new larger lipo seems worse, is it?

jimmyhackers

10 kW
Joined
May 11, 2015
Messages
600
i recently swapped out my 4 year old 13ah rc car lipos for brand new 20ah ePLB-CO20B cells. both packs are 3x6s in series for 72v nominal.

only thing is.....my new cells sag more and seem to have a similar amount of capacity/range.
i feel like ive sort-of been ripped off.

have i bought duff cells? or is the chemistry between the two so different that i should expect that kind of voltage sag and need to run my new cells to a much lower voltage to get get their actual capacity out of them?.

thanks in advance for the help
jim
 
Vendor specs for capacity testing should be published on the data sheets.

Usually 4.2V held CV until 0.05C at the top

and 2.5V at the bottom

with a CC rate at 0.1C.

Quality varies enormously from one maker to the next
 
http://www.ebaracus.com/sites/default/files/2012/12/EIG-ePLB-C020-Datasheet.pdf heres their datasheet.

my bikes max discharge should be around 21amps (1500w@72v)

the internal resistance of each cell on the new bat is about 2mohm or less, on my old bat is about 21mohm.

my old batt charged to 4.2v a cell and is empty around 3.6v a cell. this takes about 8-9ah.

my new batt is charged to 4.15v and is empty supposedly at 3.0v a cell (not gonna take em this low)

after the same ride the new cells get down to about 3.75v which is exactly storage charge. again this is about 8-9ah.

basically the voltage sag is worse under use/load....despite newer bigger cells.

only thing i can guess is that the new bat is a single pouch cell per cell, meaning one neg n one possitive tab out of the pouches.
the old batt is two pouch cells per cell, so twice the tabs in n out of the cells. menaing more potential current delivery n less sag.

just makes me wonder......these cells are rated upto 5C continuous discharge.....so 100amps constantly....atm it fells like they cant handle 1C.
 
Vendor ratings for discharge C-rate are 99.99% of the time grossly optimistic, and 90% of the time pure fiction physically impossible.

Batteries designed for high energy density need to be poor at power discharge.

But maybe you just got a bad batch?

Or maybe the maker sucks, never heard of them myself.

Link to the seller listing?
 
Based on the datasheet, these cells look like they begin to get seriously stressed beyond 1C. You don't want to see more than 0.1-0.2v drop during load per cell, ideally.

2020-11-27 12_56_50-Window.jpg

Per the included graph, it would seem prudent to run this pack at an average of 1C with up to 3C burst.

If you're running them at 1C and getting really bad sag, then two things are possible:
1) You're running them at colder temperatures than rated, which can cut the discharge rate of any lithium battery down to 25-50% of what it normally is.
2) This discharge graph is a complete lie.

By the way, internal resistance changes according to state of charge. A battery has the highest IR around mid-charge. At high and low charge, IR is increased significantly. So in comparing to your old lipo pack.. make sure you are comparing both packs together at mid charge.
 
jimmyhackers said:
http://www.ebaracus.com/sites/default/files/2012/12/EIG-ePLB-C020-Datasheet.pdf heres their datasheet.

my bikes max discharge should be around 21amps (1500w@72v)

the internal resistance of each cell on the new bat is about 2mohm or less, on my old bat is about 21mohm.

my old batt charged to 4.2v a cell and is empty around 3.6v a cell. this takes about 8-9ah.
L
my new batt is charged to 4.15v and is empty supposedly at 3.0v a cell (not gonna take em this low)

after the same ride the new cells get down to about 3.75v which is exactly storage charge. again this is about 8-9ah.

basically the voltage sag is worse under use/load....despite newer bigger cells.

only thing i can guess is that the new bat is a single pouch cell per cell, meaning one neg n one possitive tab out of the pouches.
the old batt is two pouch cells per cell, so twice the tabs in n out of the cells. menaing more potential current delivery n less sag.

just makes me wonder......these cells are rated upto 5C continuous discharge.....so 100amps constantly....atm it fells like they cant handle 1C.
Are you really meaning voltage “sag” , or just voltage change as capacity is used ?
Have you measured this “Sag” ?
Those EIG cells are known to be a quality product , used in EVs and industrial applications, so if you did buy “new” cells, from a reputable seller, i doubt there is a problem with them.
But do the tests yourself,..Capacity/voltage,..IR vs capacity (or at least mid SOC) ...voltage sag vs applied load.
EDIT .. to add.. do the IR and “sag” tests on the full pack assembled....that will cover any effects from tab arrangement and any issues with cell to cell commections !
What you have shown is that for the same range/ride, both packs use the same Ah, ( i assume you are quoting the recharge Ah ?), which is normal, and you still have 11Ah capacity left in the new cells after your ride.
So, what exactly is the problem you think you have ?
 
