EI-1034 Datasheet | LabJack
« Close

Datasheets and User Guides

App Notes

Software & Driver


EI-1034 Datasheet

The EI-1034 is made by EIC.  It is temperature probe with excellent accuracy that is easy to use with any LabJack.


The EI-1034 is a universal temperature probe that consists of a silicon type temperature sensor mounted in a waterproof 316 stainless steel tube.  It uses the LM34CAZ precision silicon temperature sensor with a typical room temperature accuracy of ±0.4 °F (±1.0 °F max). Because of the high-level linear voltage output and high accuracy, this probe is easier to use and superior to thermocouples, thermistors, or RTDs, for many applications in the range of 0 to 230 °F (temperature range varies with positive supply voltage, negative supply voltage, and LabJack model). The probe is suitable for air and liquid applications, and can be conveniently secured into pipes, vessels and chambers by using available ¼ inch compression fittings.

The EI-1034 is intended to be connected to a LabJack for 5-volt power but can be used as a stand-alone temperature sensor when connected to a DVM and a power supply in the range of 5 to 30 volts.

Electrical Connections

Three wires require connections; they are +5 volts (red), ground (black) and signal output (white). These wires can be connected to the appropriate terminal on the LabJack or other power supply in the case of using the sensor as a stand-alone unit. The output wire (white) connects to an analog input and will normally output a voltage of approximately 0.77 volts at room temperature.

Cable Length

The probe has a 10k internal resistor from signal to ground that helps keep the signal stable when sinking current or driving capacitive loads.  The cable length of the probe can be extended to 25 ft without serious degradation in performance. If the user desires to extend the length of the cable beyond 25 ft (up to 500 ft) then a resistor of 10K ohms should be inserted in series with the white wire. The resistor should be placed at the 6 ft length of the probe. When using a series resistor of 10K ohm the user should consider the voltage drop across the resistor when calculating the final temperature measurement.

Stream Mode

Temperature readings are not often acquired in high speed stream mode, but if they are note that the micropower drive circuit of the LM34 can be subject to settling errors in multiplexed applications.  Our testing on a T7 found that 50 μs of settling is required for the EI-1034.  Auto Settling can be as little as 10 on the T7, so you might need to write STREAM_SETTLING_US = 50.  Read more about settling time in the Analog Input Settling Time App Note.

Low Temperature Operation

The low temperature range of the EI-1034 can be extended to -40 °F by adding a 100K resistor to a negative supply voltage.  The Vm- supply on the U6 and T7 is handy for this.  Note that if you don't have a negative voltage available but do have an isolated voltage available such as a battery or wall-wart, you can connect it backwards to make a negative voltage. A standard wall plug-in supply can be used in the range of 5 to 15 volts. A 9-volt battery is also a good source for a negative voltage. Care must be taken to connect the positive terminal of the isolated supply to the GND wire (black) of the EI-1034 and the negative terminal of the supply in series with a 100K resistor to the white wire of the EI-1034.

Figure 1

Formulas to Calculate Temperature From Measured Probe Voltage

°F = 100*volts

°K = (55.56*volts) + 255.37

°C = (55.56*volts) + 255.37 - 273.15

LabJack U12 Quickstart

Connect the red wire to +5V, black wire to GND, and white wire to AI0.

Run LJlogger.  By default, the first row will be set to Channel = 0 SE, and the Voltage column should show something around 0.77 volts if the EI-1034 is at room temperature.

To improve resolution, you need to use gain which requires a differential channel.  Add a jumper wire from AI1 to GND, then in the desired row of LJlogger set Channel = 0-1 Diff.  Now you can adjust Gain so you are using the smallest range possible.  For example, the +/-2V range would allow temperatures up to 200 degrees F.

To make LJlogger display degrees C, enter 55.56 for multiplier and –17.78 for offset in the appropriate row.  To make LJlogger display degrees F, enter 100.0 for multiplier and 0.0 for offset. The scaled temperature will appear in the “Scaled Data” column.

U3/U6/UE9 Quickstart

Connect the red wire to VS, black wire to GND, and white wire to AIN0.

Run LJLogUD.  By default, the first row will be set to +Ch=0 and -Ch=199 (single-ended), and the Voltage column should show something around 0.77 volts if the EI-1034 is at room temperature.

To make LJLogUD display degrees C, enter a scaling equation such as "y=55.56*c - 17.78" in the desired row.  Note that "c" in this example means it will use the voltage from the 3rd row, so use the appropriate variable from "a" to "p".  To make LJLogUD display degrees F, enter a scaling equation such as "y=100.0*c" in the desired row.

U3 Comment:  You will get better resolution using a low-voltage channel (FIO or EIO) on the U3-HV, and that is the only option on the U3-LV.  Connect the white signal wire to FIO4 rather than AIN0, and in LJLogUD set +Ch=4 in any row.

T4/T7 Quickstart

Connect the red wire to VS, black wire to GND, and white wire to AIN0.  If the probe is at 70 degrees F, the voltage from AIN0 to GND should be 0.70 volts.

