LMP91000: Vout is not as expected

Hi Ikon,

Applying a 1% bias to a Vref of 1.5V will apply a 15mV bias voltage to the sensor between the WE and RE terminal. If the sensor requires a 15 mV bias, then this setting is correct.

From the datasheet of the sensor, do you know what load and bias are required for the sensor to operate? Could you share what kind of sensor this is?

Could you please provide the settings that you used, and an output value that result from those settings? For example, when you set TIA_GAIN = 350K and R_Load = 10 ohms etc..., vout is 50 mV.

I would highly suggest that you take a look at the webbench design tool for the LMP91000, located on the right hand side of the website

This tool will help to find values required for your expected output.

Best Regards,

Dan

Hi Dan,

Customer cannot set the voltage between CE and RE, their target is to get between 0 to 1 volt.

Customer is using voltage base sensor.

Thanks.

Hi Dan,
In addition, the highest output I get from testing is about 10mV with TIAGN set default 2.75K, Rload 100R.
As I understanding from datasheet, the bias voltage setting seem doesn't take 100% input from Vref. Accordingly to the block diagram, in the middle of input to output is having a voltage divider, what is the divider function as?
Thanks.

Hi Ikon,

The sensor should have a datasheet that will give the required values for bias voltage and the ohmic load required. These are required to ensure proper sensor operation. Is this a two lead or three lead sensor?

Did you mean they want 0 to 1V between RE and WE (this is the sensor bias voltage)? When the sensor detects its target gas, CE will sink/source current through WE and the TIA (providing a voltage out). RE provides the reference voltage and will not sink/source any current.

I have found that this app note is helpful in explaining the full functionality of the potentiostat in regard to gas sensors.

The voltage divider, seen in the below picture, will provide a fixed offset voltage to the TIA.

This is used to help maximize the response of the LMP91000 with respect to the ADC that will be reading VOUT (example 0.6V to 1.26V ADC sampling window). This can be controlled by setting the INT_Z register. You should see a direct change in output voltage (VOUT) when changing this value.

Hope that helps.

Best Regards,

Dan

Hi Dan,

Thank you very much for your reply. Actually the sensor that customer is going to use is a 3-lead electro-chemical sensor. They want to read the sensor data using a pulse voltametry technique in which the step voltages are applied to the sensor and its current is read. The voltages applied to the sensor are varied from 0 to 1 V with a 15mV step increment (please refer to the attached image for the voltage waveform applied to the sensor. 

To do so, they have designed a circuit (as shown in the attachment) in which the Vref of LMP is sourced from a DAC. With the combination of DAC output and LMP voltage divider circuit we are creating the desired voltages for the sensor.

Before they connect the sensor to LMP, they wanted to make sure that the circuit is able to produce the desired voltages between WE and RE. For this purpose, they have connected two 300 ohm resistors; one between WE/RE and one between CE/RE. However, they were not able to produce the voltages wanted.

For example, to apply a 100mV voltage to the sensor, they used a 5V reference voltage and the bias setting of (2% of Vref). However, the voltage read at WE is only 10mV.

Question: Understanding about LMP operation is correct? in other words, is it possible to create various voltages at WE with the combination of Vref and bias setting?" 

 

Thanks.

Hi Dan,

Thank you very much for sending the additional info and the diagram.

Just for clarification, would you please confirm that the WE & RE voltages are calculated as below?

- WE voltage is equal to Internal Zero which is the output of Vref Divider : (V_WE = y * Vref; in which y=20%, 50%, 67%,100%)

- RE voltage is equal to [V_RE = (1+x) * V_WE] in which x= 0%, +1%,+2%, ...+24%

Thanks.

Hi Dan,

After I have done several testing with different setting, I couldn't get the expected figure.

 

Vref = 2.7V

RE is probed to positive terminal voltmeter.

WE is probed to Negative terminal voltmeter.

 

According to your explanation, Vref = 2.7V, INT_Z = 50%, BIAS = 1% and the output from RE to WE will expected to get 13.5mV. Unfortunately, I only get 0.3mV. 

 

The following is the LMP91000 setting and output:

 

TIACN(0x10) - 0b00000111

TIA_GAIN = 2.7K

RLOAD    = 100R

 

MODECN(0x12) - 0b00000011

FET_SHORT = DISABLE

OP_MODE   = 3-lead amperometric cell

 

REFCN(0x11) - 0b10010001

INT_Z = 20%

1% BIAS

WE,RE = 0V

REFCN(0x11) - 0b10010010

INT_Z = 20%

2% BIAS

WE,RE = 0V

REFCN(0x11) - 0b10010011

INT_Z = 20%

4% BIAS

WE,RE = 0V

REFCN(0x11) - 0b10010100

INT_Z = 20%

6% BIAS

WE,RE = 0V   

 

TIACN(0x10) - 0b00000111

TIA_GAIN = 2.7K

RLOAD    = 100R

 

MODECN(0x12) - 0b00000011

FET_SHORT = DISABLE

OP_MODE   = 3-lead amperometric cell

 

REFCN(0x11) - 0b10110001

INT_Z = 50%

1% BIAS

WE,RE = 0.3mV

REFCN(0x11) - 0b10110010

INT_Z = 50%

2% BIAS

WE,RE = 0.3mV

REFCN(0x11) - 0b10110011

INT_Z = 50%

4% BIAS

WE,RE = 0.3mV

REFCN(0x11) - 0b10110100

INT_Z = 50%

6% BIAS

WE,RE = 0.3mV  

 

MODECN(0x12) - 0b00000011

FET_SHORT = DISABLE

OP_MODE   = 3-lead amperometric cell

 

REFCN(0x11) - 0b11010001

INT_Z = 67%

1% BIAS

WE,RE = 0.6mV

REFCN(0x11) - 0b11010010

INT_Z = 67%

2% BIAS

WE,RE = 0.6mV

REFCN(0x11) - 0b11010011

INT_Z = 67%

4% BIAS

WE,RE = 0.6mV

REFCN(0x11) - 0b11010100

INT_Z = 67%

6% BIAS

WE,RE = 0.6mV  

 

MODECN(0x12) - 0b00000011

FET_SHORT = DISABLE

OP_MODE   = 3-lead amperometric cell

 

REFCN(0x11) - 0b11110001

INT_Z = Internal zero circuitry bypassed (only in O2 ground referred measurement)

1% BIAS

WE,RE = 0.9mV

REFCN(0x11) - 0b11110010

INT_Z = Internal zero circuitry bypassed (only in O2 ground referred measurement)

2% BIAS

WE,RE = 0.9mV

REFCN(0x11) - 0b11110011

INT_Z = Internal zero circuitry bypassed (only in O2 ground referred measurement)

4% BIAS

WE,RE = 0.9mV

REFCN(0x11) - 0b11110100

INT_Z = Internal zero circuitry bypassed (only in O2 ground referred measurement)

6% BIAS

WE,RE = 0.9mV  

 

Is there any load need to be added before probe to the RE, WE?

Thanks.

 

Hi Ikon,

Was the sensor connected when taking these measurements? Have you verified that the internal zero voltage is present (Measuring RE or WE with respect to ground)?

When a bias voltage is present in your measurements, the bias voltage is not changing with respect to a change in the bias settings. I am looking into this further.

What is the impedance of your probe that you are measuring with?

Best Regards,

Dan