the sag im experiencing is closer to 5 - 7v over an 18s battery that's 0.27v to 0.38v per cell.

i have a voltage meter in my throttle that displays voltage to xx.x (one decimal point) i also temporarily plugged in a multimeter to see if this was accurate (it is)

i test the capacity/resistance through my isdt t6 lite charger (this seems fine).

ive seen it as low as 60 ish volts on a ride, but it gets back upto 67.5v (3.75v per cell).

either way its more sag than my older batteries (2-3v overall).

i can understand that im not using the full capacity of the battery, maybe 40-50% so i cold get a lot more range/use with a lower discharge voltage per cell compared to my old batteries.

but its the sag thats worrying me. i was expecting less, not more.
ive seen other people praise these cells and use them to good effect, i just wonder what (if anything) im doing wrong.

everythings the same, only thing i changed was the battery.

nothings getting hot or even warm so i dont think its a duff connection. big thick copper bus bars, 14awg wire soldered to those for main in n out, t60 connectors.
my main wiring looms always been 14awg.

only things i can think is losses through having each cell bolted rather than soldered. also that my old batt has two pouches per cell meaning more pos n neg tab area into them.
 
If you are dropping 7.0v on a 21 A max load,...that suggests a “pack” IR of 333mOhm. Which is totally out of wack with the <54 mOhm which the 18 cells would have in total.
So... yo either have one or more dud cells, ..or you have some poor connections in the pack assembly.
You will only know what the problem is by detail testing everything.
 
I can feel the differences in a 1mOh, vs 2mOh, vs 3mOh cell pack. All day.

I have large format lipos, can do over 10C easy peasy. I drop 0.9v per every 1000w.. so for a 8000w hit.. I drop 7-8v. They are 1.2mOh ea.

You need to know the IR of the pack, line, cells, busses, empirically, cell by cell. To find the bottlekneck.

I know HK lipos all come in around 2mOh... and the EIG being 3mOh.. you see.

I can feel the difference a 60mOh pack ( Hk lipos) vs a 32mOh pack... ( Chevy) One whumps ya. The Chevy. The only good thing about Hk lipos are the voltage stiffness.. the voltage stays higher longer, with the SOC... they like to give 3.8v then die fast.. the Chevy lipos are nice, longer capacity curve, down past 3.4v.. before they fall flat... and much more whumpf.. HK lipos are no longer fun for me.

www.recumbents.com said:
Yes, it's going to be hard to find a battery pack that beats LiPo. I had the same issue after replacing my 48v 10ah LiPo pack with a 20ah 48v pack. New pack is a safer formula (Samsung 25R) but more saggy.

Warren
Chevy Volt cells tromp HK lipos. I can show you sideby side datalogs.

There are others, sub1mOh or around there. More in the future too.. on there way to junkyards around you...

Ya just have to schoose the right cell.

Double the EIG cell up ( 2p) for a 1.5mOh group.. and you will be ( near ) the power of one of my Chevy cells. Entire pack IR is 32mOh to about 40mOh on a cold day. As reported virtually with the Ca3 managing the traction batt.
 
the only thing i can think is the copper bus bars may have a very thin layer of oxide on them which is contributing to the extra resistance....

its annoying as all the cells appear to be behaving identically. there's not an obviously odd one out.
all individual cell resistances and voltages are identicle after every ride/charge.

im waiting for 10awg wire to replace the 14awg before i tear them down to refit. when i do this i will take some scotch bright to the bus bar contacts and see if it helps in any way.

ive also heard lipos need about 10 proper cycles before they get into their conditioned good performance stage/part. is this true?
is it possible my only partial 40-50% use of the battery top end capacity is effecting them adversly?
 
DogDipstick said:
I can feel the differences in a 1mOh, vs 2mOh, vs 3mOh cell pack. All day.
"..."........
I know HK lipos all come in around 2mOh... and the EIG being 3mOh.. you see. ...........
Very true DD, ..but that is not jimmys issue, ...
His original lipo had a HIGHER IR than the new EIG cells??
the internal resistance of each cell on the new bat is about 2mohm or less, on my old bat is about 21mohm.
So if the info is accurate, he should be seeing less “sag” on the new cells, ...hence, his problem is not the cells, but somewhere else in the hook up.
 
turns out my info wasn't accurate. or its changed, or i mixed/messed stuff up.

its been a while..... but reading back through i dont know how i got stuff messed up so bad...

