| Electronics > Projects, Designs, and Technical Stuff |
| Regulations/standards for Lithium Cells and charge circuit in commercial product |
| (1/2) > >> |
| newto:
Hello, I'm trying to sort out what the regulations/requirements/restrictions/certifications there are for commercial products containing Lithium Ion Cells (ie: 18650 cells). I'm working on updating an old product that currently has a large lead acid battery, and want to upgrade the design to use Lithium cells instead for multiple obvious reasons. The idea is have 3-5 18650 cells in parallel with a simple 5V charge circuit (if I understand correctly, cells in parallel don't need balancing?), and then boost to 5V to power the components (old product ran on 12V, but modern equivalent components are designed for 3.3V-5V). My issue is I'm having trouble figuring out what we need to do in order to certify the new product for sale. The old design has an internal Lead Acid Battery, and an external certified off the shelf charger, so we didn't need to do any other testing other than for our CE mark. But now we want the Lithium Cells with our own charge circuit, but plan to keep an external off the shelf power supply (either 5V barrel plug, or USB if we can keep the power requirements low enough). Everything I can find so far, UN 38.3, UL 1642, IEC 62133, seems to be about cells/batteries, but if I just take properly certified cells and plug them into a board (I don't even plan to weld them, just snap in holders, maybe some zip ties or silicone), do I need to do any additional certification? We have international sales currently, so I imagine this could be a bit of a mess? |
| Peabody:
I have my doubts about your conclusion about balancing, and I know nothing about certification, but just want to say that your design should include a "load sharing" circuit that will let you safely power the device while charging the batteries at the same time. http://ww1.microchip.com/downloads/en/AppNotes/01149c.pdf |
| newto:
Yup, that was my plan (well, actually that way simpler than I thought). Everything I've read so far about parallel Lithium cells is that they will "self balance", so if some cells fill faster than others, they will draw less current, while the empty will draw more. Just need to worry about having the right cutoff current to tell when they're all full. |
| jbb:
It's a bit hard to say exactly what regulations might apply. UN38.3 etc. certainly apply if you want to ship anything! Are you looking at LiIon or LiFePo4 chemistry? In terms of cell arrangement: * I understand you can parallel cells, but they should all be 'the same' type of cell (i.e. same manufacturer and part number), and should only be placed in parallel once they have the same cell voltage to avoid damaging current flow. * I would be nervous about using holders, because their contacts aren't as reliable as welds. Also, what if one of the cells looses contact while fully charged, and makes contact again once the rest of the pack is discharged? * Once safely paralleled, you shouldn't need balancing because all cells will share the same voltage * Try to keep all the cells around the same temperature so the cells at the hot end don't do all the work * You should have a temperature sensor near the pack (preferably in the pack) so that you don't charge when it's too hot (fire risk) or too cold (can wreck cells) * For LiIon, I thoroughly recommend a battery protector circuit (over voltage, under voltage, over discharge, over charge) for safety, in case your charger chip goes haywire or something short circuits your power rail * All in all, I suggest making a pack with all the cells welded together, a protector PCB and a temperature sensor all built in. In terms of charging: * You seem to be looking at 6Ah (22Wh) or even 10Ah (37Wh) of cells * USB charging is a bit of a pain * USB spec allows supply voltage down to 4.75V * USB spec allows for cable drop, so you might only get 4.4V * If using USB for your '5V' rail, make sure your product can handle a poorly regulated supply. If not, deploy your own 5V regulator (e.g. boost converter). * Assuming a 900mA input current limit, that's approx. 4W input that you can rely on. Could be a 10 hour charge time - is that acceptable? * For batteries this size, you want a switch-mode (e.g. buck) charger, not a linear one. * As Peabody says, it's best to get a charger which has a separate system output pin and battery charge pin. This is often called 'Powerpath' or similar. This is because you don't want the charger to confuse the system load with the real battery charge current. Had a nightmare with this at work once... * Charging to a slightly lower voltage (e.g. 4.1V instead of 4.2V) may reduce your energy storage per charge a little, but increase the service life in cycles On discharge: * Ending discharge at a slightly higher voltage (e.g. 3.3V instead of 2.5V) may reduce your energy storage per charge a little, but increase the service life in cycles. Also, if using 3.3V internal rails you might be able to move from a buck-boost converter to a simple buck converter... * LiIon cell internal resistance goes up a bit as the cell gets colder (but probably not as bad as lead acid :) ) You'll also have to check the shipping rules. Generally, getting your batteries in bulk will be a pain in the ass. But you should be able to ship a product with a battery in (and possibly one spare battery) OK. I'll look up the right shipping regulations later. Talk to your shipping companies about it now. |
| aix:
For USB charging, it might be worth noting the existence of QuickCharge-type protocols, as well as USB Power Delivery. |
| Navigation |
| Message Index |
| Next page |