Time based charging / computer controlled charger?

Hi,
I'm nearly done building a bike that I plan to use for commuting 5 days a week. I'm using LiPoly packs and would like to treat them nicely to extend their lifespan. My goal is to hook the battery up to the charger when I return home at nigh then I want to have something smart that says something on the order of "If it is past 0800 (when I leave for work), and the battery is below 3.8 volts, start storage charge to 3.8V", then later "If it is 0500 and the battery is below 4.2volts, then charge the battery to 4.2V".

The idea is that when home I'd like to keep the battery at storage voltage at 3.8volts then immediately before I leave for my commute charge it up to 4.2 volts and have it be at that voltage for only a short period of time.

I can't find anything that does that. I can do it using an arduino connected to the control panel of my charger with only a little effort, but if something already exists I'd like to know.

-JIm

I've never seen anything like that. There might be a *teensy* little market for them if they did exist, but probably not many.

Something that could talk to Grin's Satiator over it's serial port would be interesting, but it would require a change in the way the port is used to do it. (presently it has to be in the "connect to PC" mode, meaning it isn't doing any charging/monitoring).

A number of RC LiPo chargers do have active serial (or USB) ports all the time, so you could monitor and possibly control them that way (not sure how many actually can be controlled by it; the couple I've had only seem able to output data on the port).

What you need is the bulk charger with three voltage switch. EM3ev is where I got mine. Or get the satiator, the most sophisticated charger you could get.

At night, charge to 4v per cell, before you sleep. Then in the am, top off to 4.2v, in about the same time it takes to shit shave shower and breakfast.

If you mean lithium cobalt pouches, like RC lipo,, under no circumstances do you charge them while you sleep, unless they are outside in the barbeque grill.

Thanks for the suggestions, I'll take a look.

As to the charging at night. I charge on a concrete floor in "fireproof" ventilated bags. I've been using RC LiPo's for almost 20 years, I was one of the first to use them in RC airplanes and used to be considered and expert in their use for RC's. I've had one cell explode during charging in that time due to an error on my setup of the charger. Since I started using balance chargers only (I ONLY balance charge) I've never even had a hint of an issue. I'll use these very carefully as I always do, but I have no issue starting a charge at night.

-Jim

Ok,, we never know if we are talking to a 14 year old, who shares the bedroom with a baby, and wants to charge next to the crib.

No doubt you have no intentions of keeping a dinged pack in service either. That's the mistake so many newbies make.

I'm not afraid at all to bulk charge, when I'm undercharging. My charging station is inside a fireplace, but I still want to be awake if the smoke alarm goes off. Or better yet, smell a hot pack even before the alarm goes off.

I just have a charger connected to outlet timer. So one is at home the other is at work. One starts charging in the morning the other just before I leave.

Elaborating on another post i think.

maybe not suited to lipo, for regular ebike paks, a cheap mechanical outlet timer and the humble charger included with the battery can achieve similar outcomes for most.

As the 24 hr dial rotates, "pegs" you decide on, toggle power on and off as it rotates thru 24 hrs.

Your commute is ~regular, as is the ~recharge time needed.

So, e.g if total 4 hrs are needed, perhaps 3 hrs are on, starting when u know u will be home by each night, and on again for 1 hr also b4 u leave each morning.

So it spends most of the nite at an approximate partial charge u decide (IE, at its equivalent approximate voltage), and it only gets fully charged soon before you ride it, which is ok - so long as not sitting fully charged for too long.

A little under charged isnt a worry either, assuming you have over-allowed some capacity for your commute, as one does.

I intuitively kinda like unplugging it in the last stages of charge - the batteryS chemical reaction all warmed up and ready to go. Manuals seem to warn u to go easy on the throttle when just starting, for this reason i assume.

If it would help baby the battery, the option is there to toggle on/off in 15 min increments on these sub $20 timers, for a gentle/gradual/slow majority recharge.

I would like to contrive things so i knew for sure it would be in the on setting when i arrived home, to minimise any damaging time spent idle at low charge.

Maybe stretch that initial charge period over a longer period, using 15 min on/off increments as above, so when u get home at ~random times, simply plug in, and u know it will be getting some charge within 15 minutes.

Fortunately, the last charge stage is mostly fiddling about with balancing. Not much charging is done. You neednt balance on each charge, so prematurely unplugging some mornings, is still to all intents, a full charge.

So yeah, its not sitting around at low or full charge for long - the point of the exercise. its cheap and simple - let the battery maker figure out the best charging parameters.

