Vhi, Vlow for thermistor measurement

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
NiNa_4043086
Level 3
Level 3
First like received

I would like to measure temperature on my system using a thermistor. Can I connect the Vhi and Vlow to Vdd and '0' internally on PSOC to save two GPIO?

If yes, how it will affect the measurements accuracy?

Many Thanks!

0 Likes
1 Solution
MiSt_296941
Level 4
Level 4
25 replies posted 10 replies posted 10 questions asked

I would not connect the Thermistor to power all the time because it will self heat and introduce an inaccuracy in your measurement.  In my project, I fed it with a VDAC so I could set the voltage to what I wanted and followed that with an Opamp in the PSoC to get enough drive current.  I could then turn the Opamp on only during the time I was going to make the measurements.

I tested using an internal 'ground' vs running a pin to the grounded side of the Thermistor and found that indeed there was a slight difference in the reading from the ADC, but it was not very significant.  Once the reading was converted to temperature I could not see a 0.1C change with the two methods.

Mike.

View solution in original post

9 Replies
odissey1
Level 9
Level 9
First comment on KBA 1000 replies posted 750 replies posted

ninatan,

Internal connections have undefined resistance of 100-700 oHm, which will affect measurement result. In such connection, when current flows through Rref and Rtherm, voltage potentials on high and low side won't be equal to Vhi / Vlow. Using 3-point measurement this effect is eliminated.

/odissey1

0 Likes

Thanks,

But as I understand there is almost no current flowing through the internal resistance because Rbuf input to ADC is minimum 10MOhm (see below my schematic) so practically high ~ Vhi and Low~gnd. What do you think? I actually don't need more than 0.5 C precision.

One more question - Is it OK to place the Thermistor at the low side (pulled low as you can see below)?

pastedImage_1.png

0 Likes

ninatan,

I would try both to see the difference. Depending on the accuracy requirements it might be OK (probably good for 1 degC, certainly not for 0.01 degC). Keep in mind that Vdd inside is not guaranteed to be exactly same as Vdd outside, same to Gnd.

/odissey1 

0 Likes

Thanks!

Can you explain why in steady state Vhi will not be equale to Vdd and Vlow to GND? What is the variations from these values? I want to calculate the error that I will get.

regarding the other question -

I did not sure that I got it. Is it is OK to place the thermistor at the low side (as I mentioned above)? I mean that the thermistor components can handle it?

0 Likes
Vasanth
Moderator
Moderator
Moderator
250 sign-ins 500 solutions authored First question asked

Hello Nimrod,

This application note should be also helpful.

https://www.cypress.com/documentation/application-notes/an66477-psoc-3-psoc-4-and-psoc-5lp-temperatu...

Best Regards,
Vasanth

0 Likes
MiSt_296941
Level 4
Level 4
25 replies posted 10 replies posted 10 questions asked

I would not connect the Thermistor to power all the time because it will self heat and introduce an inaccuracy in your measurement.  In my project, I fed it with a VDAC so I could set the voltage to what I wanted and followed that with an Opamp in the PSoC to get enough drive current.  I could then turn the Opamp on only during the time I was going to make the measurements.

I tested using an internal 'ground' vs running a pin to the grounded side of the Thermistor and found that indeed there was a slight difference in the reading from the ADC, but it was not very significant.  Once the reading was converted to temperature I could not see a 0.1C change with the two methods.

Mike.

Thanks Mike,

As I mentioned I am trying to measure the thermistor temperature with less GPIOs as I can (preferably one). So I will try to avoid adding a GPIO to fed the thermistor.

I will only measure temperature below 0C, at this range the power developed on the thermistor is less than 0.2mW. The thermistor has a dissipation factor of 23mW/C so we get in worst case a self heating of 0.0087C. At accuracy goal of 1C it is negligible.

Did you put the thermistor on the low side (pulled down)?

0 Likes

No.  In our circuit the 10K is below the Thermistor.  Not sure it makes any difference as this is a ratiometric measurement.  In other words it's just as important to accurately measure the 10K as it is the Thermistor since both of these results are fed into the function to give you the temperature. 

That being said, since you are connecting the Thermistor and 10K to the rails, then how are you measuring beyond the rails with the ADC?  Buffer bypassed?

Mike.

0 Likes

I will not need to use off rail measurements. I have to buffer the ADC because of its relative low resistance which will alter my measurements.

I plan to use a pulled up resistor of 150K Ohm. The thermistor resistance vary from 1K to 360K Ohm. Vdd is 3.3V. Therefore Vthermistor will vary from 0.22 to 2.32 V. Than I can use a buffered Rail to Rail  Mode:

pastedImage_0.png

0 Likes