like a slow display bolted to a fast converter. Divide the sample rate by the update rate in either case and the answer is the same: 4,096. Every number that appears on that screen is the average of four thousand and ninety-six separate conversions, taken over a whole number of mains cycles.
That is not a compromise. It is the entire trick. Averaging a whole number of mains cycles gives equal weight to every part of the waveform, so hum from the lighting, the motors and the drives around you sums to almost nothing rather than wandering through your reading. Averaging four thousand conversions also buries random noise. Do both at once and you can resolve to about one and a half microvolts — roughly a millionth of the voltage in an AA battery — from something running on a rechargeable cell in a plastic case.
So ultra-fast 15,360 samples per second is not a claim about how quickly the display responds. It is a claim about how thoroughly the noise has been removed, and that is the more useful claim of the two.
The same logic shows up in the data logger. The fastest logging interval is 16.666 milliseconds, which is not a round number in any obvious sense until you notice it is exactly one cycle of 60 Hz mains. Same design, applied to stored data instead of displayed data.
Three things have to happen inside the case. The instrument has to power the sensor's bridge, it has to measure the millivolts that come back, and it has to know what those millivolts mean. The first two are electronics and every indicator does them. The third is where hand-held instruments are usually let down, and it is where this one is built differently.
The excitation is 3 volts, and that is deliberate. Nearly every sensor in our catalogue is specified at 10 VDC, so three volts looks low. It buys two things. The obvious one is endurance — the excitation current is what a bridge instrument mostly spends its battery on, and dropping it is how a 2,600 mAh cell runs an instrument for sixty-five hours. The less obvious one matters more for accuracy: less excitation means less power dissipated inside the load cell, which means less self-heating. A bridge that warms itself drifts, and drift is the thing that ruins a long unattended log. Low excitation is not a concession to the battery at the expense of the measurement. It helps both.
The cost is that your signal is three times smaller, and that is why sensitivity matters here. The instrument accepts sensors from 1 to 5 mV/V into a maximum span of 15 mV. A 5 mV/V sensor at 3 volts produces exactly 15 mV at capacity and fills the input completely. A 2 mV/V load cell — the most common figure in our range — produces 6 mV and uses a little over a third of it. Since the converter divides its 15 mV span into 65,536 counts, that still leaves something on the order of twenty-six thousand counts across your sensor's working range, which is far more resolution than any load cell's accuracy can justify. Worth knowing rather than worrying about: a higher-sensitivity sensor uses more of the input, and neither choice is short of resolution.
Then there is the part that removes the commonest error in portable measurement. Under IEEE 1451.4, a sensor can carry a small memory describing itself — what it is, what excitation it wants, what its valid measurement range is, and calibration data specific to that individual unit. Published descriptions of the standard put its purpose plainly: a system reads that information automatically in order to identify, characterise, interface with and correctly use the signal from the analogue sensor, which removes manual parameter entry and the setup time that goes with it.
Think about what that actually prevents. In a facility with several sensors, the measurement that goes wrong is rarely the one with a faulty transducer or a faulty instrument. It is the one where the right sensor was connected to the right instrument with the wrong scale factor still loaded from last week. That error produces confident, stable, plausible, completely incorrect numbers, and it leaves no symptom at all. Nothing looks broken. TEDS removes the step where it can happen: the instrument asks the sensor who it is and scales itself accordingly, every time, without anybody typing anything.
Everything else follows from being genuinely portable. Ten ounces and a case the size of a chunky TV remote. A six-digit display counting from minus ninety-nine thousand to plus ninety-nine thousand, so you rarely give up a decimal place. Peak and valley capture, auto-tare, selectable units with automatic conversion between load, torque and pressure. Sixteen thousand time-and-date-stamped readings in internal memory and USB 2.0 to get them out. Two solid-state relays with LED and audible indication, isolated to several thousand volts from the signal common, so a limit can act on something rather than just being noticed.
Choosing instrumentation is mostly a question of where the reading has to happen and who has to trust it afterwards. Our applications engineers are happy to talk through a whole measuring chain rather than a single part of it.
Decide first whether the measurement travels. If your sensors live permanently in a machine and feed a control system, a panel instrument or a module is the right shape and the SSI is not. If somebody has to take a reading to where the work is — a machine on the far side of the plant, a customer's site, a goods-in bench, a rig with no mains socket near it — then a hand-held instrument with two days of battery is not a lesser version of a bench meter. It is a different tool.