my older (4 years now) cells are still actually reading between 1-2mohm. before i wrongly said it was 21mohms :S

my newer cells are now reading between 3-4mohm. before i said it was 2mohms but im sure it remember it as about 8 mohms before. :S (maybe new cells do need to "break in")

this probably explains the difference in voltage sag between the new n old..... although im not sure why its changed/lessened over time. i havn't noticed less voltage sag really. either way 4mohms isnt the stated 2mohm in their data sheet :S

i did replace my battery loom from 14 to 10awg recently. but the batteries themselves still have 14awg and xt60s.
might be placebo but my butt dyno could feel a little more pep in the acceleration.

i will get round the replacing the wire in the batteries and put it on xt90s.
i still cant imagine im losing 2mohms a cell to about 20cm of 14awg but we shall see.
 
i finally replaced my batteries wires from 14awg to 10awg and connectors from xt60s to xt90s
removed/replaced every bus bar after polishing it up.....ANNND

its done sweet fack all to improve the battery resistance reading on either of my chargers.... :(

shrodingers box has finally been opened and my cat is dead :(

looks like my cells are potentially dud/old/used/defective/overated by manufacturer. not really sure

im yet to take it for a ride.....but i will report back if i see any difference what so ever in voltage sag.
 
jimmyhackers said:
looks like my cells are potentially dud/old/used/defective/overated by manufacturer. not really sure
Those cells were cutting edge in 2009, but could be a decade old now. Yours are assuredly old.
 
jimmyhackers said:
i finally replaced my batteries wires from 14awg to 10awg and connectors from xt60s to xt90s
removed/replaced every bus bar after polishing it up.....ANNND

its done sweet fack all to improve the battery resistance reading on either of my chargers.... :(

FWIW... i would not trust an IR reading from ANY charger. !!
.. measure it/calculate it using accurate MM and careful methodology. !
 
Hillhater said:
jimmyhackers said:
i finally replaced my batteries wires from 14awg to 10awg and connectors from xt60s to xt90s
removed/replaced every bus bar after polishing it up.....ANNND

its done sweet fack all to improve the battery resistance reading on either of my chargers.... :(

FWIW... i would not trust an IR reading from ANY charger. !!
.. measure it/calculate it using accurate MM and careful methodology. !
Virtual DC IR is not accurate representation of the batter.

DC voltage reading is ony as good as the meter spec, and if you look,, you will see they are not very accurate on DC V.. ( the one that counts)... Meters do not update fast enough either.

IR should only be taken ON charge, not discharge.... At room temp (59*, whtever), from 30% SOC

A 1000hz PID loop controlled charger is much more accurate than the virtal IR you might empirically measure .

On a multimeter.. meter is onl as good as its counts and resolutions..

I spent 1600$ +$ on (3) elevctrical meters this year alone ( datalogging Flukes, Flirs, and 25 Harmonic reading, power factor analysis, clamps...) .

The Powerlabs do a better job at determine IR. My powerlabs are 1000Hz AC IR and PID controlled 16 bit machines.

The wire does make a difference. I can see it.

THey knew AC IR four terminal sensing was MUCH better than the galvometer alone in 1860 or so. Been that way since. 1000( hertz AC). Its in the accuracy of the Wheatstone Kelvin bridge inherently.
 
after a few rides performance seems unchanged.
the cells arnt dated and were sold as new but they could be old stock. they are definately over a year old now.

my chargers are ok at resistance measuring. The accucel does to the nearest 1mohm and the sdt to the nearest 0.1 mohm. accurate enough for me to see im not getting what i should of, and if i can make any improvement to them.

either way larger wire and connecters didnt help. which is annoying :(

leaving the only other option for redemption of these cells potentially being the bus bar material.

its copper, nice n thick, but i remember that copper needs to be very pure to conduct electricity well.... these came made from the seller but is it possible a sub par peice of copper sheet cut into bus bars could be my culprit?
 
jimmyhackers said:
accucel does to the nearest 1mohm and the sdt to the nearest 0.1 mohm
Do they agree with each other when conditions are identical?

 
jimmyhackers said:
leaving the only other option for redemption of these cells potentially being the bus bar material.

its copper, nice n thick, but i remember that copper needs to be very pure to conduct electricity well.... these came made from the seller but is it possible a sub par peice of copper sheet cut into bus bars could be my culprit?
Copper bus bar is assuredly not the culprit. Definitely the batteries.
 
i think i have to come to that conclusion. batteries aint great.

on the bright side, they work, have the capacity they said they would and don't get hot/warm.
i can also in theory run something upto 7000w off of them. i might try 2 or 3kw in future

im maybe loosing a few mph at the top end from the sag but it's no biggy.

EDIT: check next page, i fixed them.
 
Back
Top