• State TI Thinks Resolved
  • Locked Locked
  • Replies 19 replies
  • Answers 9 answers
  • Subscribers 30 subscribers
  • Views 1377 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LMP91000: Glucose Meter Design Using LMP91000

Nikunj Solanki
Prodigy 60 points
Part Number: LMP91000

Tool/software:

Hello TI,

I am designing a glucose meter and using the LMP91000 as the main AFE (Analog Front End). I need help with the schematic and component selection. The meter will measure multiple parameters, including:

  • Glucose
  • Hemoglobin
  • Creatinine
  • Cholesterol
  • Uric acid

1. Schematic Design Request

Could you please provide a reference schematic for using the LMP91000 in a glucose meter? I need guidance on the recommended circuit design, including connections for the working, counter, and reference electrodes.

2. Operational Amplifier Recommendation

In case I need an alternative to the LMP91000, can you suggest suitable operational amplifiers that can be used for biosensing applications?

3. Strip Detection Mechanism

What is the best way to detect when a test strip is inserted using a microcontroller? I am considering the following methods:

  • Electrical contact detection
  • Optical sensor detection
  • Mechanical switch
  • Capacitive/resistive sensing

Could you provide recommendations or best practices for strip detection?

4. Additional Suggestions

Any other important considerations or best practices from your experience in glucose meter design would be highly appreciated.

Looking forward to your expert guidance.

Thank you!

  • 0
    TI__Genius 12235 points

    Hi Nikunj,

    In the LMP91000 Datasheet we provide diagrams showing how to connect the LMP91000 with a 2-lead or 3-lead electrochemical cell. The components and internal register settings are going to vary based on what electrochemical cell you choose, and what biasing it requires. We also have a LMP91000 EVM that contains an example circuit for the AFE, but is not specific to glucose sensing or any particular electrochemical cell. You should start here to understand the AFE, then pick an electrochemical cell to use, and that cell's datasheet will inform how to configure the LMP91000.

    We also offer the LMP91200, which is designed for sensing pH electrodes. 

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Hi,

    Thank you for your response.

    We are using a 3-lead electrochemical cell (Strip) with working, reference, and counter electrodes. Could you recommend a suitable biasing current for our setup? Additionally, if possible, could you provide guidance or an example circuit for glucose sensing using the LMP91000?

    Looking forward to your insights.

    Best regards,
    Nikunj

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Nikunj,

    Have you chosen a electrochemical cell? What specifically are you trying to sense? The datasheet for the electrochemical cell you choose will tell you what bias voltage it requires, and from there you can set the registers of the LMP91000 accordingly. 

    I have not seen glucose monitoring examples, and unfortunately all the guidance I can provide is how to use the LMP91000 once you have chosen an electrochemical cell to pair it with.

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Hi,

    Thank you for your response. We are using a strip sensor for measuring glucose, hemoglobin, cholesterol, uric acid, and creatinine from a drop of blood placed on the strip. Please let us know if there are any specific recommendations for configuring the LMP91000 with this type of sensor.

    Best regards

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Hi Nikunj,

    Does your strip sensor have 2 or 3 "leads" where it could be biased by the LMP91000 and then have current output read by the AFE? If you could share the exact test strip and its datasheet, that would be helpful.

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Hi,

    Yes, we have a 3-lead strip sensor. The current is read by the AFE. The three leads are:

    • Counter Electrode (CE)
    • Working Electrode (WE)
    • Reference Electrode (RE)

    Please let me know if you need further details.

    Best regards

    Nikunj

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Nikunj,

    Please share the strip sensor datasheet so we can discuss the necessary bias voltage and other LMP91000 settings.

    For general LMP91000, please take a look at the schematic for the LMP91000EVM. There isnt much more support I can offer other than that.

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Hi,

    I am attaching the strip sensor datasheet for your reference. Let me know your thoughts so we can discuss the necessary bias voltage and other LMP91000 settings.

    Regards

    Nikunj

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Hi Nikunj,

    Does the datasheet have any information on the necessary bias voltage to the working electrode? If not, I would start by using the LMP91000EVM so that you have some known good external connections, and try using the default bias and internal zero settings to see what the strip sensor outputs.