On weekends, i think i would prefer to decide the final hour of charge myself. charge to 75% as above, & i plug in a bit b4 i set out, if i set out.

as the nother poster suggested, u could have another timer for work.

u would of course need to be sure a power outage during the night didnt pass unnoticed, and foul things up.

dogman dan said:

Ok,, we never know if we are talking to a 14 year old, who shares the bedroom with a baby, and wants to charge next to the crib.

No doubt you have no intentions of keeping a dinged pack in service either. That's the mistake so many newbies make.

I'm not afraid at all to bulk charge, when I'm undercharging. My charging station is inside a fireplace, but I still want to be awake if the smoke alarm goes off. Or better yet, smell a hot pack even before the alarm goes off.

Click to expand...

Right on, that makes sense. I really like the idea of charging in the fireplace.... that works well for me, I'm going to do that.

Looks like there really isn't something like this in the wild right now. I'm going to build one, we'll see how it goes.

-jim

Shouldn't be too hard to do that with a simple circuit: E.g. http://www.instructables.com/id/Voltage-Controlled-Switch-1/

But since I have a Comp Sci. background rather than a electronics background, I'd rather buy a Raspberry Pi with an external control board and program that up.

Well, catch 22 of the fireplace is that in the coldest part of winter, I want the darn flue damper closed.

For the winter, I move all my lipo to an old fridge in the yard. But in summer, it would fry out there. I tend not to ride the dirt in winter, and not use the lipo most of the colder season anyway. Soon as the passive solar heat can keep up,, lows in the 50's, then I reopen the damper on the fireplace. All winter,, I charge my bms equipped manganese packs in the garage, I have little fear of them going off.

So I get what you are trying to do here, have you cycle life increased by not leaving the batteries full all day. You should also realise that discharging to store down at 3.8v is still doing a discharge i.e. cycling the pack.

what would be ideal is a charger which knew your schedule for commuting and knew when you were going to use the bike (battery attached device) with enough time before the start of charging so it was ready each morning and again at work before commuting home if you are allowed to do that.

So my friend and fellow ES er on here Bobc has created a balancing board with relays which can top or bottom balance upto 26s lipo or lifepo4. I mention this cause it might be good as a storage charge leveller and will have minimal or almost not cell bias and drain unlike a bms or balance charger, but you could keep it attached to the pack. Its based on an Arduino so you could then put a rtc (real time clock) shield or wifi onto it and send commands based on your calendar, i.e. whether you will need you bike that day.

anyway the issues with this are essentially it is unattended charging and discharging, posts above covered this.

another way to do it would be circuit bend a balance charger but I don't know what the drain is leaving one attached. I.e. having the Arduino or Raspberry pi use reed relays to navigate the menu on a charger. I would say this approach isn't great as it has more complexity and if things can go wrong they will.

Also the BST 900 ming-he dcdc has a mode where it can turn on its current limited charging output when supplied with an input supply so you could have Bobc's relay Arduino board work in tandem with a relay turning on the dc supply to a ming-he (or your own non balance charger) anyway just some food for thought.

kingjamez said:

Hi,
I'm nearly done building a bike that I plan to use for commuting 5 days a week. I'm using LiPoly packs and would like to treat them nicely to extend their lifespan. My goal is to hook the battery up to the charger when I return home at nigh then I want to have something smart that says something on the order of "If it is past 0800 (when I leave for work), and the battery is below 3.8 volts, start storage charge to 3.8V", then later "If it is 0500 and the battery is below 4.2volts, then charge the battery to 4.2V".

The idea is that when home I'd like to keep the battery at storage voltage at 3.8volts then immediately before I leave for my commute charge it up to 4.2 volts and have it be at that voltage for only a short period of time.

I can't find anything that does that. I can do it using an arduino connected to the control panel of my charger with only a little effort, but if something already exists I'd like to know.

-JIm

Click to expand...

A simpler more pragmatic way of putting the question would be:

How can i charge my battery to a point where it can be fully charged in ~an hour, ie. having decided to use it, in the time it usually takes to get organised to leave?

If u dont always get it 100% charged - meh. Most have spare capacity or maybe pedal a bit more & go a bit easier that trip..

I think this topic brings an interesting question to the table, similar like @whereswally606 said

You should also realise that discharging to store down at 3.8v is still doing a discharge i.e. cycling the pack.

Click to expand...

- Is it worth to discharge to storage voltage, if the cell is left at around 3.9-4.00 volts? Let's say for 2weeks or more?

I think it would give some stress if cells are really kept at 4.1-4.2v for prolonged time but is it worth it to always discharge them to 3.7 and then charge to 4.1-4.2v before using them?

And - so the final question would be - what would be the time, when cells should be discharged / charged to the storage voltage? Is it a week? a month or more?

