Replace big power supply with trickle-charged battery pack?

[ProgramThyself]

I have a stationary application where a motor controller that sees only intermittent use is powered by a 12 volt 10 amp bench supply. That’s a bit weak and something that can deliver 250 Watt would be better. Instead of a beefier power supply that probably has a fan, I wonder if a trickle-charged power-tool battery pack or something similar can work? 50 Wh ought to be enough. The controller (VESC 6 EDU) can handle up to 6s lithium. Also, there is sometimes backward (regen) current and it would be nice to just dump it into the battery instead of burning it in a Zener diode in parallel with the power supply.

Can this Makita pack deliver current while trickle charging?
Building battery from MAKITA Konion PACK
I see adapters for $16 on eBay. Reliability matters a lot.

 

[wturber]

Lithium batteries don't like to be kept fully charged. If I were to do the trickle charge thing you are considering, I'd set the charge voltage to max out at about 80% full charge voltage. In that case, the battery would only receive a charge after it has been depleted a bit by use or perhaps if it has been idle for a long time.

If power reliability isn't a problem, then why not just get an appropriately sized power supply? If you can use 24V (6s = max 25.2V), then this Meanwell unit should work and it isn't super expensive. HLG-240H-24 It supplies 240 watts.

https://www.amazon.com/HLG-240H-24A-Price-Switching-Supply-MeanWell/dp/B0BTT3DQKK/ref=sr_1_3?sr=8-3

You can adjust the voltage output on this supply to between 22.4-25.6 volts. And if you really wanted to have a battery backup, you could use this to charge the battery. I've used a Meanwell power supply like this to charge my 36V lithium ebike batteries for about six years not. Though you'd probably want to dial the current output down depending on the charge current your ebike battery would draw.

Others on this forum know tons more about electronic than I do and I'm sure they'll have some good insight into how to run the two in parallel or why my suggestion is flawed. But if the power grid is reliable in your area, maybe just having an upgraded power supply is enough. Adding a battery and the stuff you should also add to monitor the battery increases complexity.

 

[E-HP]

What is the reasoning behind trickle charging versus charging, using the pack, the recharging? Are you proposing leaving the battery charging while unattended?

 

[wturber]

What is the reasoning behind trickle charging versus charging, using the pack, the recharging? Are you proposing leaving the battery charging while unattended?

Good point. Were it me and I was set on using batteries and reliability was a bid deal. I'd have at least two battery packs. Use one until it hit a certain level of depletion and then swap in a freshly charged battery. You always have one battery in reserve and you can pick and choose when to charge for the most part. This application doesn't seem like it is going to use up a lot of battery charge each use. Also, given the low amount of total power used, wouldn't bother with regen unless implementing it was trivial.

 

[ProgramThyself]

What is the reasoning behind trickle charging versus charging, using the pack, the recharging? Are you proposing leaving the battery charging while unattended?

Regularly swapping packs is not really an option for me. Besides my mobility constraints, I'm sure I would forget at some point.

Yes, it's like keeping a power-tool pack in the charger, which I often do. Is that a big no-no??

 

[ProgramThyself]

Lithium batteries don't like to be kept fully charged. If I were to do the trickle charge thing you are considering, I'd set the charge voltage to max out at about 80% full charge voltage. In that case, the battery would only receive a charge after it has been depleted a bit by use or perhaps if it has been idle for a long time.

If power reliability isn't a problem, then why not just get an appropriately sized power supply? If you can use 24V (6s = max 25.2V), then this Meanwell unit should work and it isn't super expensive. HLG-240H-24 It supplies 240 watts.

https://www.amazon.com/HLG-240H-24A-Price-Switching-Supply-MeanWell/dp/B0BTT3DQKK/ref=sr_1_3?sr=8-3

You can adjust the voltage output on this supply to between 22.4-25.6 volts. And if you really wanted to have a battery backup, you could use this to charge the battery. I've used a Meanwell power supply like this to charge my 36V lithium ebike batteries for about six years not. Though you'd probably want to dial the current output down depending on the charge current your ebike battery would draw.

Others on this forum know tons more about electronic than I do and I'm sure they'll have some good insight into how to run the two in parallel or why my suggestion is flawed. But if the power grid is reliable in your area, maybe just having an upgraded power supply is enough. Adding a battery and the stuff you should also add to monitor the battery increases complexity.

Yes, 80% charging could work, but I think BMS cell balancing often only kicks in at higher charge. The Meanwell unit can also have UPS backup for power outages. Thanks for that suggestion.

Still, over-voltage protection easily gets triggered by reverse current. A beefy Schottky diode in series and Zener in parallel can isolate and dissipate the backwards current. I guess a trickle-charged small battery pack has more sweetness factor and saves some space and money.

