DIY Electric Car Forums banner

Pulse charging lithium batteries

2921 Views 19 Replies 11 Participants Last post by  BrunoEV
Hi there, im designing a 10amp charger and im wondering how much voltage and current ripple the batteries are ok with?
I know sla arent bothered much but lithiums i dont know about, anyone got any info or could point me in the right direction? Google hasnt helped D:
Thanks
1 - 20 of 20 Posts
Google probably hasn't helped because you generally can't find answers to questions that aren't usually relevant to ask. It's not in a spec because it's not really relevant.

Voltage ripple, I mean, infinite, I guess.

You could feed it half-wave rectified AC and it would be fine. The battery would just only charge during the parts of the cycle that are above the battery's voltage level. I.E. If the batteries were "full" at 42 volts, and you're feeding it 40 volts that spikes up to 42v every 100th cycle, it's going to take 100x as long for the battery to reach 42v than it would have if you fed it flat 42vdc.

Just note that the batteries will charge to the peak that they're given (even if at a slower rate), so if you feed them 41vdc with a 3v ripple, you'll be overcharging your batteries to 41+3=44v instead of 42.

As to current ripple, well, current will just mirror voltage.

You can think of a battery as a big slow capacitor I suppose, voltage curve aside. It will soak up nearly bottomless current nearly instantly if all you're building is a 10a charger.

Discharge works the same. You might demand energy from the battery in all kinds of wonky manners, it doesn't care if the current wanders all over the map.
See less See more
Well...google does come up with a lot!

just search for:
ripple current lithium ion battery

one of the results:
https://www.sciencedirect.com/science/article/pii/S030626191630808X


apart from the effect on the battery...don't forget EMI / EMC
The only reason I have found to pulse charge lithium is to heat a frozen battery

Why do you want to pulse charge?
I dont want to pulse charge, i want to know how smooth i need to make the output
Lithium batteries (somewhat like other chemistries) tend to be quite forgiving. I guess it depends partly on your charge C rate. In constant current mode is current ripple really going to be that much?

Even if it is, if you're charging a 30Ah pack for example then ripple isn't going to significantly affect the cells in a negative way - and perhaps it may even reduce the recoverable solid electrolyte interphase layer - which isn't a bad thing
Some of the now defunct systems used to pulse charge at the end of charge (during balancing) other systems pulse charge to give DCIR for the BMS to calculate SOH
My advice would be don't overcharge, don't let the cells over discharge - and keep them as near to a temperature as you'd like to be kept in
Briefly looking over the link above suggests SEI is built up when cycling below 10hz, though it's unclear as to whether the SEI is recoverable or not
I dont want to pulse charge, i want to know how smooth i need to make the output
.

From that paper it looks like you just need to keep it below 254Hz
Yep, fig 3. Any high value, superimposed AC currents will heat up the battery signifantly and unnecessarily.


EDIT: (Unless you want to pre-heat the battery in wintery conditions).



Low frequency pulse charging can be used to replace the V portion of the usual charging profile. It will also reveal the battery response to a step function, allowing monitoring equipment to detect an imbalance.
I'm trying to find a solution for charging 24 V Tesla batteries without adapting the settings for the lead acid batteries in my charger, becaus I want to charge those as well. This would result in a far too high charging voltage of 29 Volts. By adding a electronic relay in the charging circuit that is triggered by the BMS I want to create a Pulse Width Modulation charge process. When the battery is indeed reacting as a big slow capacitor I expect that this way of charging will not harm the battery.
Is this analysis correct?
Details: 2 Tesla S batteries, approx 10 kWh. maximum charge current 80 Amps, but this value can be lowered. Used in off-grid residential system
Your comments will be very much appreciated.
Please don't do this!

Charging should be done with a set voltage corresponding with the target voltage.
The BMS is there to 'pull the plug' when charging / discharging is not according to the set limits. (and to adjust current when temperature is too high / low)

In your scenario, when something doesn't work, you have your batteries exploding...not a pretty sight.