Ask how many sensors are in circulation. This is the question that decides whether TEDS earns its keep. One instrument permanently married to one sensor barely needs it. An instrument that meets a different sensor every few days needs it badly, and the more sensors there are, the more certain it becomes that somebody will eventually measure with yesterday's scale factor. If you are ordering sensors alongside the instrument, specify the Cal-Teds plug and play option on them so the pairing works as intended from the first day.
Check the sensitivity of what you already own. The input accepts 1 to 5 mV/V. Most of our load cells and torque sensors sit comfortably inside that, but it is a five-minute check against your existing certificates and it is better done before the order than after.
Work out whether you need the trace or the number. Peak and valley capture, an auto-tare and a pair of setpoint relays cover a great deal of practical work with nothing recorded at all. If you need evidence — a time-stamped record of what a machine did overnight, a file to attach to a report, a series of readings to compare against last quarter's — then the internal memory and the logging software are the reason to choose this instrument rather than a simpler one.
Then decide what the relays should do. They are solid-state, isolated by thousands of volts from the signal, and rated for modest currents at high voltage, which suits driving a lamp, a beacon, a small solenoid or a logic input on somebody else's controller. They are a signalling output rather than a power output, so anything with a real current draw wants a contactor between the instrument and the load.
Finally, consider the software separately from the instrument. There is a package for reading and editing what is written in a sensor's TEDS memory, and a package for data logging to a PC. They solve different problems — one manages the sensors, the other manages the results — and a facility that owns many sensors often wants the first more urgently than it expects to.
Tell us what you are measuring, what sensors you already have, whether the readings need to be recorded and where the work happens, and we will tell you whether this is the right instrument or point you at a better fit. We keep this instrument on the shelf, and universities and research groups should ask about the academic discount when they enquire.
SSI Smart Sensor Indicator Applications.
The Transducer Techniques SSI is a plug and play smart TEDS IEEE 1451.4 compliant portable hand-held indicator that self-calibrates to the sensor connected to it, and is used wherever a reading has to be taken away from a bench.
- Field Service and Maintenance: Engineers carry one instrument to equipment across a site and take traceable readings from whatever sensor is fitted, without configuring anything by hand.
- Incoming Inspection: Goods-in departments verify that a delivered sensor performs to its certificate before it is accepted into stock or fitted to a machine.
- Production Line Verification: Quality staff spot-check force, load and torque at the machine, using the setpoint relays to signal a result rather than reading a number and writing it down.
- Laboratory and Test Benches: Researchers use the peak and valley capture and the fast averaged conversion to record test events accurately without building a data acquisition system around them.
- Calibration Support: Laboratories read and edit sensor TEDS memories so that each transducer carries its own current calibration data wherever it goes next.
- Long Duration Monitoring: Sixteen thousand time and date stamped readings and up to sixty-five hours of battery life allow unattended recording where mains power is impractical.
- Multi-Sensor Facilities: Organisations running many transducers use automatic TEDS scaling to remove the setup step where the wrong factor gets applied to the right sensor.
- Peak Force and Break Testing: Peak capture records the maximum reached during a destructive or proof test, where the value exists only for a moment.
- Limit and Alarm Checking: Dual solid state relays with LED and audible indication drive a lamp, beacon or logic input when a reading passes a set threshold.
- Portable Data Collection: USB logging to a PC turns a hand-held reading into a file that can be attached to a report, compared over time or archived as evidence.
The SSI works with our full range of load cells, torque sensors and pressure transducers, and supports selectable units with automatic conversion across load, torque and pressure measurement.
Frequently Asked Questions
What does the SSI do, and what can I connect to it?
It is a hand-held indicator that supplies excitation to a strain gauge bridge, reads the millivolt signal it returns, and displays the result in engineering units on a six-digit screen. It will work with load cells, torque sensors and pressure transducers with sensitivities between 1 and 5 mV/V, connecting through a DB9. If the sensor carries a TEDS memory the instrument configures itself from it on connection; if it does not, you scale it once through the keypad or from a PC and it remembers. It runs on its own battery, so no mains outlet is needed where the measurement happens.
What actually happens when I plug a TEDS sensor in?
