How to Verify a Vishay Strain Gauge with a Multimeter: A Five-Step Receiving Checklist

I handle component buying for a mid-size sensor company. We spend roughly $80,000 a year on strain gauges and force sensors, and for the past six years I have run the receiving checks on every Vishay order that comes through our door. That habit started because I got burned twice on parts that looked right but were not. It is also why I still personally check the load cells we use in our calibration jacks. A Vishay strain gauge is usually at the heart of those load cells, and a multimeter catches most common problems before they become expensive field failures.

This is not a calibration procedure. It is a practical 15-minute checklist. If your company buys Vishay strain gauges, or if you have a hydraulic jack load cell that started giving erratic readings, these five steps will catch about 80% of the usual trouble with a basic digital multimeter.

Step 1: Verify the part number before you trust the box

The first step does not require a meter at all. Read the label and compare it against the datasheet before you open the package. Vishay strain gauge part numbers carry a lot of information. A part like CEA-06-250UN-350 tells you the carrier material, the grid length, and the nominal resistance. If the part number does not match your order, nothing else on this checklist matters.

I once had a supplier send a visually identical box that was three characters off from what we ordered. The gauge pattern was different, and if we had installed it without checking, every load cell in that batch would have had the wrong geometry. It would not have shown up as a short circuit. It would have shown up as inconsistent output once the cells were already assembled.

Ask the distributor for lot traceability and a copy of the manufacturer datasheet before you pay, not after. A vendor that lists these things upfront is usually the one whose total price matches the invoice. A vendor that hides them until after the order usually adds a line item called "documentation" later. I have seen that more than once.

Step 2: Measure the gauge resistance with the DMM in ohms mode

Set your digital multimeter to the ohms range. Auto-ranging is fine for this. If the meter does not have auto-ranging, use the 200 ohm range for a 120 ohm gauge or the 2000 ohm range for a 350 ohm gauge.

Touch the two probes together first and note the residual reading. Most meter leads add a little resistance, usually around 0.2 ohms. That is small enough to ignore for this check, but it is good practice to subtract it if you are recording exact numbers.

Now place the probes on the gauge solder tabs or lead wires. The resistance should settle quickly near the nominal value on the label. Vishay strain gauges typically come with a fairly tight resistance tolerance, often around 0.3%. So a 350 ohm gauge should read somewhere around 348.95 to 351.05. A 120 ohm gauge should sit close to 120 ohms as well.

What I care about more than the exact number is stability. The reading should not drift. If the resistance slowly climbs or falls while the probes stay still, that is a warning sign. A cracked grid can open and close with temperature changes, and a bad solder joint does the same thing.

Step 3: Wiggle the lead wires and watch the display

Keep the meter in ohms mode. Gently move the lead wires where they connect to the gauge tabs. If the reading jumps around or briefly goes to over-range, the connection is suspect.

This is basically a poor mans continuity test. It will not catch every defect, but it catches the common ones: cracked foil, a poorly crimped lead, or a solder joint that passes when it is cold and fails when it heats up.

If you are checking a rosette or a multi-element gauge, measure every element separately. Each element should read about the same nominal resistance. If two elements are supposed to be identical and one reads 350 ohms while the other reads 340 ohms, do not use that gauge. I have rejected gauges for exactly that reason.

Step 4: Check isolation to ground, especially on mounted gauges

This is the step most people skip, and it is the one that has saved us the most money.

If the strain gauge is already mounted to a metal part, or if you are testing a load cell, switch the multimeter to the megohm range. Put one probe on the metal body or the cable shield, and put the other probe on each gauge lead one at a time.

The meter should show over-range or a very high reading. On most digital multimeters, you will see OL or a value in the hundreds of megohms. If you see a few megohms, or a reading that slowly drops while you watch it, that usually means moisture has gotten into the system. I have seen new-looking load cells fail this test because they were stored in a damp warehouse.

Important warning: do not use a high-voltage insulation tester or megger on a strain gauge circuit. Many insulation testers apply 500 volts or more, which can damage the foil grid or the bonding system. A normal digital multimeter uses a much lower voltage in resistance mode, which is safer for this kind of check. For a formal inspection, Vishay Micro-Measurements publishes technical notes on insulation resistance, and it is worth reading those before you set up your own procedure.

On bare foil gauges that are still in the package, this test is less relevant because there is nothing mounted yet. Once the gauge is installed on a metal specimen or inside a load cell, though, I would not skip it.

Step 5: If the gauge is part of a bridge, check the zero output

Most force sensors are not a single gauge. They use four strain gauges wired into a Wheatstone bridge. A load cell inside a jack is a good example. If one of those four gauges fails, the bridge will often still pass a basic resistance check, but the output will be wrong under load.

First, identify the excitation and signal wires from the datasheet. Set the meter to ohms and measure across the excitation pair, then across the signal pair. A healthy bridge with four 350 ohm gauges will usually read somewhere around 350 ohms at both pairs. If you see a much lower value, like 120 ohms, or a completely open circuit, one of the internal connections is wrong.

Then check the live zero. This part needs a little more than the multimeter itself because most handheld meters do not supply excitation voltage. I use a small regulated DC power supply set to 5 volts. Power the sensor, let it warm up for a couple of minutes, then set the meter to millivolts DC and read the output with no mechanical load applied. The value should be close to zero, though it does not have to be exactly zero. If the datasheet says the sensor is rated for 2 mV per volt of excitation, then at 5 volts full scale is about 10 millivolts. A zero balance within about 0.1 millivolts is reasonable for many sensors. If your reading is several millivolts at zero load, something is not right.

Finally, press on the load cell or jack by hand and watch the meter. The voltage should change smoothly and return to the original zero reading when you release the pressure. If the reading jumps around or does not respond at all, one of the gauges is likely bad.

Notes from the receiving bench

This five-step list works for our shop because we put a small number of gauges into every load cell, and a field failure is expensive. If you are doing low-volume prototype work, you might not need to check every single gauge this carefully. But I would not skip the resistance measurement and the isolation check. Those two steps catch the most problems for the least effort.

I am not telling you to ignore cheaper alternatives to Vishay. I am telling you to verify what the cheaper number actually includes. In Q2 2023, we accepted a lot of supposedly equivalent gauges to save about $300 on the order. Eleven out of sixty gauges gave unusable bridge balance readings in the same week. The rework, engineering time, and repeat calibration cost us roughly $1,800. No one sent us a separate invoice that said "hidden cost." It just showed up in our labor hours.

Honestly, I am not sure if that bad lot was damaged in transit or if it was never good to begin with. My best guess is moisture exposure, because the isolation readings were all over the place. What I know for sure is that if we had run this checklist on a sample before installing them, we would have caught it on day one.

There is something satisfying about catching a bad part at the receiving bench instead of in the field. The whole process takes about fifteen minutes. It is boring, but it is a lot cheaper than explaining to a customer why their calibration jack drifted.