Author Topic: Buck converter at 100%  (Read 1666 times)

0 Members and 1 Guest are viewing this topic.

Offline cv007Topic starter

  • Frequent Contributor
  • **
  • Posts: 861
Buck converter at 100%
« on: January 12, 2019, 06:50:15 pm »
I have a circuit with a couple pairs of p-channel mosfets that control output from a couple inputs (see attached picture). The inputs are 5v on one side which also powers a buck converter, and the other input is the buck converter output of 3v3. The mosfet 'outputs' are tied together to get either 5v or 3v3, or no output. The back-to-back mosfets prevent the 3v3 side from conducting via the body diode when the 5v side is enabled. The extra p-channel on the 5v side is not needed, but these are 2-mosfets in one package so just added it.

A micro controls the output selection via a couple pins in open drain mode. The buck converter enable pin is also controlled by the micro. The micro runs from 5v.

None of this is actually tested, only simulated somewhat in my (parts limited) simulation software.


Now, looking closer at the buck converter datasheet I noticed it has 100% duty cycle capability.
Maybe I misunderstand, but it seems I can eliminate my mosfet switching circuit-

buck converter EN low- internal mosfets off = no Vin->LX = 0V out
buck converter EN high- FB = R1,R2 (to get 3V3 out) = 3V3 out
buck converter EN high- FB = >0.3v (prevent over current protection I think), <0.6v (keep duty at 100%) = Vin->LX = 5V (less pch mosfet loss)
(dac pin could drive fb pin to something less than 0.6v to get 5v/max output, or just use the pin to provide a path to ground for another resistor in fb)

these are my current buck converter choices (just need up to 1A, easy to use, low parts count, and larger than a grain of sand)
Diodes Inc. AP3924A or AP3417C
https://www.mouser.com/datasheet/2/115/AP3429_A-766309.pdf
https://www.mouser.com/datasheet/2/115/AP3417C-271484.pdf

Am I missing something, or can the buck converter take over my switching duties? And am I correct in assuming just keeping FB below the reference voltage of 0.6v gets me 100% duty cycle (in these parts)?

Thanks.
 

Offline MrAl

  • Super Contributor
  • ***
  • Posts: 1666
Re: Buck converter at 100%
« Reply #1 on: January 12, 2019, 09:14:21 pm »
I have a circuit with a couple pairs of p-channel mosfets that control output from a couple inputs (see attached picture). The inputs are 5v on one side which also powers a buck converter, and the other input is the buck converter output of 3v3. The mosfet 'outputs' are tied together to get either 5v or 3v3, or no output. The back-to-back mosfets prevent the 3v3 side from conducting via the body diode when the 5v side is enabled. The extra p-channel on the 5v side is not needed, but these are 2-mosfets in one package so just added it.

A micro controls the output selection via a couple pins in open drain mode. The buck converter enable pin is also controlled by the micro. The micro runs from 5v.

None of this is actually tested, only simulated somewhat in my (parts limited) simulation software.


Now, looking closer at the buck converter datasheet I noticed it has 100% duty cycle capability.
Maybe I misunderstand, but it seems I can eliminate my mosfet switching circuit-

buck converter EN low- internal mosfets off = no Vin->LX = 0V out
buck converter EN high- FB = R1,R2 (to get 3V3 out) = 3V3 out
buck converter EN high- FB = >0.3v (prevent over current protection I think), <0.6v (keep duty at 100%) = Vin->LX = 5V (less pch mosfet loss)
(dac pin could drive fb pin to something less than 0.6v to get 5v/max output, or just use the pin to provide a path to ground for another resistor in fb)

these are my current buck converter choices (just need up to 1A, easy to use, low parts count, and larger than a grain of sand)
Diodes Inc. AP3924A or AP3417C
https://www.mouser.com/datasheet/2/115/AP3429_A-766309.pdf
https://www.mouser.com/datasheet/2/115/AP3417C-271484.pdf

Am I missing something, or can the buck converter take over my switching duties? And am I correct in assuming just keeping FB below the reference voltage of 0.6v gets me 100% duty cycle (in these parts)?

Thanks.

Hello,

100 percent usually means it can connect the input directly to the output for a 1:1 ratio input to output.
So that usually does not mean 0 to 100 percent.
The data sheet for that first one says the output is 0.6v min, so that's not completely off.
You could look to see if they have a shutdown pin though.
 

