ΔVbe + DMM = Celsius, Kelvin, Fahrenheit, and Rankine thermometer
Combining an accurate temperature sensor with a standard digital multimeter can make an inexpensive, accurate, and useful thermometer.
A recent Design Idea, BJT is accurate sensor for absolute temperature in Kelvin and Rankine, was based on a 1991 application note (PDF) by a legendary guru, the forever remembered Jim Williams. In his article, Williams demonstrated that, when used as ΔVbe sensors, ordinary unselected transistors give temperature readings accurate to a fraction of a degree without calibration:
“…randomly selected 2N3904s and 2N2222s … showed less than 0.4°C spread over 25 devices from various manufacturers.”
Wow the engineering world with your unique design: Design Ideas Submission Guide
As shown in BJT is accurate…, the basic math of ΔVbe can be cooked down to a simple and easy to remember (hah!) linear-in-absolute-temperature relationship: ΔVbe/°C = Log10(Current-ratio)/5050. Therefore, if we want any given ΔVbe/°C, the required is just Current-ratio = 10^(5050 ΔVbe/°C).
For example, for ΔVbe/°C = 100uV, Current-ratio = 10^(5050 * 100uV) = 10^(0.5050) = 3.20. This ratio is implemented in Figure 1’s simple circuit for a 100uV per Kelvin output.
Figure 1 An ordinary BJT Q1 makes an accurate 100uV per unit Kelvin absolute temperature sensor.
Okay. So. What’s it good for? One plausible application is, as frequent contributor Nick Cornford has shown in several ingenious designs:
- Newer, shinier DMM RTDs—part 1 and part 2
- Dropping a PRTD into a thermistor slot—impossible?
- DIY RTD for a DMM
that the combination of an accurate temperature sensor with a standard digital multimeter can make an inexpensive, accurate, and useful thermometer.
Nick’s favorite sensor is the super-versatile platinum RTD, but as Williams showed, a humble (and super cheap) 2N3904 (or similar) BJT might also fill the bill. That’s assuming that its package-limited −55 to +150°C temperature range is adequate. And that’s also assuming that it gets a little help from its friends, such as Figure 2’s zero-drift op amp that boosts the output span to a DMM-friendly 1mV per unit Celsius, Kelvin, Fahrenheit, and Rankine.
Figure 2 A zero drift, 5uV max offset A1 rescales 100uV/°K by 10x to 1mV/°C and by 18x to 1mV/°F.
Of course, Kelvin and Rankine absolute temperature measurements are absolutely less frequently useful than the common Celsius and Fahrenheit scales…which is where Figure 3 comes in:
Figure 3 Connect the DMM’s plus lead to the appropriate figure 2 output, and the minus lead to the correct precision 0° offset terminal, to re-zero 273K to 0°C and 460R to °0F.
V+ can be anywhere from 3 to 6 volts. Current consumption at 3v is barely more than 1mA, dominated by the Z1 shunt reference, so two AAs will support 2000 hours (nearly three months) of continuous operation. A single CR2032 lithium coin will hold up for 10 non-stop days.
Thanks, Nick and Jim!
Stephen Woodward‘s relationship with EDN’s DI column goes back quite a long way. Over 200 submissions have been accepted since his first contribution back in 1974. They have included best Design Idea of the year in 1974 and 2001.
Related Content
- BJT is accurate sensor for absolute temperature in Kelvin and Rankine
- Newer, shinier DMM RTDs—part 1
- Newer, shinier DMM RTDs—part 2
- Dropping a PRTD into a thermistor slot—impossible?
- DIY RTD for a DMM
The post ΔVbe + DMM = Celsius, Kelvin, Fahrenheit, and Rankine thermometer appeared first on EDN.