Though for this there should probably be some scientific experiments and analysis to determine which action is worth it in the long run (charging to storage or just keeping fully charged, if the battery is gonna be used soon enough. I think I will try to look some info up on the net and If I find something I will publish it here.

There are scientific experiment on calendar aging in function of State of Charge (also depending on storage temps ans chemistry :
https://endless-sphere.com/forums/viewtopic.php?f=14&t=86777

Let's say is best to store a 50% SOC if you were to store at 6 months...

The thing that's hard to grasp is, what is 50% SoC? For instance, a brand new VTC4 cell wight have 2200 mAh capacity, thus 50% SOC is at 1100 mAh, at 3.60V (which happens to be nominal voltage here)... But in 5 years the cells might be 1600 mAh, so 50% SOC is at 800 mAh...
Is it that in both cases, 50% SOC will be at the same resting voltage or not ? (ex : at around 3.6V nominal). Or will the nominal voltage change according to capacity decay over the years ?

Matador said:

There are scientific experiment on calendar aging in function of State of Charge (also depending on storage temps ans chemistry :
https://endless-sphere.com/forums/viewtopic.php?f=14&t=86777

Let's say is best to store a 50% SOC if you were to store at 6 months...

The thing that's hard to grasp is, what is 50% SoC? For instance, a brand new VTC4 cell wight have 2200 mAh capacity, thus 50% SOC is at 1100 mAh, at 3.60V (which happens to be nominal voltage here)... But in 5 years the cells might be 1600 mAh, so 50% SOC is at 800 mAh...
Is it that in both cases, 50% SOC will be at the same resting voltage or not ? (ex : at around 3.6V nominal). Or will the nominal voltage change according to capacity decay over the years ?

Click to expand...

Im no expert but I think loosing 400mah is not that big difference.. ok it is 400mah / 2200mah = 18.2 %.. but Im not sure is it needed to ''go for maximum precision in this

We all know that in general if you keep the cells at max voltage it is no good in long run.. so recommended is actually at around 3.9v or below that.. So if you really wish to store for long time.. 3.8v might as well be your point..

Okami said:

We all know that in general if you keep the cells at max voltage it is no good in long run.. so recommended is actually at around 3.9v or below that.. So if you really wish to store for long time.. 3.8v might as well be your point..

Click to expand...

Yes, it makes sense.... But I ask the question because in my case, I'm doing lots of testing on slightly used cells (Makita VTC4), testing ONE cell at a time for capacity (i want to capacity match the cells in the 14P10S pack I want to build)... I'm doing single cell capacity testing with a Foxnovo 4S digital charger (see pic 1 below).

The thing that sucks with the Foxnovo 4S charger is that, as soon as the capacity test discharge cycle finishes for one cell, the charger automatically restarts the recharges cycle up to 4.20 Volts for that cell. So for example when Cell #1 finishes discharge cycle at 12:00 it then automatically engages into recharge cycle at that precise moment... then Cell #2 at 12:21 then cell #4 at 13:01 then cell #3 at 13:02... (times varies from cells to cells). So I cannot really stop them all at the exact same State of reCharge (in mAh) or the exact same voltage to just put them to storage (unless fully charged, which is 4.205V ± 0.016V at 95% confidence interval (n = 85)). I can't just stay in front of the charger, waiting for each cell to reach 1037 mAh (50%SOC), then pulling out each cell one by one at different moments of the day... Not practical ! So I just let them recharge completely to 4.20V. But I don't want to store them at 4.20V for two months. So I found a solution.

I have 290 cells to test, that will take me two months! And i can only test in batches of 4... In my case consistant reproductible SoC is important because I also want to test each cell with load testing for DC-IR determination, and also, I'm using very high drain/low IR cells,... So when I will parallel cells to make battery pack, if there is a significant voltage difference before paralleling cells, I could damage some cells. For exemple, if I add one unbalanced cell resting at 4.100V (20 mOhms) to a string of 9 paralleled cells sitting at 4.200 V (2.2 mOhms) to make my battery 10P, then there will suddently be 45 AMPS dumped in the one unbalanced cell sitting at the lower voltage (4.100 V). One cell "charging" at 45 amps when the manufacturer's specified max charge rate is known to be 4 amps/cell maximum is certainely not a good thing !

So I need reproducible conditions 1) to make DCIR testing in same conditions, and 2) to parallel my cells without damaging them (the VTC4 are low internal resistance : 12 mOhm impedance at 1kHz AC or around 20 mOhms DC resistance).

