TMP513 shunt voltage measurement offset voltage

Hi Suresh,

Thanks for posting on TI E2E forums. I am applications engineer from the sensing team and I will be helping you with this request. I will ask you a few questions to have a better understanding of your scenario. 

1. This 1.7mV offset is the constant when varying the input voltage range?

2. This 1.7mV is  Referred-to-input (RTI)?

3. Which gain are you using? I am assuming you are using PGA= +1 because according with your specs the maximum shunt voltage yoiu have is 10mV. Please confirm. 

4. How you have the 3.3V subregulator configured?

5. You have any schematic that you can share about your system? This will give me more clues on the issue. If you don't want to share in public forum you can email me directly. 

Best Regards,

Mayrim Verdejo

Texas Instruments Inc.

Applications Support | MHR Sensing Products

Email: mayrim.verdejo@ti.com

Hi Mayrim

Thanks for your quick reply, find the details below.

1. This 1.7mV offset is the constant when varying the input voltage range?

            The 1.7mV offset voltage reduces when the input bus voltage reduced.

2. This 1.7mV is  Referred-to-input (RTI)?

            The 1.7mV offset voltage reduces when the input bus voltage reduced.

3. Which gain are you using? I am assuming you are using PGA= +1 because according with your specs the maximum shunt voltage you have is 10mV. Please confirm. 

            With a gain of PGA= +1 and  PGA= +8, i am getting the same offset voltage.

4. How you have the 3.3V sub regulator configured?

           Sub regulator configuration :3, refer attached schematic

 5. You have any schematic that you can share about your system? This will give me more clues on the issue. If you don't want to share in public forum you can email me directly.

           Find the attached schematic.

And one more query is , the offset voltage measured with the same bus voltage will vary for all TMP513 IC's are constant or will it vary ?

Hi

I could not get good technical support from india TI technical support team, kindly improve the TI support in india.

Hi Suresh,

I am sorry to heard you are not receiving good support from our team from India. Please use the TI E2E forums for technical questions about your designs, we will try to reply in a timely manner.

Thanks for all answers. I received your email with the schematic and I will consult with some of my colleagues the issue to try to detect the issue. We will probably have to recreate this in the lab so please allow us some time for replies.

Also for question 1, what I wanted to ask is if this 1.7mV offset is the constant when varying the input CURRENT range. I mentioned voltage instead of current. You mentioned you vary from 0A to 2A. Is this offset constant or decrease/increase?

Best Regards,

Mayrim

Hi mayrim,

The 1.7mV offset is constant for the current of 0A to 2A.

Suresh,

Mayrim alerted me to your question and I would like to try to help.  I have a few questions about your setup, and if you're willing to modify your board a little, we can do some remote debug.

1. Why do you have R33 in place?  It seems to me it will just create a variable voltage drop depending on the current drawn by the TMP513.  This seems unnecessary and possibly problematic.

2. Why is C113 before R33 and not directly on U10, pin 2?  Standard design practice is to decouple at the pin; as close to the pin as possible.

3. A constant offset voltage could be any number of things.  Because the offset voltage changes with bus voltage, the first thing that comes to mind is that you have a resistive path to ground parallel to your load.  When you measure the current for your load externally for confirmation, do you do it on the load side of the shunt or the supply side?  Changing between them may give us some clues.

4. Another thing to try is to remove the filtering.  Short R30 and R31, and remove C28.  This will eliminate possible effects of the passive components in the filter.

5. Another thing to try after the filter is removed is to remove the shunt resistor and short VIN+ and VIN- together.  I believe this should remove the effects of input bias currents and just leave you with the offset voltage measurement.  Remove the load and just adjust the bus voltage to see if there is any change.  You can also remove the load and Vbus for a totally clean measurement, if possible, or leave the Vbus connected and sweep it to see if there is still an effect on the offset voltage present in the measurement.

Let me know how flexible your setup is and if you are able to test these suggestions, and what the results of the debug experiments are.

I hope to solve this issue quickly!

Jason Bridgmon

Hi Jason

Thanks for your reply

1. R33 used for current limiting purpose, Voltage drop in the resistor R33 is minimum.

2. C112  decoupling provided,

3. current measured in  the input side.

i will do the changes recommanded by you in the module and try to give you the results within two days