Offline cv007Topic starter

  • Frequent Contributor
  • **
  • Posts: 861
Re: Buck converter at 100%
« Reply #2 on: January 13, 2019, 12:34:34 am »
Quote
The data sheet for that first one says the output is 0.6v min, so that's not completely off.
When enabled. I would assume the minimum output is simply due to the internal reference at 0.6v.

It would seem odd to me that they have an enable pin, have control over the output mosfets so the high side (p-ch) could be turned off (via a built-in pullup), list the shutdown current as <1ua, and still let 0.6v out of the output pin when not enabled.

the 3429 lists 'Output Voltage: 0.6V to VIN', '100% Duty Cycle Operation'
the 3417 lists 'Adjustable Output Voltage: 0.6V to 0.9×VIN' , 'Low Dropout Operation at 100% Duty Cycle'

so it seems the 3417 has more loss when at 100% duty cycle, or maybe its just a little newer where they have more a accurate description.
the 'Low Dropout' and 0.9xVin does not inspire confidence

May have to just do trial and error.
 

Offline MrAl

  • Super Contributor
  • ***
  • Posts: 1666
Re: Buck converter at 100%
« Reply #3 on: January 13, 2019, 01:43:59 am »
Quote
The data sheet for that first one says the output is 0.6v min, so that's not completely off.
When enabled. I would assume the minimum output is simply due to the internal reference at 0.6v.

It would seem odd to me that they have an enable pin, have control over the output mosfets so the high side (p-ch) could be turned off (via a built-in pullup), list the shutdown current as <1ua, and still let 0.6v out of the output pin when not enabled.

the 3429 lists 'Output Voltage: 0.6V to VIN', '100% Duty Cycle Operation'
the 3417 lists 'Adjustable Output Voltage: 0.6V to 0.9×VIN' , 'Low Dropout Operation at 100% Duty Cycle'

so it seems the 3417 has more loss when at 100% duty cycle, or maybe its just a little newer where they have more a accurate description.
the 'Low Dropout' and 0.9xVin does not inspire confidence

May have to just do trial and error.

I dont think those two have an enable pin.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9318
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Buck converter at 100%
« Reply #4 on: January 13, 2019, 02:05:21 am »
They both do, listed as EN on the datasheets.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Buck converter at 100%
« Reply #5 on: January 13, 2019, 03:03:36 am »
Yes, this sounds like a fine solution -- and FAR more pleasant than having to hack in a pair of switches!

If you did need the switch (or still need one elsewhere), a quirky alternative I've seen is: a four terminal (substrate taken out to pin!) PMOS.  As quirky as this sounds, it's apparently useful enough to exist, and for exactly this sort of application because with the substrate pinned out, you can tie it to the highest voltage in the circuit (say with ORing diodes) to prevent backflow.
https://www.digikey.com/product-detail/en/microchip-technology/MIC94051YM4-TR/MIC94051YM4-CT-ND/9643046

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline cv007Topic starter

  • Frequent Contributor
  • **
  • Posts: 861
Re: Buck converter at 100%
« Reply #6 on: January 13, 2019, 03:32:51 am »
Quote
If you did need the switch
I looked for such a thing, but could not find one. Now I know such a thing exists (I'll bookmark that, and download the datasheet).

I originally had 3 sot23 p-channel low-rds mosfets, then switched to a pair of double p-channels in so-8 simply to get 1 less part to lay down (and just looks a little nicer on the board). With the above part I'm down to 2 small sot and 2 less pullups.  And with the buck converter doing it all, I can eliminate all the above.

I'll probably get the 3429A as it sounds like I may get closer to Vin with that one (or just try both).

Now I can tear up a third of my pcb design. But that is good (for me, less parts=better).

Thanks.
 

Offline MrAl

  • Super Contributor
  • ***
  • Posts: 1666
Re: Buck converter at 100%
« Reply #7 on: January 13, 2019, 01:47:27 pm »
Quote
If you did need the switch
I looked for such a thing, but could not find one. Now I know such a thing exists (I'll bookmark that, and download the datasheet).

I originally had 3 sot23 p-channel low-rds mosfets, then switched to a pair of double p-channels in so-8 simply to get 1 less part to lay down (and just looks a little nicer on the board). With the above part I'm down to 2 small sot and 2 less pullups.  And with the buck converter doing it all, I can eliminate all the above.

I'll probably get the 3429A as it sounds like I may get closer to Vin with that one (or just try both).

Now I can tear up a third of my pcb design. But that is good (for me, less parts=better).

Thanks.

Hi,

Well you could also look for one with an external 'slow start' mechanism.  You might be able to use that to disable the chip.  In all cases though check the data sheet and look for application notes.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf