1
\$\begingroup\$

I'm looking for suggestions on what type of temperature sensor to use. I need to have it attached to the heat sync of a 150V, 40A 3-phase motor controller. The motor controller is in a enclosed space and I need the monitor the heat being dissipated by the controller remotely.

Whenever the motor controller is run, I see a massive spike in the temperature reading, due to it being analog, coming from the induced voltage from the EMF produced by the 3-phase output.

Currently I have this sensor attached http://www.analog.com/media/en/technical-documentation/data-sheets/TMP35_36_37.pdf

My current thoughts are to use a simple thermistor with a shielded twisted pair attached to it, but I thought I'd see if anyone had an idea about a way to have a sensor that would be agnostic to the EMF.

\$\endgroup\$
  • \$\begingroup\$ I think you should add how the sensor is wired up for better help. Is it just three/four long wires going to the remote monitoring device? \$\endgroup\$ – Dejvid_no1 Oct 3 '17 at 19:06
  • \$\begingroup\$ It is 3 long wires going to the device. My problem is I didn't comply with the datasheet as Transistor pointed out. \$\endgroup\$ – eenelson Oct 3 '17 at 19:43
1
\$\begingroup\$

enter image description here

The datasheet gives some strong advice on decoupling.

Note the 0.1 µF bypass capacitor on the input. This capacitor should be a ceramic type, have very short leads (surface-mount is preferable), and be located as close as possible in physical proximity to the temperature sensor supply pin. Because these temperature sensors operate on very little supply current and may be exposed to very hostile electrical environments, it is important to minimize the effects of radio frequency interference (RFI) on these devices. The effect of RFI on these temperature sensors specifically and on analog ICs in general is manifested as abnormal dc shifts in the output voltage due to the rectification of the high frequency ambient noise by the IC. When the devices are operated in the presence of high frequency radiated or conducted noise, a large value tantalum capacitor (±2.2 µF) placed across the 0.1 µF ceramic capacitor may offer additional noise immunity.

Did you comply?

The datasheet continues ...

enter image description here

Figure 32 illustrates a way to convert the output voltage of a TMP35/TMP36/TMP37 sensor into a current to be transmitted down a long twisted pair shielded cable to a ground referenced receiver. The temperature sensors are not capable of high output current operation; thus, a standard PNP transistor is used to boost the output current drive of the circuit. As shown in the table in Figure 32, the values of R2 and R3 were chosen to produce an arbitrary full-scale output current of 2 mA. Lower values for the full-scale current are not recommended. The minimum-scale output current produced by the circuit could be contaminated by ambient magnetic fields operating in the near vicinity of the circuit/cable pair. Because the circuit uses an external transistor, the minimum recommended operating voltage for this circuit is 5 V. To minimize the effects of EMI (or RFI), both the circuit and the temperature sensor supply pins are bypassed with good quality ceramic capacitors.

See also the 4-20 mA section.

\$\endgroup\$
  • \$\begingroup\$ I guess I need to learn how to read. I did not comply with the datasheet, nor did I bother to check this section. I will implement the circuit specified and report back with the results, but from what I'm seeing here explains why I'm seeing issues. I made the assumption I could run the sensor the same way as I did in initial testing. Thanks. \$\endgroup\$ – eenelson Oct 3 '17 at 19:47
  • 1
    \$\begingroup\$ When thinking about leaving out recommended capacitors always ask yourself, "What new information do I have that the guys who designed it don't?" The answer is usually "none". Obey their recommendations. I hope it works out. Thanks for the vote. \$\endgroup\$ – Transistor Oct 3 '17 at 19:51
0
\$\begingroup\$

Proper shielding of the sensor and cable is likely your best bet. If the thermal response time of whatever you are measuring is not extremely fast the entire sensor can be shielded provided sufficient thermal conduction is provided.

Failing all that you may need to add circuitry, and send power, to the business end to send the temperature information as a differential voltage over two wires, or even make it digital. That is, make it an active sensor.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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