Traditional Technique:  Acquire the voltage and in software multiply it by 100 to convert to degrees F.

Run LJLogM.  By default, the first row will be set to AIN0, and the Voltage column should show something around 0.77 volts if the EI-1034 is at room temperature.

To make LJLogM display degrees C, enter a scaling equation such as "y=55.56*c - 17.78" in the desired row.  Note that "c" in this example means it will use the voltage from the 3rd row, so use the appropriate variable from "a" to "p".  To make LJLogM display degrees F, enter a scaling equation such as "y=100.0*c" in the desired row.

AIN-EF Technique:  Configure an extended feature on the applicable channel to apply a slope of 100, and then read AIN0_EF_READ_A to get the scaled value.

In Kipling, go to the Analog Inputs tab, click the "+" at the far right of the AIN0 row, set Extended Feature to "Slope/Offset", and set Slope=100.  Now you can read AIN0_EF_READ_A in any software, such as LJLogM, and get the scaled value.

T4 Comment:  You will get better resolution using a low-voltage channel (FIO or EIO) on the T4.  Connect the white signal wire to FIO4 rather than AIN0, and in LJLogM use "AIN4" in any row.



Range with 0/5 volt supply:

+10 to +230 °F (-12 to +110 °C) with the LabJack U12

0 to +230 °F (-17 to +110 °C) for the LabJack U3 or UE9


+/- 0.4°F Typical Room Temperature

+/- 1°F Max Room Temperature

+/- 2°F Max 0°F to 230°F

+/- 3°F Max -40°F to 0°F

Sensor device in probe: LM34CAZ

Cable length: 6 ft supplied max 25 ft user extended

Power: +4 to 35 VDC at 100-400 µA

Output Current: 10 mA

Note: When operating at voltages less than 5 Volts the maximum operating temperature is reduced, typically at 4 Volts supply the maximum temperature limit is 200 °F

Probe dimensions: 6 in x 0.25in diameter.  Metal tube is 316 stainless steel, product number SS-14-6CLOSED from Omega.

Possible fittings:

More information:
LM34 Datasheet
Temperature Sensors Application Note


Special Notice Regarding The EI-1034 Cable

Although the temperature sensor and associated electronics are rated for 110 degrees C, the normal cable is only rated for 80 degrees C. We have tested the cable and probe at 150 degrees C, and have noticed the cable gets soft at the high temperatures but continues to function. When the cable and probe were returned to normal temperatures, no degrading was observed in the cable or probe. Also at the low temperatures, the cable is only rated to -20 degrees C where the sensor and associated electronics are rated lower. Testing the probe with the wire at the lower temperatures showed normal operation and no degrading of the cable when returned to normal temperatures. The user should be aware that even though the probe itself can operate at the rated temperatures the use of the cable in environments of over 80 degrees C and lower than 20 degrees C is at your own risk.

Manufactured by:


Email: [email protected]



If I use EI1034 Thermo sensor to connec to LJU3-LV DAC, I am not sure which wires to which connectors on DAC. Suppose I wan to connect to DAC0 on LJU3-LV. Should I connect Red wire of the sensor to an external power supply, white wire to DAC0, and black wire to GND port above DAC0.


Can I connect the black wire to one of the connector on U3 so that I don't need any power supply?


Thank you!

The DAC is a Digital to Analog Converter, so it takes a digital control signal and outputs a corresponding voltage. I don't think that will help us with the EI-1034.

The normal connections to a U3 are Red to a Vs terminal (+5V), Black to GND terminal, and White to an FIO or EIO line.


I am using this with a U-6 Pro. I have the red wire to VS the white wire to FIO3 and the black to ground. i am trying to use the U6 single io example to read the voltage, but it does not seem to be working. I have set the channel to #3 and I have tried all the drop down choices. it will not give me a voltage. I checked the voltage using a multimeter at FIO3 and ground and it gives me the correct voltage using the formula on the data sheet. Any advice on what I may be doing wrong with the LabView code?



The EI-1034 outputs an analog signal (10mV/degF) that you need to connect to an analog input such as AIN3.

I don't have any of the AIN connections open. What are my options for hooking up this temperature probe?

If you have any of AIN4-AIN13 available you can access them through the DB37 connector.  Otherwise you need to add a Mux80 or a 2nd LabJack.

Hi, I just want to check some resolution calcs on this sensor, when it is measured in Celsius and configured for the extended temp range as outlined in the data sheet: The AIN for the device, with 16-bit resolution on a UE9, is still swinging 0-5v on 65536(2exp16) levels, yielding voltage resolution of 76microVolts. The temp sensor range in this config can swing from -50F to 300F, or -45.5C to 148.9C, a range of 194.4.  I understood, from other comments, that the sensor resolution for the EI1034 is 10mV per degF, which I convert to 21.2mV per degC. 

21,200 microVolt per degC / 76 microVolt = ~ 279

Temp swing: 194.4 / 279 = ~0.67 degC resolution

Have I done this calc correctly? If not, can you point me in the right direction?


