Just for the challenge, just for fun, just for bragging rights, and just to do a little showing off – all perfectly valid reasons to take on a project. It seems like one or more of those are behind this tiny ESP32 board that’s barely larger than the coin cell that powers it.
From the video below, [Mike Rankin] has been working down the scale in terms of powering and sizing his ESP32 builds. He recently completed a project with an ESP32 Pico D4 and an OLED display that fits exactly on an AA battery holder, which he populated with a rechargeable 14550. Not satisfied with that form factor, he designed another board, this time barely larger than the LIR2450 rechargeable coin cell in its battery holder. In addition to the Pico D4, the board sports a USB charging and programming socket, a low drop-out (LDO) voltage regulator, an accelerometer, a tiny RGB LED, and a 96×16 OLED display. Rather than claim real estate for switches, [Mike] chose to add a pair of pads to the back of the board and use them as capacitive touch sensors. We found that bit very clever.
Sadly, the board doesn’t do much – yet – but that doesn’t mean we’re not impressed. And [Mike]’s no stranger to miniaturization projects, of course; last year’s Open Hardware Summit badge was his brainchild.
[Ewald] tipped us off to this one. Thanks!
27 thoughts on “A Coin Cell Powers This Tiny ESP32 Dev Board”
That is very, very cool. If I could buy one of these now I would, although I’d probably want at least one pin exposed somehow for interfacing…maybe in place of the capacitive pad?
Also, kudos for using the CP2102N instead a you-know-what.
The you know what CH340E comes in a MSOP10 3*3mm package and does’t need a crystal.
I’ve put a few exposed pad son the backside. +3.3v, GND, a touch pad, SCL,SDA and two other GPIO: https://github.com/mike-rankin/ESP32_CoinCell/blob/master/Images/Exposed_GPIO_Pads.jpg
I have one for sale on Tindie now but i’m not sure if the whole coincell thing was such a good idea at only 5 minutes of run time or 10 hours waking every 10 minutes. I’ve just added the Altium Design files now in hopes that someone can build off of it or it can be helpful in some way. I lean towards the hardware design side of things so that’s why the software is so poorly done.
Doesn’t ship to Australia :( Oh well, I’ll keep an eye out for your future projects instead.
Ah ha another to add to the list for review as in maybe could be more palatable to some eg suit diverse more personality traits, it’s odd how some people seem to have occadional negative emotional reactions to industrial/hobby form factors virtually regardless of capacity or potential, yeah subtle I know ;-)
…and thus starts the micro-badge craze. :grin:
Using the MCP1703 LDO (2µA Quiescent Current) might have been a better choice over the SE5218 with 65µA for a low battery power device..
The LDO i’m using now has crazy high sleep current. I’ve order a much better one from LCSC, part number
HT7833 that is much better at around 7uA sleeping. Its drop in replacements which is also nice.
You could run a laptop off that cell, for all we know for just as long! Without telling us any useful info about battery life, or the power of the device, this isn’t that useful as an article.
“Current consumption is around 0.45mA when sleeping but up to 85mA running. Runtime testing is still in the works but battery life is only 5 minutes when the display is on and constantly connected to wifi. Connecting to wifi and turning on the display every 10 minutes appears to last 10 hours or so.”
Lately I’ve popped out the coincell and tacked on a regular square LiPo to the two pads to get more life out of it. For the next revision I may add on a micro JST connector in case you want to use another battery. The CP2102N has been re-done as well. Now it will only be powered by the microUSB connector to reduce sleep current. This along with an LDO that draws 7uA instead of 45uA makes for a better design.
I’m curious about the antenna:
What’s the performance like? Worth the extra height vs a pcb trace?
Thanks for sharing!
No idea. I’ve seen others use the antenna so I just used that one. A pcb trace would require a lot more effort and might not work as well.
You should use e-paper – more adapted to devices with limited energy which will need to power up for a very limited time…
Except that comes with the trade off of needing a more complex circuit to drive the E-ink device, taking up precious PCB space for this project. I believe most of those require negative voltage sources in addition to positive ones.
An SPI interface with a tiny DC-DC circuit.
The trINKet used an e-ink display. Amazing technology. The displays are physically very, very large though. If you want a low power design with a large display then e-ink is the way to go: https://github.com/mike-rankin/ESP_trINKet
Very cool. Really liked the form factor.
I would prefer a buck regulator because these esp32 at it’s peak performance takes a decent amount of current. Also I am wondering the battery discharge rate because again the pulse current when wifi activity happens is also kind of a bit higher, if we probe oscilloscope on the power rail or battery then it will definitely show some huge ripple for a low discharge rate battery. I remember I tested with a 150mAh lipo with 1C discharge the ripple is around 0.4V i guess and in my case when battery voltage reaches 3.4v average, it started dipping below 3v due to ripple.
But yes definitely this can be solved with a high discharge rate lipos, those toy drone lipos are perfect for these (may be a bit overkilling as well), no matter how small the capacity is.
I have a few of these tiny buck regulators that might work better than an LDO. The inductor I ordered at the same time is only 2x2mm as well: https://datasheet.lcsc.com/szlcsc/Richtek-Tech-RT8096CHGJ5_C250418.pdf
fantastic project. What was the cost of the materials with PCB roughly?
Around $20 or so.
How long does it run on a coin cell ?
A very sad 5 minutes with the display on a connected to wifi. 10 hours in sleep mode waking every 10 minutes.
LIR2450 has a nominal capacity of 100mAh, so 5 min running time implies 1.2A current draw—significantly more than 85mA the article claims. In other words, if it really used 85mA it should last about an hour. This estimate could be off because of 1) losses in the power supply circuit, 2) loss of capacity at large discharge current (it’s specified for 200mA max, so even 80 is quite a draw (0.8C).
The bottom line is that the 5 minute time is an anomaly—something draws much too much current, and it stresses the battery way beyond its specs. I don’t see what could be the culprit—it shouldn’t be the display, and it isn’t WiFi (because on WiFi alone the battery lasts 10hr).
I agree that the math doesn’t add up. The cells that I have from Aliexpress have a bit of rust and are scratched up so it looks like they are used ones. I’ve ordered more from another supplier and will re-test.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)