The instrument reads the memory built into the sensor and configures itself from what it finds there — the sensor's identity, its excitation requirements, its valid measurement range and calibration data belonging to that individual unit rather than to its model. You do not enter a capacity, a sensitivity or a scale factor, because the sensor has already told the instrument. The practical effect is that connecting a different sensor is a mechanical act rather than a setup procedure, and the reading is right the moment the display settles.
Can I use the SSI with a sensor that has no TEDS memory?
Yes. Non-TEDS sensors are scaled manually, either through the menu on the keypad or from a PC using the instrument setup software, and the configuration stays in the instrument afterwards. Everything else — peak capture, tare, units, relays, logging — behaves identically. The difference is not in what the instrument can do, it is in who is responsible for the scale factor being correct. With TEDS the sensor is; without it, you are, which is worth remembering if that instrument is going to meet several different sensors.
Why does the display update under four times a second if it samples fifteen thousand times a second?
Because the two numbers are the same design decision. Divide the sample rate by the display rate and you get 4,096 — every number you see is the average of that many separate conversions. Divide the sample rate by the mains frequency and you get 256, meaning the conversions are taken a whole number of times per power line cycle. Averaging complete cycles cancels mains hum almost exactly, and averaging thousands of readings buries random noise, which together is how the instrument resolves to around a microvolt and a half. A display that flickered fifteen thousand times a second would be unreadable and would show you far more noise than signal. The speed is spent on quality instead.
What sensor sensitivity does the SSI accept, and does it affect resolution?
It accepts 1 to 5 mV/V into a maximum input span of 15 mV. Since excitation is 3 volts, a 5 mV/V sensor delivers exactly 15 mV at capacity and fills the input; a 2 mV/V sensor delivers 6 mV and uses a bit over a third of it. The converter splits its full span into 65,536 counts, so a 2 mV/V sensor still gets on the order of twenty-six thousand counts across its working range. That is far finer than the accuracy of any load cell, so in practice neither choice leaves you short. It is worth a glance at your existing certificates before ordering, simply to confirm your sensors fall inside the 1 to 5 mV/V window.
Why is the bridge excitation only 3 volts?
Two reasons, and both are advantages. Excitation current is where a battery-powered bridge instrument spends most of its energy, so a lower voltage is what makes sixty-five hours from a single charge possible at all. It also puts less power into the load cell, and a bridge that is not heating itself does not drift — which matters a great deal if the instrument is going to sit logging unattended overnight. Your sensors do not need 10 volts to work correctly; sensitivity is specified per volt precisely because output is a proportion of whatever you excite with. The signal is smaller at 3 volts and the instrument is designed around that.
How long does the battery last, and how is it charged?
Up to sixty-five hours from a full charge of the internal 2,600 mAh lithium-ion cell, with a full recharge taking up to eight hours. The supplied charger runs on any mains supply worldwide, 100 to 240 volts at either mains frequency, and delivers 5 volts DC at an amp. Sixty-five hours is roughly a working week of intermittent use, or two full days and nights of continuous unattended logging, which is the figure that actually decides whether a monitoring job is practical without running a cable to it.
How much can the SSI store, and how quickly can it log?
Sixteen thousand readings in internal memory, each stamped with the time and date, with a memory protection setting that prevents new data overwriting what is already there. The logging interval is programmable from 16.666 milliseconds up to about an hour and a half, which spans everything from capturing a fast event to recording a slow drift over days. That fastest interval is exactly one cycle of 60 Hz mains, which is the same power-line-locked design that governs the display. Readings come out over USB 2.0 to a PC, either as you go or afterwards.
What can the two relay outputs actually switch?
They are solid-state, rated at 120 mA and up to 350 volts on either AC or DC, and isolated from the signal common by several thousand volts. That combination — high voltage, modest current, very high isolation — suits driving an indicator lamp, a beacon, a small solenoid, a buzzer or a logic input on somebody else's controller, and the isolation means a fault on the switched side has a long way to travel before it reaches your measurement. What they are not is a power output. Anything that draws real current should be switched by a contactor or relay that the SSI drives, rather than by the SSI directly.
What software is available and what does each package do?
Three separate things. The instrument setup software is a free download and configures the SSI itself from a PC, which is easier than a keypad when there is a lot to enter. The TEDS reader-editor package reads and writes what is stored inside a sensor's memory, which is what a calibration laboratory needs when a transducer is recalibrated and its stored data should be brought up to date. The data logging package handles recording and exporting readings to a PC. They address different problems — one manages the instrument, one manages the sensors, one manages the results — and which you need depends on whether you are measuring, maintaining a fleet of sensors, or producing records.
