Power Monitoring IC Required

Hari T O0

Other Parts Discussed in Thread: INA226, INA3221, SN74CBTLV1G125, TS5A3159, INA219

Hi ,

I need to measure the current consumed by my DSP in it's different operating modes.It contains two in built LDO which will receive the input from outside and its out put will

be connected back to DSP.We will be placing current sense resistors in the output path .The current sense resistor output will be connected to a power monitoring IC.Please see the attached

block diagram.The out put of the power monitor will be connected to Microcontroller.

May I know TI have any IC which supports this intent

block diagram

over 6 years ago

0 Manuel Chavez over 6 years ago

TI__Genius 17100 points

Hi Hari,

TI power monitoring ICs can be found on TI's current sensing web page here: www.ti.com/.../products.html

The INA226 is a versatile device, but I can help you find a more tailored solution based on operating conditions and device package you are looking for - which is how devices are categorized on TI.com as linked above.

How may I help you select a device?

Thanks,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

HI Manuel,
We are in concept phase now ,will give you full updates soon.May be within 1month.
Regards
Hari

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

HI Manuel,
I need to monitor two currents at a time.INA226 can monitor only one current rt?.Can you suggest an IC which can monitor two currents at a time.
Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hari,

If you do not want to use two independent power monitors, the INA3221 can monitor up to three currents. If choosing this device ensure to configure the amount of channels being used.

Best,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,
Thank you very much,will contact you at the design phase:)
Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

You're welcome, Hari; I will be available

Best,
Manuel Chavez

0 Manuel Chavez over 6 years ago in reply to Manuel Chavez

TI__Genius 17100 points

Hi Hari,

It has been a couple weeks since our last post, so I will close this thread. Please reply or create a new thread when you reach the design phase.

Thank you for choosing TI parts!
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel
Is it possilble to measure 1uA to 100mA using the above mentioned IC's
Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hi Hari,

With the device mentioned above it could be possible to measure that current range, however there are a few requirements to keep in mind for it to work:

The setup would require all three channels to sense current from each line or LDO (so 2 INA3221s for the application above)
A switch network would need to be implemented to toggle the setup either manually or automatically
Calibration would be required

Let me know what you are considering.

Thank you for your time,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,

How to design an automatic switching network. Please provide me with some ideas
May I know you have any reference designs.

Regards
hari

0 Hari T O0 over 6 years ago in reply to Hari T O0

Genius 3820 points

Power Monitor IC Comparison.xlsxHi Manuel,

I am doing a comparison of current measurement methods using power monitor IC vs Conventional current sense methods

I am able to see one advantage of using current power monitor IC is ADC is present inside.

If you find any additional advantages than this please let me know

Please find the attached excel.

Regards

hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hari,

Automatic switching circuits will be complex due to losses from additional components and switching. Comparators or transistors can be used here and will need to be low leakage parts if measuring with a current shunt monitor.

I added some notes to your Excel document which you can see attached. Below is also a link to a TI Precision Design to measure µA-mA of current.

http://www.ti.com/lit/ug/tidu033/tidu033.pdf

I hope this is helpful!

Thank you,
Manuel Chavez

Copy of Power Monitor IC Comparison.xlsx

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,

Thank you very much.
Let me study the reference design provided by you,will contact you if I need any clarification

Regards
Hari

0 Hari T O0 over 6 years ago in reply to Hari T O0

Genius 3820 points

Hi Manuel,

For INA3221,ADC refrence voltage is 168mV ,correct?

Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hi Hari,

The internal ADC reference voltage is ~1.2V. This is significantly higher than the shunt voltage input because it is boosted up for the delta sigma converter.

Best,

Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,
May I know how that value is obtained.I am not able to see the REF voltage value in the data sheet.
Regards
hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hari,

This value is internally derived for some devices and not included on the datasheet. What is the specific reason you would like to know this value?

Thank you,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,
Thank you very much.
Some colleagues where telling ADC reference is maximum voltage allowed across the shunt.
So I am confused If you can help me how you got 1.2V that will help me make them understand maximum voltage across shunt is different from ADC reference.

Regards
Hari

0 Hari T O0 over 6 years ago in reply to Hari T O0

Genius 3820 points

Hi Manuel

We have almost finalized our specs

Minimum current to be measured=.2mA
Maximum Current to be measured=100mA
Rsense=.1Ohm
Minimum Current Step=0.01mA

I need one more clarification
My required step is 10uA so ADC and sense resistor is .1Ohm,so ADC should detect 10uA*.1OhmVolts..correct me if I am wrong.
May I know is this possible with INA3221.
Some time we will increase the sense resistor value to 5Ohm

Do we need pre amplifier before Power monitor IC

Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hello Hari,

The reference voltage of an ADC is separate from its input pins. This voltage, usually externally set or derived from Vs, is a constant that input signals can be compared to for accurate and precise conversions.

The maximum measurable voltage across the shunt is given on the INA3221 datasheet as Vshunt, ranging from -163.8mV to +163.8mV.

Regarding the system specs above, voltage step calculation for 10uA increments is correct, but the ADC’s LSB step size is 40µV – too high to sense 10µA x 0.1Ω. For this step size to be accurately converted, Rshunt would have to be 4Ω minimum. However on the 100mA end of the expected current range, shunt voltage would be too high for the device (400mV > 163.8mV), which means increasing Rshunt to 5Ω would be excessive as well. An optimal Rshunt value is 1.6Ω, which will yield a resolution of 25µA steps.

Tradeoffs should be considered as your design is finalized and will come down to:

Smaller current range for desired resolution
Desired current range for less resolution

No preamp is needed for our current shunt monitoring devices, so they are ready to use!

Best,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,

Thank you very much.
INA3221 data-sheet contains no information about ADC ENOB.
May I know will we get full 13 bit resolution from ADC

Regards
Hari

0 Hari T O0 over 6 years ago in reply to Hari T O0

Genius 3820 points

Hi Manuel,

I need one more clarification.
The maximum measurable voltage across the shunt is given on the INA3221 datasheet as Vshunt, ranging from -163.8mV to +163.8mV.
So the maximum voltage allowed across shunt is 327.6mV Correct.
Is it possible to drop 327.6mV across shunt

Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hari,

The effective number of bits for an ADC is dependent on noise and its ratio to the LSB. For the INA3221 conversion time will be crucial to determining your noise level. Different conversion times vs. noise are shown in Figure 25 of the device’s datasheet and conversion time configurations are listed on Table 5. For the default conversion time and above, ENOB is basically 13.

It is not possible to measure +327.6mV across the shunt resistor because the full scale range of the ADC inputs is 163.8mV.

Thank you,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Current Monitor Design Calculations.xlsx Calculation Explanation.pdfHi Manuel,

Thank you very much. Attaching my design calculations and it's explanation. Kindly go through it.

May I know switching between multiple resistors as shown in the below link will work well with INA3221

Regards

Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hi Hari,

The system calculations and your explanation are clear. I agree with how the numbers are derived and they provide a dynamic understanding of the parameters you are designing. Thank you for the documents.

Regarding the switching sensing system from the PDF above, this should be possible with the INA3221’s critical pin in partnership with an ISO (TI isolation) device and switches like the TS5A3159 and SN74CBTLV1G125. Losses may factor into measurement accuracy, but it is worth investigating. Below is a hand-drawn schematic of what a system like this may look like - let me know how I may be of more help.

Thank you for your time,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,
Thank you for your time and consideration.
May I know why isolation is required.
Regards
Hari

0 Hari T O0 over 6 years ago in reply to Hari T O0

Genius 3820 points

Hi Manuel,
If my voltage at the i/p of INA3221(Rsense*Isense) is less than 40uV.can I use a preamplifier .
Is it allowed to use pre amlifiers

Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hi Hari,

Isolation is not a requirement for the circuit, but can be useful for protecting its components.

What do you have in mind as far as the pre-amplifier?

Have a good day,
Manuel Chavez

0 Manuel Chavez over 6 years ago in reply to Manuel Chavez

TI__Genius 17100 points

Hari,

A pre-amplifier is not necessary for this circuit to be optimal, but if you are considering a more complex system, ADCs are capable of sensing voltages although some require a signal buffer to drive inputs. TI has low sampling frequency ADCs to choose from depending on additional requirements.

If you would like to continue the conversation, please reply or open a new thread if this one is locked!

Thank you for your time,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,

Definitely I need to continue with this conversation.Past 3-4 days I was busy with some other work.
Monday onwards I will come with my queries .

Regards
Hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hari,

Okay great! I will be out of the office next week so feel free to create a new thread.

Enjoy your weekend,
Manuel Chavez

0 Hari T O0 over 6 years ago in reply to Manuel Chavez

Genius 3820 points

Hi Manuel,

Hope you are doing great....
May I know how power monitor " input measurements can exceed power rails".
And also why their maximum voltage allowed on shunt is limited to a psecific value.

Regards
hari

0 Hari T O0 over 6 years ago in reply to Hari T O0

Genius 3820 points

Hi manuel,

For INA219 1LSB is 10uV and maximum voltage allowed on shunt is 320mV correct me if I am wrong.

Regards
hari

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hi Hari!

Yes, the LSB size for shunt voltage measurements is 10uV and if you are using /8 gain setting the maximum recommended shunt voltage is +/-320mV.

Best,
Manuel Chavez

0 Manuel Chavez over 6 years ago in reply to Hari T O0

TI__Genius 17100 points

Hari,

I am well thank you. Current shunt monitors can measure voltages which exceed the device's power supplies thanks to a bias on the front end. The maximum shunt voltage the device can tolerate is limited by this same bias and its protective cells (maximum operating conditions), but the measurable range is much smaller (recommended operating conditions). If this range is exceeded the device will simply saturate to GND or the supply.

If you'd like details on current shunt monitor topics, please check out this informative video series: training.ti.com/getting-started-current-sense-amplifiers

Hope to have been helpful.
Manuel Chavez

Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

Data converters

Data converters forum

Power Monitoring IC Required