To minimise capacity loss of my Sony VTC4 NMC cells, I wanted to discharge them to around 50% SoC before storing them, then I can capacity test some more cells....
Why 50% SoC ? well here's why (see Pic 2).
View attachment 18
And in my case I determined that, at least at very low current draw of 0.2C (this is the closest simulation I could find to the unloaded "float voltage" conditions), 50% SoC is 3.65 Volts (or 1037mAh/2075 mAh @ 0.2C from 4.20 to 3.00V discharge cycling) according to this graph :

So what I did is one afternoon, I took all my capacity tested cells, sitting at around 4.20 volts, and I checked their voltage precisely with one CellLog 8M device...

My "precison" voltage tests results with the CellLogs8M are here in column "T" of this Excel sheet : View attachment Makita VTC4 cells from doctorbass.xlsx
Then I made groups for different cell voltage :
Group #1 : 4.213V
Group #2 : 4.209 V
Group #3 : 4.207 V
Group #4 : 4.204 V
[...]
Group #13 : 4.179V

Then I used some spring modules that I fabricated myself to put all cells in one group in parallel. I call it « The Matador Spring-loaded Module » 8) :

Then I used a dummy load of around 0.5 to 0.6 amps and another optional, but faster, dummy load at around 4 amps (to discharge Group #1 sitting at 4.213V) while monitoring voltage slowly going down with my CellLog 8M . As soon as Group #1 was discharged to the exact same voltage as the Group #2 (4.209V), I then paralled Group #1 with Group #2. I then continued discharge while still monitoring voltage. As soon as Merged "Group #1-2" reached voltage of Group #3, I added Group #3 in parallel connection. And so on and so forth, parallel adding more and more Groups with lower voltages until reaching last Group #13 with the lowest voltage. I did this for 130 cells yet. Lastly Merged "Group #1-2-3-4-5-6-7-8-9-10-11-12-13" was siting at 4.179V IIRC.

When all my 130 cells were finally paralled but still sitting at around 99.5% SoC. I decided to discharge them to 50% SoC (I think 3.65V for my type of cell) and put them to storage (While still being paralled so they stay ultra well balanced between them). It looks like this :

Paralleled groups :



My homemade parallel connection bussbars : Flattened 1/4 inch copper pipe with hammer, drilled holes, sanded with wetsandpaper, nickel plated the copper with nickel acetate ! Still shiny (unlike bare copper wich tarnishes with time) !
View attachment 5
Storage voltage of 130 cells sitting in parallel :

My dummy load setup (using halogen lamp):

a) Just using the 12V20W Halogen bulb as the dummy gives me 0.54A load @3.651 V.

b) when i use the complete light as dummy load it pulls 3.84A @ 3.650V for faster discharge time ( Bulb dissipates 2W in this setting but it's paralleled with the lamp's transformer coil which dissipates 9.3W here (3.30 A @ 3.650V) ). But I dont let the coil on for too long as it gest hot after one hour.... I'm afraid varnish will end up melting, shorting the coil (and possibly making low amp short to my battery making the wires melt).

Well long story short, I now have balanced 130 VTC4 cells to storage voltage (3.652V@rest).
When I finish up capacity testing of more Makita pack I have to tear up (i'll reach 290 cells), i can then go on and do some DC Internal resistance testing... to match cells by resistance too !

Matador

PS : In case anybody want to know how I made the matador modules... It was initially inteded to be used as a solderless battery : https://endless-sphere.com/forums/viewtopic.php?f=14&t=57810&start=150#p1213147
Few more pics of that old battery just for nostaliga and sharing (1 picture is a thousoud words) :

View attachment 1
View attachment 2

The graphs show that storage at 0% SOC would be best which i believe is referring to 2,5V.
Can this be true?
otherwise it seems like 10-50% would be optimal.

madin88 said:

The graphs show that storage at 0% SOC would be best which i believe is referring to 2,5V.
Can this be true?
otherwise it seems like 10-50% would be optimal.

Click to expand...

I also noticed that... I always wondered if there is a reason for not storing the cells at let's say at 5% SoC (headroom for the steep dive when aproaching 3.0V).
People seem to say it's best at 50 SoC.

Maybe the 50% SOC number is a good compromise so that people that are not so well aware of BMS potential pitfall (overdischarging your cells to voltages making the unrecoverable useless bricks) dont spoil their costly battery.
For myself, I will not connect any BMS while storing.... So less than 50% SOC would not matter, unless there is another reason that I'm not aware of....
But these graph shows otherwise !

madin88 said:

The graphs show that storage at 0% SOC would be best which i believe is referring to 2,5V.
Can this be true ? Otherwise it seems like 10-50% would be optimal.

Click to expand...

I'm sure there's some battery expert (electrochemist ?) somewhere on this forum who knows the exact answer....