Solar powering a rover

I think that n paper the CE10144 would be more suitable with the higher discharge capability and capacity.

I will say this one more time.
You have to consider the start up current for both servo and wheel motor. This will be the stall current which should be documented, if not go somewhere else. This is worst case scenario but has to be considered as it will be highly likely that all 6 motors will start up together. Things would improve if the motors have a soft start or speed increase built in to the controller. But I always suggest that the stall current be the one considered as this will be absolutely worst case and you are not guessing what might be and eventually run into trouble.

The stall current can be closely calculated fairly easily if you can’t find it anywhere. If you don’t know how to do this let me know and I will attempt to enlighten you.

If contemplating using the larger 18650 please note this in the product description.
“This Battery is almost 70mm long; it will not fit the majority of 18650 Battery Holders.”
This despite the fact it is still sold as 18650 which implies a length of 65mm. The self protection in the nipple end is the culprit.
Cheers Bob

Hey all,

Just wanted to post one more quick question here now I have made some progress. I have designed and made a custom solar charging module using a CN3722 chip and have successfully got it working with a 2s BMS module.

My question is, now that I am wiring everything up, is it best to connect the output of the charging module and power rail going to the buck converter in parallel on the output terminals of the BMS? That is, the BMS output terminals would have four wires going out, two to the output terminals of the charger, and two going to the buck to power the MCU and motors. Or will this damage the charger when it and the motors are running at the same time?

Cheers

1 Like

Hi Euan,

How sweet! What a sub-project!

I’m having a bit of trouble visualising how exactly you have everything connected - could you please send through a quick diagram of how you intend to connect it?

Liam

1 Like

Hi Liam,

Thanks for the response!

This is a rough sketch of how I have it wired up. My concern is that maybe the 1A-ish continuous load from the motors might damage the CN3722 charger, or will the batteries be the ones taking up the slack?

Right now I am coding the MCU and so I haven’t actually driven it with the solar panels on yet since I need access to the top for wiring and stuff while I am still bug fixing.

Thanks

1 Like

Hi Euan,

Looks sweet! That sketch helps a lot.

The bit I’d want to verify before saying “Yes, run and charge at the same time” is the exact CN3722 circuit you’ve built. I don’t have the schematic for your module from the sketch alone, so I wouldn’t assume it has proper power-path/load-sharing behaviour unless the design notes explicitly say so.

Again, without the full design context, it might be worth adding reverse current protection on the charger (as this part of the circuit should only be charging the battery, or helping it provide power to the motors.

PS: Will you be able to supplement the solar charging with a mains adapter?

Liam

For the circuit, I basically copied the example circuit in the chip’s datasheet, and set resistor / capacitor values based on the formulae they gave and the parameters of the panels and batteries.

Maybe to play It safe i can keep the solar panels attached, but not connect them when the robot is turned on, only connect them when i have shut it off.

At this stage no, to charge the batteries when I received them I just set up the charging circuit on a breadboard on a table outside for a few days until it charged. I should probably invest in a mains charger as well.

1 Like

Hi Euan,

I’d keep going with that “charge only while shut off” plan for now.
You could even use an SPDT switch/Relay to hand over power from one system to another.

That avoids the motor current pulses confusing the charger or pulling current through a path that wasn’t designed for it. The battery/BMS should be treated as the energy source for the rover, and the CN3722 board as the charger.

Liam

2 Likes

Hi Euan, I agree with keeping charge and run separate for now, but the bigger issue is that most charger chips decide the battery is full by watching the current taper off. If the motors pull current from the same terminals, the charger never sees that taper, so it can sit in constant-voltage mode and keep pushing the cells while the load is running. It is not usually destructive, it just stops the charge termination working properly. Also check that the BMS charge and discharge paths are both rated for your motor stall current, not just the average. If you want charge-while-driving later, a power-path stage that feeds the load first and the battery second is the clean fix. A fuse on the battery side is cheap insurance too.

The battery may remain as the primary power source. Let the solar panel recharge it continuously. The battery must handle motor startup and current peaks. A 3W panel seems undersized for this application. Its output cannot sustain a 3–4A load. Consider a larger panel with sufficient mounting space. Also, separate the motor and logic power rails. Use regulated supplies for the controller and camera. Check battery discharge ratings and protection circuit limits. Choose motor drivers based on continuous and peak currents. Add current sensing for stall detection. PWM ramping can reduce startup current spikes. However, software cannot prevent sustained hardware overloads. I’d avoid paralleling power managers without manufacturer approval. First, measure actual motor currents under realistic loads. Then establish your complete power budget. Unikeyic is relevant when engineers need to move from component specifications to purchasing checks. If you do not find all of your desired parts here, you can search there for the exact parts or alternatives.