Pico 2W PCB design (KiCAD)

41

Hi Trevor

Good attitude. If you lose power the pump probably won’t work anyway.
I am a great believer in keeping things as simple as possible. Complexity is probably OK at experimental and playing stage but if you have something that you want to go on working for a useful period any unneeded added complexity can and often is a source of problems down stream.

I think everybody has done some learning here. I for one will find out a bit more about strange little boxes simply marked “Lipo” before I weigh in with any suggestions. If I had realised this box was “charge only” I would have had a different approach. And I think that you may have learned that enough info on the schematic will not cause or increase confusion.

We will all get there.
Any more help just give a yell.
Cheers Bob

42

@Robert93820 this box was “charge only”

I actually had some second thoughts overnight about my “give up” stance!

I am still suffering from being pulled in too many different directions on this and consequently losing focus… but… your comment that the “LiPo” box is charge-only puzzles me, as waaaaay back, I did a test. I actually have code to monitor VBUS on the Pico, and if USB power is disconnected for, say, 3 minutes, my system would do an orderly shutdown, all whilst running off the LiPo battery. THIS WORKED! (Well, unless I am really losing my marbles…which is a definite possibility, so I better retest this, hey :wink: )

The other misgiving I have about abandoning my plan is this: I have always intended to have a secondary box, housing things like 24V transformers (to drive a solenoid), relays, 5V P/S, some other bits’n’pieces - but NO 240V GPO outlet for plugging a USB charger into… and feed all the necessary wires into my Pico box, which kinda needs to be close to the top of my tank, for I2C length issues if nothing else, whilst the secondary box needs to be closer to the ground for other reasons.

Bottom line: I am mentally exhausted from the last week…(working until 4AM on PCB design didn’t help!), but will likely persist with this a bit yet - even if I drop the LiPo, surely using the Raspberry-supplied circuit for adding an external supply, as per the Pico datasheet, should be feasible?

PS One of the other contributing factors to mental exhaustion was my battles with an obstinate KiCAD issue… in the end, I started a new KiCAD project, dropped in a minmal set of components (A Pico, an op-amp) and an external power connector, with +ve pin on the VDD net. Pico Pin 40 was labelled VDD Everything was fine… I could run a trace from VSYS (Pin 39) to my opamp pin 4, on the VSYS net. Great.. it all works as expected… so, the question is, why, in my pump project KiCAD simple refused to run a trace from Pin39/VSYS to my other VSYS net? That’s right. No mistake.

So, I thought, I’ll now make my pump project schematic “more like” my little test circuit. This meant changing my VCC net name to VDD, which is what Pin 40 is labelled on the Pico footprint. And, I think that is the ONLY significant change I made… and KiCAD was then perfectly happy to route Pin 39 to my op-amp etc. on VSYS Go Figure!!

Anyway… that’s another little “learning”, I guess, but it sure was driving me to despair!

I might take a wee break from this… I have other Python programming to add a SLEEP mode (so I don’t keep sampling my water level every 5 seconds, and pipline pressure every second) when I really don’t need to!)

Thanks again for sticking with me on this

Cheers, T.

1 Like
43

Hi Trevor

To comment here we would have to know EXACTLY how all this was connected. If you have not got a paper trail (if not you should have) you might have a bit of trouble unless your memory is pretty good.

If your BATTERY was connected to VSYS somehow then certainly the Pico would power to enable shut down but you would have no 5V. Your battery V (whatever that is at the moment) would appear here instead.
Cheers Bob

1 Like
44

Hey @Trevor277988,

Sometimes pausing, reassessing, and choosing the simpler path is the real engineering win. Great to see you racking up PCB experience; each board gets easier… though let’s be honest, the next board is usually just a remix of fixing all the “learning opportunities” from the last one :grinning_face_with_smiling_eyes:

Yeah, following the Pico datasheet gives some well-thought-out, and more importantly, well-tested, examples that will provide a reliable external power setup.

Unfortunately, as I’ve mentioned, I’m not well-versed in KiCAD at all, so I can’t help with the weird net issues, but it sounds like you’ve been able to get that one sorted.

Best of luck, and we’ll be here for any further issues :slightly_smiling_face:

45

Good news @Robert93820 … I am NOT losing my marbles! (Well, not completely, anyway :wink: )

My VBUS monitorig code continues to run AFTER removing USB power, an extract from my log file proves it.

2025/09/16 18:07:29 VBUS LOST
2025/09/16 18:07:39 VBUS RESTORED
2025/09/16 18:11:24 VBUS LOST
2025/09/16 18:13:15 VBUS RESTORED

My Pico, LCD and Transceivers (I2C devices) all continue to run off the LiPo. This is with my Pico connected to the Core PicoDev platform… which uses that same LiPo charge device, .the MCP73831 That’s why I chose it for my project!

Core’s circuit attached… it doesn’t show the part number either, but I got it straight from Core support.

PiicoDev LiPo Expansion Board.pdf (164.3 KB)

Anyway, I really am going to (try) to put this aside for a few days while I sort out some other issues. It will become critical as I get closer to sending of Gerber files!

Cheers, T.

PS… got a bunch of probe ground springs the other day. And my logic analyser, so am raring to go next time a thorny I2C issue arises!

1 Like
46

Hi Trevor

OK That is fine.
I had a look at that link to the expansion board and noted the cap in question and the charging circuit. OK.
I also noted the Mosfet switching circuit which IS NOT how it is portrayed earlier. That circuit as shown, Gate to VBUS and 100kR to ground Drain to VBATT and Source to VSYS will connect VBATT to VSYS when VBUS is removed. VBUS will hold the Mosfet OFF while present.

This will put VBATT to your sensor and LMC6484 instead of 5V which is the case when VBUS is present.

I know the LMC6484 would be happy at VBATT voltage but I don’t know about the sensor. VSYS (VBATT or 5V) is regulated down to 3.3V for everything else so would keep the Pico etc active until organised shut down.

It would appear that your sensor is OK with 3.3V. If you are shutting down when you lose VBUS it would not matter anyway.
Cheers Bob

1 Like