ModRobotics is now offering a 3000mAh 3A 3S1P Lithium-Ion rechargeable battery for the GoPiGo3:
Prior GoPiGo kits shipped with an (empty) 8-cell battery holder that sometimes did not hold the cells securely, and required users to either purchase individual cells and a charger, or to purchase a Dexter Industries Rechargeable battery and Charger set.
In another thread:
Out of an abundance of caution, and to future users’ benefit the GoPiGo3 kit now ships with the above rechargeable battery at a price much less than the prior kit plus the separate rechargeable battery/charger.
Some questions:
Does the new battery come with a DC supply for charging?
I believe the current GoPiGo3 “red board” has a low voltage shutdown feature, and I read there is a new GoPiGo3 “red board” in process. Is the new board going to continue to support the existing low voltage shutdown feature (even though it cannot be triggered by the new 3S1P Li-Ion battery)?
I believe the battery voltage sensing on the current red board maxes out under the fully charged 3S Li-Ion voltage. Will the coming GoPiGo3 board have a different voltage sensing range?
Back when I was investigating Li-Ion, I ordered a 3S1P 18650 holder with the battery board in it and a “1A” 12v wall wart (“1A” turned out to be for 220v, only 395mA at US voltage). My tests (outside on the patio) showed I would need to buy high quality Li-Ion 18650 cells to do better than my NiMHs.
The rechargeable battery pack is definitely Li-ion 18650 cells. We hope they are high quality! They passed all our tests, and TalentCell is a serious battery manufacturer.
Folks talk about charge and discharge rates in terms of x.yC where C = the capacity of the cells in milli-amp hours mAH. So charging is often 0.2C to 1C, and discharge is some “design rated” based on the application.
For my my FPV quad’s “high voltage” (4.35v) 450mAh 1S LiPo cells are rated 30C (~13.5A) but my quad only draws a max of 6A for a discharge rate of 13C.
The 18650 cells I tested were low quality, and delivered only 1700-1800 mAH to total shutdown.
Using 1750 as C, the 395mA discharge rate is around 0.2C.
“Playing” till the 10.5v “knee” would give about 4.5 hrs of playtime and only 15 minutes of “reserve for poorly aligned dock retries”. I probably would choose a voltage that allows more reserve with these particular cells.
The Dexter/ModRobotics battery pack contains better quality cells with a higher capacity, so I would need to cycle it a few times to get it to settle, then measure the curve at Carl’s “typical” discharge rate to identify the knee voltage and the slope after the knee.
Selecting a “time to get on the dock” is a trade-off between wanting a good long playtime but saving enough reserve for two failed docking attempts in Carl’s case.
As a comparison, I also tested a 3S1P “battery pack” which had a slightly better discharge profile:
Quality 3SP1 Li-ion could provide up to 30% longer playtimes
3SP1 Li-ion is heavier than 8 NiMH - which may affect turning accuracy (unknown)
I would need to build a new (simpler) ruleset for juicer - my charge management program
I do not have any cycle life data to compare to Carl’s Eneloop NiMH 750 cycles
3SP1 Li-ion with onboard BMS does not fit inside Carl’s current body
I feel safer with NiMH chemistry,
Took study cells and battery to recycler to get them out of the house
Carl hasn’t complained that he wants a longer playtime.
And now, the new ModRobotics pack seems to offer benefits but does not fit inside Carl.
(The “inside Carl” position improves turning accuracy and repeatability greatly. Rear deck mounting is totally a non-option.)
The “local Amazon” is called “Yandex” - by the same guy who invented Google - and the only thing I can find ships from the US and will take quite a while to get here.