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Hello,
    The strip sensor datasheet does not mention the necessary bias voltage. What voltage should I connect at VDD and VREF?
    Thanks and Regards
    Nikunj

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Hi Nikunj,

    Without any information on the bias requirements, you may just be guessing what to use. I would reach out to the strip sensor manufacturer and see if they can provide more details.

    In the meantime, I would try to following settings just to get started. i am choosing these since I have observed them as relatively common selections. Please try these if you have no other information, and see what the LMP91000 does.

    • LOCK register to 0x00, so that the registers can be written to.
    • Set the TIA feedback resistance to 14KOhms
    • Ser R_load to 100Ohms
    • leave the reference voltage source to internal
    • leave the internal zero to 50%
    • leave the bias polarity as negative
    • use a 20% bias percentage
    • leave the shorting FET disabled
    • set the operating mode to 3-lead amperometric cell.

    we can see what these settings do. You can try tweaking the settings from there to get your desired results.

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Dear Alex,


    I sincerely appreciate your help.

    I must now get a response from the strip manufacturer, and they say that the bias voltage for the strip is 40 mV. Therefore, does this setting work with this bias voltage, or do I need to adjust any other settings?

    Thanks and Regards
    Nikunj

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Hi Nikunj,

    I would adjust to using an external reference voltage. I would use a 2V reference voltage on VREF, then set the bias percentage to 2%. This will get a bias of 40mV. If that actually ends up being a 20mV output because of the 50% internal zero setting, then use the 4% bias voltage setting to make the CE output 40mV.

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Dear Alex,

    Thank you for your response.
    What other settings do I need to adjust if I use a fixed external 2.5V as VREF? Should the other settings remain the same as given in the previous reply?

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Nikunj,

    In that case I would adjust the bias voltage percentage to get as close as possible to 40mV, and I would go slightly over than slightly under. So using the 2% bias setting would get you 50mV, try that.

    Regards

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Dear Alex,

    Thank you so much for your support.

    I have one common voltage, which is 3V3. Can I use it for VREF? If so, what settings should I apply?

    Additionally, as we measure different parameters by placing a blood drop on the strip, the bias voltage of the strip for hemoglobin and creatinine is 300 mV. What settings should I use to achieve this voltage?

    Could you also explain the calculation for selecting the appropriate bias voltage percentage?

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Hi Nikunj,

    Yes you can use 3.3V as the Vref as well as VDD. Alternatively, you can use the internal 2.5V reference. The bias voltage percentage calculation is very simple, I am just multiplying the chosen reference voltage by the difference percentages to get close to the necessary voltage. For example, when you mentioned 2.5V internal reference, I multiplied that by 2% to get 50mV, which is closest to the 40mV requirement than any other option.

    Thanks

    -Alex Thompson

  • 0 in reply to Alex Thompson
    Prodigy 60 points

    Hi Alex,

    Thank you so much for your support.

    We can apply an external 3.3V at VDD and Vref. The biasing voltage requirements are as follows:

    • For glucose, uric acid, and cholesterol strip sensors: 40 mV
    • For hemoglobin and creatinine strip sensors: 300 mV

    Could you please guide us in selecting the appropriate bias voltage percentage for these cases?

    1. For 40mV with VDD and Vref at 3.3V

      • What would be the correct bias voltage percentage selection and other relevant settings?

    2. For 300mV with VDD and Vref at 3.3V

      • What would be the appropriate bias voltage percentage selection and other settings?

    Looking forward to your insights.

    Best regards,
    Nikunj

  • 0 in reply to Nikunj Solanki
    TI__Genius 12235 points

    Nikung,

    For the other settings, as I've said before, start with the defaults or what I listed above, and experiment from there. What gain is desired or how to set the right output settings is going to depend on your system and what is receiving the output of the LMP91000. I can't do that for you.

    Since we are dealing with positive bias voltages, set the bias polarity bit in register 0x11 to 1, so that the bias polarity is positive.

    You won't be able to get exactly 40mV and 300mV with a 3.3V VREF, but that should be OK. For 3.3V, set the bias to 2% to 66mV to get close to 40mV, and set the bias to 10% to get 330mV, which is closest to 300mV.

    Regards

    -Alex Thompson