Questions From The Field
My readings disagree with the sensor's calibration certificate.
Check the scale factor before anything else, and check it against the certificate for that exact serial number rather than the model. This is far and away the most common cause and it is entirely silent — a wrong factor produces steady, believable numbers that are simply the wrong size. On a TEDS sensor, confirm the instrument actually read the memory rather than falling back to a stored manual configuration. On a non-TEDS sensor, re-enter the figures from the certificate in front of you. If the factor is confirmed correct and a known load still reads wrong by a consistent percentage, that is worth reporting to us with both serial numbers.
The last digit never settles. Is something wrong?
Almost certainly not. Six digits of display on a microvolt-level input means the final digit is showing you real, tiny variations — in the load, in the structure, in the temperature of the room. A reading that never moved at all in its last digit would suggest something was rounding away information. Worry only if the movement is large, if it is rhythmic rather than random, which suggests interference finding its way in, or if it grows over a measurement, which suggests something warming up. Check the cable is not running alongside a motor lead, that the connector is fully seated, and that nothing is leaning on the sensor.
I swapped sensors and forgot to change the setup. How would I ever know?
Without TEDS, you very likely would not, and that is precisely the problem it exists to solve. The instrument has no way to detect that the sensor changed, so it applies the last configuration it was given and produces confident, stable, entirely wrong numbers with no symptom whatsoever. There is no error message because nothing has failed. With TEDS the question does not arise, because the instrument asks the sensor what it is every time it is connected. If your sensors are not TEDS-equipped, build a habit of confirming the displayed capacity or units against the sensor in your hand before the first reading, and consider specifying the TEDS option on future purchases.
My readings drift slowly over a long unattended log.
Look at the mechanical and thermal picture before the electronics. Overnight logs almost always span a temperature swing, and both the sensor and whatever it is mounted to will move with it — a fixture expanding is a real load, not an error. The instrument's low excitation deliberately minimises self-heating in the bridge, so the sensor warming itself is one of the less likely explanations. Look for something settling, creeping or relaxing in the mounting, and check whether the drift correlates with the building's own temperature cycle. If it does, the reading is telling the truth about something you did not intend to measure.
I need more than four readings a second on the screen.
The display rate is fixed by the averaging that gives the instrument its resolution, so it is not something to turn up. Two routes exist instead. For fast events, use peak and valley capture — the instrument follows the input far faster than it refreshes the screen, so it catches a maximum you could never have read off a display anyway. For a fast record, use the data logger, which goes down to one reading per mains cycle. If you genuinely need continuous high-rate output into a control system rather than a record, that is a signal conditioner and data acquisition job rather than a hand-held indicator one, and we can point you at the right modules.
The unit spent the night in a cold vehicle and reads oddly.
Let it come up to the temperature of the place it is working in before trusting it, and let the sensor do the same. The instrument is rated to operate from freezing to well above normal room temperature and to be stored colder still, so a cold night has not harmed it, but an instrument and a sensor at two different temperatures, both changing, will not agree with each other or with themselves. Give it time to stabilise, then zero. The same applies to a sensor brought in from outdoors — the steel takes considerably longer to settle than the electronics do.
Can I leave the SSI on charge permanently between jobs?
The supplied charger is designed for the cell in the instrument and a full charge takes up to eight hours, so ordinary practice is to charge it after use and disconnect it. With sixty-five hours available there is rarely a reason to keep it connected. For genuinely continuous monitoring over many days, plan the job around the battery figure and the logging interval rather than assuming mains will be available, and if it will not stretch far enough, tell us what you are trying to record and we will suggest an arrangement that does.
Can one SSI serve several sensors across a plant?
That is exactly the situation it was designed for, and it is where the TEDS capability stops being a convenience and starts being the reason to buy it. One instrument moving between many sensors is the arrangement most likely to produce a wrong scale factor, and it is the arrangement where automatic configuration removes that risk entirely. Specify the TEDS option on the sensors, and the instrument identifies each one as it is connected. It also keeps your records coherent, since every sensor carries its own calibration data with it rather than depending on somebody maintaining a list.