LDC1000: Step function response over temperature

Hi Richard,

Can you clarify how long the data is stuck at a single value? Just in case you haven't seen it, we do have an application note about temperature compensation with the LDC10xx devices: http://www.ti.com/lit/an/snaa212a/snaa212a.pdf

We do have a drop-in replacement for the LDC1000, which is the LDC1001. The key difference is that the LDC1001 cannot drive an external crystal and needs an external clock input for L sensing. This does not affect Rp sensing at all, so you do not need to change your design. However, please note that we have many newer LDC devices in our portfolio that may be suitable for your application. These newer devices are higher resolution and lower power. I recommend looking through the LDC Device Selection Guide if you can still make changes to the part selection in your design.

Best Regards,

Hi Kristin,

Attached is an Excel spreadsheet which shows the response I describe. The x-axis is temperature in degC, recorded over a long duration while the equipment warms up. The y-axis is the proximity data value. There are 32 sensors connected to one LDC1000 by analogue mutiplexers. All sensor positions are empty, i.e. no targets are present. 

The AppNote doesn't explain the plateaus in the response.

Changing to LDC1001 has the least impact on the existing design. What’s its YTEOL ?

Regards

Richard

/cfs-file/__key/communityserver-discussions-components-files/1023/variation-avec-temperature-TI.xlsx

Hi Richard,

What are your RP_MIN and RP_MAX values? I'd like to look at the data in terms of actual resistance compared to the LSB resistance value.

If you look at the Rs across temperature graph in the application note I shared, our data points are all ~20 degrees C apart. You have a much finer scale on your graph, and I wonder if it would look similar to our graph if you looked at a larger temperature step. 

Do you see similar steps when you measure the Rp response versus target distance?

We have no plans to EOL the LDC1001. You can find our EOL policy here: http://www.ti.com/support-quality/quality-policies-procedures/product-life-cycle.html

Best Regards,

Hi Kristin,

Rp response v distance.xlsx

Apologies for my delayed response, I was side tracked onto another project. 

RP_MIN=19h , RP_MAX=3Fh

Yes, the Rp response versus target distance has a similar response with steps.

I’ve done an experiment to show this. Please see the attached spreadsheet, in particular the second chart. Basically column A is distance, column N is proximity data. As a target is moved away from the sensor there are apparent steps in the proximity data values, at the same absolute values as the previous test over temperature. If proximity data is the output from an ADC why does it behave in this way ? (data sheet says « This range of RP can be viewed as the maximum input range of an ADC. ») Does the ADC have missing codes ?

Best Regards

Richard Kelly

Hi,

I can do some experiments to determine whether changing the value of Rpmax fixes the issue. If the steps in the response and the non-monotonicities are no longer present then is that result valid for all LDC1000s we may receive in production, over temperature etc ? 

Is there an erratum you can share so that I can actually define the problem I face?

As my previous posts I am intending to replace the end-of-life LDC1000 with the LDC1001. Does LDC1001 have the same problem ?

Best Regards

Richard

Hi Kristin,

Thanks for your assistance.

Regards