AMC1311-Q1: AMC1311-Q1

FATMA BAY

Part Number: AMC1311-Q1
Other Parts Discussed in Thread: ISO224, TLC555, AMC1311

Hello team,

I would like to use AMC1311-Q1 in my design to measure temperature isolated. NTC has a very low resistance in high temperature like 160 ohm @ 150C and 125kohm @-40C. When I sclaed the input from 0V to 2V, I have about 3mv input @150C and it doesn't make sense to use it as that due to it's hard to tell it from any noise.

Do you have any suggestion regarding this issue?

Thank you so much!

Regards,

Fatma

over 3 years ago

0 Alexander Smith over 3 years ago

TI__Mastermind 47595 points

Hi Fatma,

Thanks for your question! I have a few ideas:

One option would be to use a device with a larger input voltage range such as the ISO224, however this device is not automotive qualified.

Another option is to perform an offset calibration which would help differentiate the 3mV input signal from the system noise.

Last but not least, you could consider adding some buffer voltage above the noise floor to bias your input signal. For example, you could sweep from 500mV to 2V instead of 0 to 2V. Depending on the magnitude of the bias voltage, this would reduce your full-scale input range but your signal would no longer be hidden by the noise floor.

0 kai klaas69 over 3 years ago

Guru 140200 points

Hi Fatma,

which NTC do you use?

As Alexander already mentioned you would need to decrease the range of change of NTC resistance to a suited level. This is usually done by putting a precision resistor in parallel to the NTC as discussed here, e.g.:

Another option is to take the TLC555 and use the NTC as frequency determining component. Here again the paralleling with a resistor will help. Then add a simple optocoupler to provide galvanic isolation :-)

Kai

0 FATMA BAY over 3 years ago in reply to Alexander Smith

Prodigy 50 points

Hi Alex,

Thank you for quick answer.

I will consider first option that is not automotive qualified one.

When I gave an offset to my input signal, for lower temperature I have also an offset. So, the input range would have been exceeded. Is it sufficient?

In the last option, I can lost the best part of measurement range like e.g. upwards of 80°C.

I can separate the measurement range with 2 isolator opamp but it means double quantity and 2 signal on the output.

Regards,

Fatma

0 FATMA BAY over 3 years ago in reply to kai klaas69

Prodigy 50 points

Hi Kai,

It's an internal NTC of a power module. The specs are below.

I will consider that application, then I can give a feedback.

Thank you for your help.

Regards,

Fatma

0 kai klaas69 over 3 years ago in reply to FATMA BAY

Guru 140200 points

Hi Fatma,

the specs didn't come through. Please post again and use the "Insert File" button (paperclip) in the header of textbox :-)

Kai

0 Alexander Smith over 3 years ago in reply to FATMA BAY

TI__Mastermind 47595 points

Hi Fatma,

I would like to confirm what you are seeing.

For the offset calibration, it would be applied in software. Such that when the NTC is disconnected, you record the value measured by the AMC1311. Then you apply the inverse of this value to your measurements once the NTC is connected. What value of voltage are you seeing when the input range is exceeded?

The input voltage range of the device is specified to be -0.1V to 2V, with a clipping voltage of 2.516V. You can safely measure input voltage up to 90% of the clipping voltage without significant additional errors, 2.26V.

For the last option, you would have to attenuate the signal to match the new input voltage range of 2V - the bias voltage. Is the output voltage from the NTC 0-2V or do you have a gain/attenuation opamp before the AMC1311?

It looks like the datasheet for the NTC didn't post correctly, can you please re-post?