Author Topic: Power Delivery issues with boost converter  (Read 1021 times)

0 Members and 1 Guest are viewing this topic.

Offline tyguy2Topic starter

  • Supporter
  • ****
  • Posts: 303
  • Country: us
Power Delivery issues with boost converter
« on: July 18, 2019, 11:26:08 pm »
Hello all,

I've recently designed a 5v rail that takes a minimum of 3.3v from a battery and steps it up to 5v, with a maximum current draw of 500ma (schematic attached, here's the part https://www.digikey.com/product-detail/en/maxim-integrated/MAX1522EUT-T/MAX1522EUT-TCT-ND/6680188). I'm already imbedded the design into a PCB, but I'm getting strange voltage sags when I try to draw anything over around 200ma. Does anyone have any ideas on how I can fix this without having to rerun a new PCB?
[Sarcastic comment] clever joke [/sarcastic comment]
Bitcoin:
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Power Delivery issues with boost converter
« Reply #1 on: July 19, 2019, 12:21:32 am »
I can't see anything obvious from the data sheet, are you sure the input 3.3V isn't dropping.
.  That took much longer than I thought it would.
 

Offline tyguy2Topic starter

  • Supporter
  • ****
  • Posts: 303
  • Country: us
Re: Power Delivery issues with boost converter
« Reply #2 on: July 19, 2019, 01:34:43 am »
I've checked that, and I'm not getting any issues with that either. 3.3v comes directly from a LiPo
[Sarcastic comment] clever joke [/sarcastic comment]
Bitcoin:
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
 

Offline Audioguru again

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ca
Re: Power Delivery issues with boost converter
« Reply #3 on: July 19, 2019, 02:05:30 am »
Your 3.3V Li-PO is almost dead. It is 4.2V when fully charged and is 3.7V when half-charged. it should be disconnected and re-charged when it is 3.2V.
Are you drawing 5A from the 3.3V Li-PO? That is 5A x 3.3V= 16.5W. Maybe 1.6W is wasted on heat then the 5V output will be 15.5W/5V= 3.1A but a Li-PO that is almost dead cannot produce the original 5A.
 

Offline tyguy2Topic starter

  • Supporter
  • ****
  • Posts: 303
  • Country: us
Re: Power Delivery issues with boost converter
« Reply #4 on: July 19, 2019, 03:12:10 am »
Your 3.3V Li-PO is almost dead. It is 4.2V when fully charged and is 3.7V when half-charged. it should be disconnected and re-charged when it is 3.2V.
Are you drawing 5A from the 3.3V Li-PO? That is 5A x 3.3V= 16.5W. Maybe 1.6W is wasted on heat then the 5V output will be 15.5W/5V= 3.1A but a Li-PO that is almost dead cannot produce the original 5A.
No, this Lipo is fully charged to 4.2v, I'm just having it operate down to the drop off voltage of the battery. I said I was drawing 500ma, not 5A.
[Sarcastic comment] clever joke [/sarcastic comment]
Bitcoin:
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 10385
  • Country: nz
Re: Power Delivery issues with boost converter
« Reply #5 on: July 19, 2019, 03:27:12 am »
D9 (DB2J310) is only rated to 200mA absolute max
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline tyguy2Topic starter

  • Supporter
  • ****
  • Posts: 303
  • Country: us
Re: Power Delivery issues with boost converter
« Reply #6 on: July 19, 2019, 04:21:31 am »
D9 (DB2J310) is only rated to 200mA absolute max

 :palm: Right in my face the entire time. Thanks!
[Sarcastic comment] clever joke [/sarcastic comment]
Bitcoin:
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Power Delivery issues with boost converter
« Reply #7 on: July 19, 2019, 10:13:10 am »
D9 (DB2J310) is only rated to 200mA absolute max

 :palm: Right in my face the entire time. Thanks!

Yeah, that is a problem, but not one that would cause Vo to sag. A more likely cause is saturation of the boost inductor. It needs to withstand a peak current considerably higher than the average dc input current (e.g. - 40-100% more) without losing too much inductance in the process.

EDIT - corrected "wouldn't" to "would"
« Last Edit: July 19, 2019, 03:33:05 pm by MagicSmoker »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf