LTC2971 VOUT_SNS1 About the maximum voltage

Hi Uboti,

If I may ask, which figure number are you referring to?  The LTC2971 only senses voltages/currents and enables DC/DC converters.  It doesn't know about or touch PWM pins.  You can use any buck ,boost, invert, SEPIC, Cuk, etc that you like as long as you select the proper 2971 option for voltage sensing.

The VIN_SNS specification is found on page 9.  The Abs Max rating is 63V independent of the base LTC2971, -1, -2, or -3 option.

You might want to look at the demo board DC2875. It has various DC/DC converters on the board, one is a boost that takes 12V in and creates 48V out.

https://www.analog.com/en/design-center/evaluation-hardware-and-software/evaluation-boards-kits/dc2875a.html

Regards

Mike

thank you for your reply.
VPWM was a typo. VPWR was correct.

I saw DC2875A. This controls the step-down DCDC.
Q１：In the case of boosting, the connection will be as follows, is there any problem?
(I would like to use LTC2971 to improve voltage variation)

Let me ask you another question.
I am wondering if it can be used even with a high voltage DCDC.

Q2:Is it possible to use it as shown in the figure below?

OK got it, VPWR. I do not recommend that you tie VPWR to 48V.  The chip will get hot.  Almost 0.5W power dissipation.

On the first page of the DC2875 schematic, I show a block diagram for all power stages.  You'll notice that I used a 48V to 5V switcher to power the 2971.  The part number is ADP2360-5.  It and external components are very small in overall footprint size.  With an LDO, you're just transferring the power dissipation from one chip to another.  A switcher is the best option.

Typically you would like the 2971 to sense the voltage of the input supply that powers the managed DC/DC converters.  So, your first drawing is OK but does not have the 5V supply for VPWR.  Your second drawing is not OK for 2 reasons: VIN_SNS needs to be tied directly to the input of the buck-boost, and VPWR ought to be powered from the furthest upstream supply in the system (in this case 24V Vin of the buck-boost or a 12V or 5V buck from Vin). 

Think of the 5V buck (VPWR) as a standby supply that is powered up immediately when Vin is applied.  It takes ~30ms for the 2971 to initialize which is another reason why you want the 2971 powered up early.

Mike

OK got it, VPWR. I do not recommend that you tie VPWR to 48V.  The chip will get hot.  Almost 0.5W power dissipation.

On the first page of the DC2875 schematic, I show a block diagram for all power stages.  You'll notice that I used a 48V to 5V switcher to power the 2971.  The part number is ADP2360-5.  It and external components are very small in overall footprint size.  With an LDO, you're just transferring the power dissipation from one chip to another.  A switcher is the best option.

Typically you would like the 2971 to sense the voltage of the input supply that powers the managed DC/DC converters.  So, your first drawing is OK but does not have the 5V supply for VPWR.  Your second drawing is not OK for 2 reasons: VIN_SNS needs to be tied directly to the input of the buck-boost, and VPWR ought to be powered from the furthest upstream supply in the system (in this case 24V Vin of the buck-boost or a 12V or 5V buck from Vin). 

Think of the 5V buck (VPWR) as a standby supply that is powered up immediately when Vin is applied.  It takes ~30ms for the 2971 to initialize which is another reason why you want the 2971 powered up early.

Mike

Thank you for your answer

Excuse me, please tell me one more thing.

Based on the story I received, do I understand that the following is okay?

Yes, that looks good.  You can tie the 2971's VOUT_EN pin to the converter's RUN pin.  And you can add an output Rsense to measure the converter's output current.

Mike