This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LDC1101 SNOSD01A data sheet - questions (June-2015 Version)

Other Parts Discussed in Thread: LDC1101, LDC1612

Dear TI-Team!

I've some experience with LDC1000 and have ordered the new LDC1101 now. Studying the datasheet, the following questions came up:

ad 1  Features:  What is meant with "Remote Sensor Placement Isolating the LDC from Harsh Environments". Is this a new "Feature" in comparison to LDC1000?

ad 3 Description: "Inductive sensing Technology enables precise measurement of....":  If the same tank-configuration is used and the same external clock frequency is used: Is the LDC1000 accurracy for L-measurment (24bit) the same to LDC1101 accurracy in LHR mode?  Or does the new LDC1101 incorporate new design features that improve accurracy and resolution? And same question for RP-resolution and accurracy...

ad 8.4.1 Measurement Modes: "Both measurment subsystems can convert simultaneously but at different sample intervals - the completion... is asynchronous...": Is it possible to gather data from both subsystems querying RPL and then LHR at "High Speed" without reconfiguring the chip to the other mode, so that no sleep mode cycle is needed?

What can be done, to get RP with 16BIT AND fres with 24 Bit as with LDC1000?

ad 8.4.4 Reference Count Setting: Are the given values 8,6µs to 87,38ms correct?

The formula for Measurment Time (tconv)=(RCOUNT*16+55)/fclkin. RCOUNT ranges from 512 to 65535 (right?). fclkin max is 16MHz. So tconvmin=515µs and tconvmax=65,5ms...

What is wrong here?

8.6.24 Register LHR_RCOUNT_MSB...: Table 26: Description: Valid range: 2<= RCOUNT[15:8]<=65535. Is this correct?

Shouldn't it say 2<=RCOUNT[15:8]<=255 or so?

8.6.25 and 8.6.26: the LHR_OFFSET_LSB and LHR_OFFSET_MSB  Fields: The LHR_OFFSET Register(s) should be set by the user , but there is no description anywhere, what the purpose of these registers is. They are only used in formula 11, but what is their purpose and how should they be set?

8.6.28 Table 30: LHR_DATA_LSB Field Descriptions: Is the formula given for fsensor correct? If yes, how is it related to formula 11?

9.1.9 LHR Sample Rate Configuration with RCOUNT:

What is "really" meant with ENOB (Effecibe number of bits?). It is introduced here without giving a context: is it introduced to be able to calculate the effective resolution with ENOB?  

If yes, how can I calculate it?

9.2.3: Application Courves: how can I calculate such courves for my specific application (simple distance measurment)

I thought to understand that LHR Resulution is: 1/(RCOUNT*16) at Target Distance /Sensor Diameter =1. This is true for the black courve. But for the red courve it is not. Can you give the calculus for that? What is the calculus for values <1?  (i took one example of your application courve in 9.2.3).

Looking forward to your reply!

Franz

  • What is meant with "Remote Sensor Placement Isolating the LDC from Harsh Environments". Is this a new "Feature" in comparison to LDC1000?

    LDC devices, including the LDC1000, can use sensors which are separated from the device by many cm; this is a feature when compared to many technologies in which the sensor is inside the IC or require the sensing element to be adjacent to the IC. For example, an inductive sensor can be placed in an environment which exceeds the operating range of the IC, which can be remotely located in a compliant environment.

    ad 3 Description: "Inductive sensing Technology enables precise measurement of....":  If the same tank-configuration is used and the same external clock frequency is used: Is the LDC1000 accurracy for L-measurment (24bit) the same to LDC1101 accurracy in LHR mode?  Or does the new LDC1101 incorporate new design features that improve accurracy and resolution? And same question for RP-resolution and accurracy...

    The LDC1000 24 bit L measurement is equivalent to the LDC1101 16 bit L measurement, only the 16 bit measurement from the LDC1101 is effectively higher resolution because the LDC1101 can use a 16MHz reference frequency, while the LDC1000 is limited to only 8MHz.

    The LDC1101 LHR measurement is equivalent to a single channel LDC1612.

    ad 8.4.1 Measurement Modes: "Both measurment subsystems can convert simultaneously but at different sample intervals - the completion... is asynchronous...": Is it possible to gather data from both subsystems querying RPL and then LHR at "High Speed" without reconfiguring the chip to the other mode, so that no sleep mode cycle is needed?

    Yes, all 3 sets of data (RP, L, LHR) can be retrieved while the LDC1101 is converting. There is no need to enter sleep mode or reconfigure the device. 

    What can be done, to get RP with 16BIT AND fres with 24 Bit as with LDC1000?

    The LDC1000 can only provide a 24bit resolution measurement with a sensor frequency of 5kHz (not 5MHz). The LDC1101 has 2 L measurement paths – a 16 bit L measurement which is similar to the LDC1000 implementation and provides best resolution at the lowest sensor frequency, and the LDC1101 has a LHR mode, which can provide a higher resolution measurement when configured with a high reference count.

    ad 8.4.4 Reference Count Setting: Are the given values 8,6µs to 87,38ms correct?

    Actually, the time interval is 5.44μs to 65.5ms; this will corrected in the datasheet soon.

    The formula for Measurment Time (tconv)=(RCOUNT*16+55)/fclkin. RCOUNT ranges from 512 to 65535 (right?). fclkin max is 16MHz. So tconvmin=515µs and tconvmax=65,5ms...

    What is wrong here?

    RCOUNT[15:0] can be set to a minimum value of 2; although the effective measurement resolution will be quite low.

    8.6.24 Register LHR_RCOUNT_MSB...: Table 26: Description: Valid range: 2<= RCOUNT[15:8]<=65535. Is this correct?

    This should be 2<= RCOUNT[15:0]<=65535

    Shouldn't it say 2<=RCOUNT[15:8]<=255 or so?

    8.6.25 and 8.6.26: the LHR_OFFSET_LSB and LHR_OFFSET_MSB  Fields: The LHR_OFFSET Register(s) should be set by the user , but there is no description anywhere, what the purpose of these registers is. They are only used in formula 11, but what is their purpose and how should they be set?

    There is no need to ever set the OFFSET; the functionality is similar to the LDC131x devices; however the LDC1101 does not have a gain function which requires the OFFSET.

    8.6.28 Table 30: LHR_DATA_LSB Field Descriptions: Is the formula given for fsensor correct? If yes, how is it related to formula 11?

    Table 30 should be f­ = fCLKIN ˣ 2SENSOR_DIV ˣ LHR_DATA ÷ 224, the assumption is OFFSET =0.

    9.1.9 LHR Sample Rate Configuration with RCOUNT:

    What is "really" meant with ENOB (Effecibe number of bits?). It is introduced here without giving a context: is it introduced to be able to calculate the effective resolution with ENOB?  

    If yes, how can I calculate it?

    The ENOB is determined by the effective resolution of the configuration for the LDC and is corresponds to the output code change.

    9.2.3: Application Courves: how can I calculate such courves for my specific application (simple distance measurment).

    These curves were calculated from measured output codes using a precision mechanical setup with an aluminum target which was precisely positioned. 100 measurements were made at each position, and the slope of the output code was divided by the standard deviation at the given position to calculate the resolution.

    I thought to understand that LHR Resulution is: 1/(RCOUNT*16) at Target Distance /Sensor Diameter =1. This is true for the black courve. But for the red courve it is not. Can you give the calculus for that? What is the calculus for values <1?  (i took one example of your application courve in 9.2.3).

    The LHR resolution is actually more accurately estimated as 1/(RCOUNT*32); however this estimate underestimates the actual resolution by 2-4 bits.