 

[ProgramThyself]

Also, given the low amount of total power used, wouldn't bother with regen unless implementing it was trivial.

The reverse current has to go somewhere, doesn't it? Most power supplies can't absorb current. Then voltage rises, triggering over-voltage protection unless I add high-current isolation and Zener diodes.

 

[E-HP]

Yes, 80% charging could work, but I think BMS cell balancing often only kicks in at higher charge. The Meanwell unit can also have UPS backup for power outages.

Does the Makita bms balance? I looked at a few replacement bms specs and none of them listed balancing as a function, and didn't provide related specs, like balancing current. Is there a model number? 80% route is a good suggestion.

 

[wturber]

The reverse current has to go somewhere, doesn't it? Most power supplies can't absorb current. Then voltage rises, triggering over-voltage protection unless I add high-current isolation and Zener diodes.

I'm not familiar with how a VESC works. I ran an ebike for years with a direct drive and no regen. I assume the motor controller dissipated any reverse current that was generated. I further assumed that the controller (which is what actually faces the motor) does something so that there is little or no load across the motor's coils. With little or no load, there would be little or no current and hence little or nothing to dissipate. Also, no significant drag from the motor. I also operated that bike with a DC-DC boost converter that took the battery voltage up to 52V. That also worked fine. But again, that faced the controller, not the motor.

That controller did have an e-brake option that I assumed would put a load toward the motor. I never enabled it. But like I said, I'm not the electronics guy and I don't know anything about VESCs. I think of these things in a very basic way - mostly as magic boxes arranged in a flow diagram. I assumed that anything your bench power supply could handle from the motor/VESC could be handled by the Meanwell. But that may be a bad assumption. Maybe I'll read up on the differences between a VESC and a typical ebike controller. :^)

 

[ProgramThyself]

I'm not familiar with how a VESC works. I ran an ebike for years with a direct drive and no regen. I assume the motor controller dissipated any reverse current that was generated. I further assumed that the controller (which is what actually faces the motor) does something so that there is little or no load across the motor's coils. With little or no load, there would be little or no current and hence little or nothing to dissipate. Also, no significant drag from the motor. I also operated that bike with a DC-DC boost converter that took the battery voltage up to 52V. That also worked fine. But again, that faced the controller, not the motor.

That controller did have an e-brake option that I assumed would put a load toward the motor. I never enabled it. But like I said, I'm not the electronics guy and I don't know anything about VESCs. I think of these things in a very basic way - mostly as magic boxes arranged in a flow diagram. I assumed that anything your bench power supply could handle from the motor/VESC could be handled by the Meanwell. But that may be a bad assumption. Maybe I'll read up on the differences between a VESC and a typical ebike controller. :^)

For the intended use, there does need to be something like "e-brake" resulting in backwards current. The cheap adjustable bench supplies I've used don't detect 15V on a 12V rail as over voltage. Fixed supplies seem to shut off.

 

[ProgramThyself]

Does the Makita bms balance? I looked at a few replacement bms specs and none of them listed balancing as a function, and didn't provide related specs, like balancing current. Is there a model number? 80% route is a good suggestion.

OK, maybe Makita was a bad example. These DeWalt packs balance though and charge cells to only 4.1V.

DeWalt 20V Max 3.0Ah Battery Pack Teardown & Analysis

It’s been a little bit since I’ve torn apart a new battery pack! The last new-to-me pack I pulled apart was a 26v BionX battery (which, I’d add, I rebuilt to nearly twice the stock capacity by filling all the space with cells). And I’ve got this cute little DeWalt 20V MAX battery pack (model...

www.sevarg.net

That website's many pix of corroded cells make me reconsider the wisdom of using batteries though. Then again, dead power supplies with blown capacitors would also look ugly. Perhaps the right approach is to estimate ballpark values of MTBF.

 

[wturber]

I know this is a DIY channel and that sometimes the fun of something is simply the exercise of developing a solution from the ground up. I started down that road when I was looking down the road to store my bicycles vertically on one end of the garage. it is easy to pick up a regular bike and place it on a hook. But a long tail electric cargo bike or even my regular electric bike becomes unwieldy. So like you, I started thinking about how to repurpose some old hub motors to wind up a rope. Then it occurred to me that there are inexpensive solutions readily available that would be much better. Would one of these inexpensive hoists be a better/simpler solution for you?

https://www.amazon.com/VEVOR-Electric-Factories-Warehouses-Construction/dp/B0DDXS7P8V/ref=sr_1_4?sr=8-4

As for cells corroding, if kept indoors this seems like an unlikely problem. My ebike cells look fine after 5-6 years on the bike. They were stored in a plastic case much like those sold by Pelican. The case is water resistant - or was until I drilled holes in the side for the wires to exit.