Just looking at resolution, and not considering noise/accuracy/linearity/etc, I get something different:


5V span / 16-bit resolution = 76 uV/count


1.8 degF/degC

10000 uV/degF * 1.8 degF/degC = 18000 uV/degC

18000 uV/degC => 0.0000555 degC/uV


0.0000555 degC/uV * 76 uV/count = 0.0042 degC/count


Yup, I see the error in my Celsius conversion. 18mV per degC is correct, not 21.2mV / degC -- I was using some goofy conversion app, not my trusty calculator.

A couple of questions:

1) Is the temperature sensor at the tip of the probe?

2) How do temperature gradients along the stainless steel tube affect the performance?  For example, if the probe is mounted using metal compression fittings to a metal housing that is heated to and maintained at 40 degrees C, over time the probe's stainless steel tube will equilibrate to the temperature of the metal housing (40C).  If air at a constant 30 degrees C is then flowed over the tip of the probe and everything is allowed to thermally equilibrate, a temperature gradient will exist (40 C where is compression fitted to 30C where the cooler air is passing over the tip).  In this situation will the probe temperature read the air temperature at 30C, or will it read slightly higher due to the temperature gradient?

1.  Yes.

2.  I would see that yes there will always be gradients, the question is how much.  In your example, if the air in the center of the pipe is 30 degrees, and the air 6 feet away from the outside of the pipe is 40 degrees, there will be a gradient of temperature change from the center of the pipe to the point 6 feet away from the pipe.  There will be a gradient, for example, from the center of the pipe (30.0 deg) to the inner surface of the pipe (e.g. 30.1 deg), however slight.

In your example, I suspect the error would depend on how much of the probe was in the pipe compared to out, among other things.  I don't have a feel for the magnitude of the errors.  If it seems like it might be a concern in your setup, you will have to do a couple experiments.

What is the IP rating of this sensor? Can it be buried at 15 meters underground and survive?



I would contact EIC for a comment on that, but I suspect they will just say it does not have an IP rating.  The weak link is likely the junction of the cable and tube, which is protected/sealed with a product from Sumitomo.

Better to have T7 (including Kipling's EF) Quickstart just like U3/U6/UE9 Quickstart

Added a T7 Quickstart section that uses LJLogM.

Thanks for T7 Quickstart section that uses LJLogM, but I still think it will be much better to have Kipling 3 EF


I've connected the EI-1034 and have been attempting to make some measurements in order to gage the accuracy of the probe. When I submerge it in ice water, the voltage drops steadily downwards, but at about .55 V, the voltage stops going downward and returns a static 2.45 V value. I have tried several times and this result is consistently returned. The data sheet says the sensor should be valid 0 to 230.

What hardware are you using?  Is this a U3-HV, AIN0?  The voltage range corresponding with 0°F to 230°F should be 0V to 2.30V.  Since the LabJack hardware does not have a negative voltage available, the readings will stop getting lower at 0°F, but it should be possible to measure ice water. I would expect the voltage to be near 0.32V when the probe is in ice water.  Can you compare readings with a second EI-1034 probe?  I would also advise removing all other connections from the LabJack device except VS/GND/signal from your EI-1034, just in case something else is causing a significant ground offset error, which is impacting your readings from the sensor.

I'm using a U3-LV, and I've tried the probe on the FIO1, FIO2, and FIO3 ports. I've tried the same thing with a separate EI-1034 probe, it it does not have this problem on the same ports and is able to measure down to into the 30s (F). However, this brings me to another point, is there an easy way to calibrate these probes? The data sheet lists +/- 1 F for the accuracy though in an ice water bath it returns ~ 36F. I don't know if it's as simple as adjusting the scaling equation, assuming the probe does not drift, or if there is a more sophisticated solution.

But to the original concern, it seems the initial EI-1034 is likely defective hardware.

Yes, it seems that the initial EI-1034 is defective in some way, since it should agree with other sensors in the water.  Please send an email to [email protected] for an RMA.

There is no way to calibrate the probe, though you could add your own calibration within software.  I think that the water is actually 36F instead of 32F.  36F is well within the expected realm of water with ice in it, especially if the water is in a non-insulated glass in open air. You could also check the temperature of boiling water (accounting for the expected delta associated with atmospheric pressure/elevation).

BinarySolo's picture

Can someone confirm that this will work properly if used on a U3-LV using the Vs as the supply voltage for the sensor? The Vs I believe is sourced from the USB bus and is relatively stable but not perfect. Is the output voltage from this sensor independent of the supply voltage, or should this be used with LJTick-VRef in order to achieve accuracy?

labjack support's picture

Yes, VS is a fine supply voltage.  The LM34 is not ratiometric, but rather is specified as giving 10 mV/degF for any supply voltage.  The applicable spec from the LM34 datasheet is "Line regulation", and TI says typical +/-0.01 mV/V and max +/-0.05 mV/V.  Per USB spec the USB 5 volt supply voltage should be 4.75-5.25 volts.  Worst case the supply voltage could change by 0.5V which could cause a worst case change in output of 0.025 mV which equates to only 0.0025 degrees F.