Other thing is to charge well you really need a CC-CV charge, and turn of charging after reaching 100%SOC.
By adding a electronic relay in the charging circuit that is triggered by the BMS I want to create a Pulse Width Modulation charge process. When the battery is indeed reacting as a big slow capacitor I expect that this way of charging will not harm the battery. Is this analysis correct?
It'll work.

If something goes wrong and the relay doesn't shut off, you'll start a fire and burn the vehicle and garage to the ground.

That said, I suppose that's true of just about any switchmode charger. Something can go wrong and the output might get the full voltage or a higher voltage than it should.

I'd say, don't let it be the reason you don't do something, but, plan on upgrading soon. Nothing as permanent as a temporary solution that's good enough.
I'm trying to find a solution for charging 24 V Tesla batteries without adapting the settings for the lead acid batteries in my charger, becaus I want to charge those as well. This would result in a far too high charging voltage of 29 Volts. By adding a electronic relay in the charging circuit that is triggered by the BMS I want to create a Pulse Width Modulation charge process. When the battery is indeed reacting as a big slow capacitor I expect that this way of charging will not harm the battery.
Is this analysis correct?
Details: 2 Tesla S batteries, approx 10 kWh. maximum charge current 80 Amps, but this value can be lowered. Used in off-grid residential system
Your comments will be very much appreciated.
I find it hard to understand what you are attempting to do...... If you mean you are adding a circuit that takes the output of the charger and generates a PWM pulse to lower the voltage to 24 then yes that will work. I am not sure I would want to switch a relay fast enough to produce the desired effect but you might be able to find some that switch fast enough with reliability (doubtful) Most designers use Mosfets or IGBT's for this function in which case you are better off buying a charger for you purpose than trying to recreate the wheel.
I am not sure I would want to switch a relay fast enough to produce the desired effect
Oh, I interpreted it as, he would just feed it the 29v from the lead-acid charger, but have his BMS trigger the relay off when his batteries reached 24v. Not, using the relay to clicky-clack average 24v on a 29v charger... which will not work, as it will still charge the batteries to 29v, just at a slower rate.
Not exactly what your looking for as they don’t reach the voltage your looking for but this is all I have referenced as I use these to rescue bricks at cell level.
https://www.ebay.ca/itm/DC-DC-Step-...e=STRK:MEBIDX:IT&_trksid=p2057872.m2749.l2649
As long as you are supplying a dc voltage 5v above the output then they taper and cut off charging, they are not accurate, some were reading .26v higher than they were outputting but none read under which is important, but they are super stable and once they were set they worked perfectly tappering nicely at my max current. I’m sure there’s some dc/dc converters that are higher output or even an ebike lithium charger that will safely do what you want, unless your just playing with PWM to see if it will work, if so get a load of cheap laptop cells and make a simple 24v pack and try on those first in a safe area (an out door BBQ is a favorite).
See less See more
I'm trying to find a solution for charging 24 V Tesla batteries without adapting the settings for the lead acid batteries in my charger...
Just to be clear... you know that there are no "24 volt" Tesla modules, right? They are 6S modules, so their nominal voltage is 6 x 3.75 V, or 22.5 volts. if you are charging to 4 volts/cell, then yes that's 24 volts... but don't think of these modules as in any way equivalent to a nominally 24 volt lead-acid configuration.
put a voltage sense on the pack that turns the charger off once the voltage reaches a set point, in your case 24v


http://ghurd.info/
this is used for wind turbines but u could use it in this case aswell..


if you are any good with arduino ill send u the code i used for my charger turn off relay
Thanks a lot guys for all the useful advice and suggestions. I am aware that when anything goes wrong this will end ia a fire, so there must be an extra safety system based on another principle like the one suggested by Arklan.
Testing PWM on a simple 24V pack is also quite interesting, as well as looking at a stepdown converter.
All in all, plenty to think about!
After considering all the options I opted for the safe but more expensive solution by choosing a 230 VAC charger dedicated for charging and protecting the Tesla batteries. The charger is linked to the BMS to keep the batteries healthy and safe.
1 - 20 of 20 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top