How do control units typically measure PT1000 (RTD) temperature sensors ?

Post Reply
freeflyer
Posts: 2
Joined: Sun Dec 19, 2021 11:57 am

How do control units typically measure PT1000 (RTD) temperature sensors ?

Post by freeflyer »

I am using a CCS Type 2 connector which has PT1000 RTD temperature sensors on the DC and AC power pins.

What method do control units typically use to measure RTD sensors ?

I have used NTC thermistors before, consisting of a pull up resistor on an analogue input, but I have never used RTDs before.

Do control units use a special interface to measure RTD sensors, such as a wheatstone bridge or a dedicated IC designed for measuring RTDs ? Or do control units simply use a pull up resistor on an analogue input ?

I understand cost is an important factor in automotive, so thought control units would use the simplest and cheapest method to read RTD sensors.

The CCS connector states two temperatures thresholds:
  • 80 degrees C to derate the charging current
    90 degrees C to stop charging
I looked at using a pull up resistor with an analogue input, but the problem is there is a very small difference in resistance:
  • At 80 degrees C the resistance is 1308.97R
    At 90 degrees C the resistance is 1347.07R
Using a 1k pull up to 5V, this produces a voltage drop of:
  • At 80 degrees C the voltage drop is 2.835V
    At 90 degrees C the voltage drop is 2.870V
This is a difference of 35mV which is going to be difficult to measure in a noisy environment (EMI).

There would also need to be hysteresis on each of the two threshold measurements which would be an even smaller voltage.
User avatar
johu
Site Admin
Posts: 3549
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Contact:

Re: How do control units typically measure PT1000 (RTD) temperature sensors ?

Post by johu »

You could compare your sensor voltage to a 2.5V reference and then amplify the result. So basically v_out=(v_sens - 2.5) * 10

And you can have pretty low cutoff low pass filtering, like 1Hz to get rid of the noise
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
freeflyer
Posts: 2
Joined: Sun Dec 19, 2021 11:57 am

Re: How do control units typically measure PT1000 (RTD) temperature sensors ?

Post by freeflyer »

Thanks johu

I am using an off the shelf controller which has 0 to 5V analogue inputs, so cannot change the interface without designing and building additional hardware that sits outside the controller which I would like to avoid.

I am curious how these sensors are interfaced in automotive controllers seeing as the CCS type 2 connectors are common place and all use these PT1000 sensors.

Are they using a typical analogue input or do they have a special interface for PT1000 sensors ? Such as a wheatstone bridge or a fixed current source with a PGA (programmable gain amplifier) etc
User avatar
johu
Site Admin
Posts: 3549
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Contact:

Re: How do control units typically measure PT1000 (RTD) temperature sensors ?

Post by johu »

A very typical CCS choice. Use something that is really expensive and doesn't fit the purpose ;) Sorry for the rant.

https://www.adafruit.com/product/3648
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
User avatar
EV_Builder
Posts: 799
Joined: Tue Apr 28, 2020 3:50 pm
Location: The Netherlands

Re: How do control units typically measure PT1000 (RTD) temperature sensors ?

Post by EV_Builder »

It depends on the bits of your ADC? Is it 10 or higher?

I don't know if I would bother much with 80 or 90 degrees .
At 80 I would already stop charging I think. Especially if it rose rapidly.

Adding an cap will filter electronic noise and adding a mathematic model will make the conversion smarter .
Sample 4Hz and go with the 1Hz value etc

I would say more then 55 degrees is worth an attention to the user already. Cut off by 80?
Post Reply