First board layout - LiFePo4 Gas Gauge

Hi, I have finally tried to fully layout a board in Kicad having played around with the software for something like 15 years, on and off! I am seriously impressed with Kicad 5 and really appreciate all the hard work that has gone into it - thank you.

This is a very simple breakout board for the STC3100 battery monitor / Coulomb counter. There is a commercial eval board available, but it costs around $60 so I thought I would have a go at building my own. I am hoping it will work with a LiFePo4 14500 single cell battery that is powering a ESP8266 board.

The circuit is heavily influenced by the reference design and the app. note that suggests adding ESD protection.

Datasheet: http://www.st.com/resource/en/datasheet/stc3100.pdf
App Note: http://www.st.com/resource/en/application_note/dm00027741.pdf

I have tried to follow the manufacturer’s PCB layout guidance in the app. note. One issue that I faced was connecting the GND terminal of the sense resistor to the GND terminal of the IC, but not through the ground plane. I found a post on this forum that suggested using keep-out areas, so this is what I have done.

All the components are 1206 because I have a set of these and this will also be my first attempt at hand soldering SMD.

I am hoping to upload the design to OSHPark soon, so any comments or critique are very welcome :slight_smile:

Thanks!

Congrats on taking the plunge! Your circuit looks like a good start.

Some comments:

  • PT_CG doesn’t need the trace on top when exiting pin 7. Just start on the bottom layer.
  • PT_SCL, PT_SDA and PT_IO0 don’t need to be on the bottom layer. You can just route them on the top.
  • I’m not certain about pin 9 (not mentioned in my cursory look at the datasheet) but unless there is a reason not to, you’ll usually want to connect those to the ground plane.
  • You’ve got a bit of extra trace on the top layer next to pin 1.
  • C1 is meant to decouple VCC. As such, it should be inbetween your input and the pin. You have done this correctly for C2. I’d move C1 above U1, allowing C2 to move closer to the pin it decouples.
  • You are using a thicker trace for PT_CG than for your power input PT_VCC and PT_VBATT. Is this intentional?
1 Like

Seth_h,

Really thank you for taking the time to review my board, it is incredibly useful.

I think I have improved the layout according to your comments.

You are using a thicker trace for PT_CG than for your power input PT_VCC and PT_VBATT. Is this intentional?

This should be intentional (I think). IIUC the entire battery current (~100mA in my case, but could be up to 2A) flows through PC_CG trace and the shunt, whilst only the current drawn by this board flows on PT_VCC / PT_VBATT.

Here is the revised board:

Thanks again :slight_smile:

Final version:

  • Increased default track width from 0.3mm to 0.35mm
  • Labels on connector pins
  • Keepout around mounting holes
  • Re-positioned some legends
  • Corrected D1 footprint to SOD-123

And off to OSHPark!

2 Likes

I’ll nominate that as the most HUMBLY magnificent STC3100 demo board in the galaxy! :grinning:

Dale

1 Like

You could make that totally single sided if you wanted to ? (and smaller too…)

2 Likes

@dchisholm

Ha ha, thanks!!

@PCB_Wiz

Goodness, yes I could have easily made it single sided - I see that now. Oh well, I think the OSHPark charge is the same for single and double sided so it’s okay for this batch.

With regard to size, I am planning to make another version with a charging circuit, either a BQ25070 or a MCP73123 in the same form factor.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.