I am working on a project where we have to record breathing data. For that we ordered a LabJack U3-LV along a with an amplifier (datasheet) and a belt which is some kind of strain gauge I think. We have no background in electronics so having a hard time extracting data from the setup. There are certain dip switches on the amplifier. We just connected the belt / strain gauge to the equipment as shown in the figure but we just get some random spikes (we are using Labjacks' LJStream UD) and not continuous data as shown in this video.Can somebody please help as to how we can use the setup?

  • \$\begingroup\$ Please use the schematic button to add a schematic and link datasheets for all devices. If you have multimeter voltage readings, tell us. \$\endgroup\$ Apr 24, 2019 at 9:25
  • \$\begingroup\$ Most likely all you need is shielded twisted pair cable (STP) with shield grounded to ADC card \$\endgroup\$ Apr 24, 2019 at 9:44
  • \$\begingroup\$ @ Sunnyskyguy EE75, the person in the demo video seems to be having the same kind of wire that we have. \$\endgroup\$
    – ArkanSaaS
    Apr 24, 2019 at 9:48
  • \$\begingroup\$ I cannot tell from that. Then use earth ground to direct or via 10nF cap to 0V due to common mode power supply noise \$\endgroup\$ Apr 24, 2019 at 10:00
  • \$\begingroup\$ A piezo-electric is a very different thing from a strain gauge. \$\endgroup\$ Apr 24, 2019 at 22:45

1 Answer 1


Since your sensor has a piezoelectric output, it has a charge output that will provide a voltage that swings both positive and negative with respect to ground into a high value load resistance. I would try tying one output to both the "IN A-" input of the amplifier and the VREF pin, with the other output tied to the "IN A+" pin, with switch #6 on and switch #5 off. This will allow the amplifier to remain in the active region as the voltage goes negative. The belt spec sheet does not indicate whether a load resistor is built into the belt sensor, so you may need a high-value resistor between "IN A-" and "IN A+", in the 100K to 1M range. The resistance you choose will affect the low frequency response, so you may have to experiment to find the right value.

  • \$\begingroup\$ Thank you John, I will try this and come back. Also, the current position of the dip switches is 4, 5 and 10 ON (this is how i got it) while others are OFF. How do I find what should be the configuration of the dip switches for my particular application? \$\endgroup\$
    – ArkanSaaS
    Apr 24, 2019 at 13:48
  • \$\begingroup\$ The datasheet mentions that pin 6 will center a bipolar signal at 1.25 volts, allowing it to swing positive and negative around this point. If you use channel A, you set the gain with switches 8,9, and 10. Since the belt data sheet mentions a voltage output of 20 to 50 millivolts (0.02 to 0.05 volts) I would start with a gain of 51 (switch 9 on). This would mean your output is multiplied by 51 so you would get .02 X 51 or 1.02 volts on the output. Also note that if you install as shown in figure 2 on the amplifier link, your channel "A" data will be on analog channel 4 in the labjack. \$\endgroup\$ Apr 24, 2019 at 15:05
  • \$\begingroup\$ I got this from the consumer service "The UFI model 1132 transducer is a piezo-electric sensor with a built-in charge reservoir capacitor. The output of the 1132 is approximately 20 mV for a normal breath, and so a gain of about 100 will be sufficient. The 1132 output can be connected single-ended and DC coupled, but the input impedance should be 1 MegOhm or more. This information should help you get your signal conditioning configured correctly. It sounds like maybe you have the amp setup for AC coupling?" \$\endgroup\$
    – ArkanSaaS
    Apr 24, 2019 at 16:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.