0

Our team is currently working on a Battery Management System with specifications as below:

  • 12 x 3.3 V, 50 Ah LiFePO4 Batteries, connected in series
  • LTC6803-4 Battery Controller
  • Monitored via Arduino Uno with SPI

Here's the board schematic that has been created. All resistor values are indicated with their 3-digit SMD code (221 for 220 Ω, 222 for 2200 Ω, etc.):

Revised BMS Board Schematic

However, the LTC6803-4 didn't turn on when the battery pack was connected. It's indicated with the Vreg pin not supplying 5V properly, just around 2V or less. So LED13 connected to it won't turn on.

We already attempted to do this:

  • Re-validate the board, now it supposed to comply the datasheet based on three attempts of redesign the schematic and re-checked several times.
  • Replaced different LTC6803-4 multiple times and all have the same result, until we worried that we might just have burned all of them.
  • Changed the SDO pullup resistor from 220 Ω to 2200 Ω, then to 10 kΩ. The LTC turns on for a while and sends proper SPI signals, but not for long before the LED13 connected to it flickers and dies again. It can still be turned on after waiting for some time before connecting it again, at which the problem just repeats.
  • Shorted the D2, although for now we just put the D2 back and didn't care much about it, as the LTC6803 won't turn on at all regardless of whether the Arduino is connected or not.

We wonder if we actually missed something crucial.

NOTE: VDD and GND at bottom-right corner are put sideways on purpose due to viewport dimension constraint

Null
  • 7,448
  • 17
  • 36
  • 48
Thor-x86_128
  • 13
  • 7
  • 3
    The schematic is simply not readable, please redraw it in conventional ways. Supply voltage is up, ground is down, N-channel drain/source is to the right not the left, and so on. [Rules and guidelines for drawing good schematics](https://electronics.stackexchange.com/questions/28251/rules-and-guidelines-for-drawing-good-schematics) – Lundin Jan 04 '22 at 14:05
  • "All resistor values are indicated with their 3-digit SMD code (221 for 220 Ω ... ". Well, that's random. – Graham Nye Jan 04 '22 at 16:56
  • @Lundin I'm confused which supply voltage that doesn't up. Anyway, I'm going to ask my teammate who designed the schematic to make it more readable. Thanks for your information – Thor-x86_128 Jan 04 '22 at 17:12
  • 2
    @GrahamNye It's clearly written on top of the SMD resistors. So it's not random as you think – Thor-x86_128 Jan 04 '22 at 17:14
  • 1
    Oh, it's random. The circuit diagram should show the actual numeric values, not the SMD codes. If you were using through-hole resistors would your circuit diagram show the colour bands instead? (Perhaps I shouldn't mention that in case it gives people ideas.) – Graham Nye Jan 04 '22 at 21:11
  • It's because you didn't connect anything to the V+ pin. – Davide Andrea Jan 05 '22 at 00:28
  • @Lundin please review the revised schematic, it's now updated – Thor-x86_128 Jan 05 '22 at 03:05
  • @Thor-x86_128 have you found the solution to that problem im also facing the same problem. thank you – gaurav kolte Aug 29 '23 at 13:37
  • _”All resistor values are indicated with their 3-digit SMD code (221 for 220 Ω, 222 for 2200 Ω, etc.)”_ please don’t! – winny Aug 29 '23 at 13:43
  • VREG must feed the isolator IC. An SI8841, as mentioned in the schematic, does not exist. So please specify the used isolator IC to verify that VREG can feed this IC. – Jens Aug 29 '23 at 20:37

0 Answers0