Acceptable Battery Sag

Batteries, Chargers, and Battery Management Systems.
Post Reply
newb123   10 W

10 W
Posts: 93
Joined: Jun 07 2019 4:12am

Acceptable Battery Sag

Post by newb123 » Nov 14 2019 2:25am

I have a 10S4P pack, and my sag is averaging 3 V (ex. 41 no throttle, 39 half throttle) on even terrain.

was wondering is this is average, or quite bad. The cells are a bit used

Also,

Ive read that most BMS start to balance at 4.2 V, but its best to charge to 4.0 for battery life extension

how do we get balanced cells and extend their life if we cant charge to 4.2V?

john61ct   1.21 GW

1.21 GW
Posts: 3639
Joined: Dec 18 2018 2:06pm

Re: Acceptable Battery Sag

Post by john61ct » Nov 14 2019 3:34am

Don't use a BMS that does not let you adjust the parameters to your liking.

Or use another method for balancing entirely.

Hillhater   100 GW

100 GW
Posts: 10671
Joined: Aug 03 2010 10:33pm
Location: Sydney ..(Hilly part !) .. Australia/ Down under !

Re: Acceptable Battery Sag

Post by Hillhater » Nov 14 2019 6:19am

% throttle tells us nothing about the load on the pack.
Need to know the actual current that gives 3 v sag.
Where /how are you measuring the voltage ?
Sag can have many causes..
Several bad cells
One dud cell
Bad connections
Poor wiring
Etc
This forum owes its existence to Justin of ebikes.ca

Jan-Erik-86   10 W

10 W
Posts: 91
Joined: Jan 23 2019 4:57am

Re: Acceptable Battery Sag

Post by Jan-Erik-86 » Nov 14 2019 6:31am

I'd say 3V is normal, but it all depends on what cells you're using, what wires, what connectors, and most of all how much power you're pulling from it.

As for balancing, there are basically 3 ways:
1, best) Use a programmable BMS and set it to balance at lower voltage. Also set overcharge voltage to 4.15-4.2V instead of the more common (on cheap none-programmable BMS) 4.3V +/- 0.05V.
2, good) Use a none-programmable BMS mainly for added protection, but use a balance charger instead of charging trough the BMS.
3, acceptable) Use a none-programmable BMS and charge trough it like usual, but stop at the desired voltage. Battery will slowly go out of balance, so once in a while charge it to 100% and let it balance.

Option 1 is what i recommend these days. Option 2 works as well, but is more work and requires a special charger.
Option 3 is cheap and (used to be) common, but this does require you to overcharge (at least some of) the cells once in a while to be able to balance the pack. This is simply because the balance does not kick in before the cells are already 100% charged, and if the pack is very out of balance when you do this it might harm some of the cells.

john61ct   1.21 GW

1.21 GW
Posts: 3639
Joined: Dec 18 2018 2:06pm

Re: Acceptable Battery Sag

Post by john61ct » Nov 14 2019 7:20am

There are plenty of other options too, both approaches and devices, but rarely used in this context.

User avatar
neptronix   100 GW

100 GW
Posts: 14168
Joined: Jun 15 2010 5:56pm
Location: California refugee living in Utah, USA
Contact:

Re: Acceptable Battery Sag

Post by neptronix » Nov 14 2019 4:59pm

That's a lot of sag for a 36v battery pack. About 8% of your energy is being lost to heat.
Colder temperatures make this worse.

Might be time to get a heartier battery. Or draw less current ( IE limit your amps and top speed down ) to stretch out the life of the current battery.
"Love and compassion are necessities, not luxuries. Without them, humanity cannot survive." - Dalai Lama

My first major build: 1.6kW 8T MAC motor on a Trek 4500.
The new all-arounder: Leafmotor 1500w @ 4kW on a Turner O2 full suspension.
The monster scooter: 20" eZee on a Cannondale Semi Recumbent.
Whipper-snapper: ? on a lightweight BikeE Semi Recumbent

999zip999   100 GW

100 GW
Posts: 9177
Joined: Jan 10 2010 1:40pm
Location: Dana Point So. Cal

Re: Acceptable Battery Sag

Post by 999zip999 » Nov 14 2019 5:11pm

What. ? What matters is how many amp draw is your controller and what motor ? How much you weigh ? Flat or hills ect.

newb123   10 W

10 W
Posts: 93
Joined: Jun 07 2019 4:12am

Re: Acceptable Battery Sag

Post by newb123 » Nov 16 2019 1:13am

ok will post more data as I ride like avg amps etc

on a throttle twist I noticed that I was using almost 7 amps sometimes, seems some power is getting lost somewhere

User avatar
dogman dan   100 GW

100 GW
Posts: 35406
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: Acceptable Battery Sag

Post by dogman dan » Nov 16 2019 7:16am

Power is always lost, always to heat. Heat in the battery cells, heat in the controller, and heat in the motor. The question is, how much. It would be great to ride around at 90% efficiency all the time, but its not going to get you much assist riding that way. Maybe not even enough to overcome the added weight. On the bright side, if you ride around at 75% efficiency, you just beat most gas engines by a ton.

I've certainly had batteries that sagged a lot more than 3v under full acceleration load.

As for balancing the pack, do you have any way to tell the voltage of each cell? if not, then you should just balance the pack often, meaning charge it to full. To minimize the damage of this, which is slight anyway, start riding soon after it gets full. Its time charged full, more than just that it ever got full.

If you can tell when the battery is balanced, then charge it full to balance only when its not balanced. Personally, I don't ever fret if a battery cell is .02v out of balance. More like .1v matters, or perhaps .05v. Tenth of a volt is close to balanced enough IMO.

newb123   10 W

10 W
Posts: 93
Joined: Jun 07 2019 4:12am

Re: Acceptable Battery Sag

Post by newb123 » Nov 17 2019 7:23am

at about 5 amps Im sagging at 3 volts, and loosing about 1V per KM

Im using a watt meter to monitor this

as for measuring each cell, I just use a DMM to do it after a ride. The BMS cuts out the pack at anything over 37 volts or so, so I am not using it except for charging. After a ride, the cells are not out of balance that much

docware   100 W

100 W
Posts: 293
Joined: Sep 16 2018 5:36am

Re: Acceptable Battery Sag

Post by docware » Nov 17 2019 7:41am

What cells are in the battery ? What is peak/average load (Amps or wats) ? How old (years/cycles) is the battery ?

Hillhater   100 GW

100 GW
Posts: 10671
Joined: Aug 03 2010 10:33pm
Location: Sydney ..(Hilly part !) .. Australia/ Down under !

Re: Acceptable Battery Sag

Post by Hillhater » Nov 17 2019 4:21pm

newb123 wrote:
Nov 17 2019 7:23am
at about 5 amps Im sagging at 3 volts, and loosing about 1V per KM
That is not a normal situation, !
You either have some bad cells, or a bad connection somewhere.
This forum owes its existence to Justin of ebikes.ca

newb123   10 W

10 W
Posts: 93
Joined: Jun 07 2019 4:12am

Re: Acceptable Battery Sag

Post by newb123 » Nov 18 2019 5:51am

The data I collected after a ride is as follows:

Time elaseped: 1 hr
Start voltage : 42V
End voltage: 36 V
Ah : 2.78
Ap : 20.47
Wp : 630
Vm : 30.57

Hillhater   100 GW

100 GW
Posts: 10671
Joined: Aug 03 2010 10:33pm
Location: Sydney ..(Hilly part !) .. Australia/ Down under !

Re: Acceptable Battery Sag

Post by Hillhater » Nov 18 2019 8:34am

So, basicly that means you had approx 10 volts of “sag” at 20amps.
As i said,.. that is not normal !
And since the pack dropped 6volts with only 2.78ah used, i would suspect you have a few dud cells or some cell groups way out of balance.
This forum owes its existence to Justin of ebikes.ca

newb123   10 W

10 W
Posts: 93
Joined: Jun 07 2019 4:12am

Re: Acceptable Battery Sag

Post by newb123 » Nov 18 2019 9:14pm

ok after another ride I will check the cells balance voltage again, for any duds or out of balance. I checked my phase wires to ground, all at infinity so nothing leaking there.

is 1 volt an acceptable number for out of balance tolerance? for example most cells at 4.0 and one at 3.2?

User avatar
st35326   100 W

100 W
Posts: 249
Joined: Jan 26 2016 9:46am
Location: Modesto, CA

Re: Acceptable Battery Sag

Post by st35326 » Nov 18 2019 9:33pm

Hillhater wrote:
Nov 18 2019 8:34am
So, basicly that means you had approx 10 volts of “sag” at 20amps.
As i said,.. that is not normal !
And since the pack dropped 6volts with only 2.78ah used, i would suspect you have a few dud cells or some cell groups way out of balance.
I've got a 72 volt 20s 6p pack made with Samsung 40T cell that Sags exactly 10 volts when FLux weakening kicks in (at about 6kw or anything above 45mph). Its been that way for some time now, it likely could use one more parallel groups but the pack was bought and paid for before the more powerful controller was installed. Seems to handling just fine though.

john61ct   1.21 GW

1.21 GW
Posts: 3639
Joined: Dec 18 2018 2:06pm

Re: Acceptable Battery Sag

Post by john61ct » Nov 18 2019 10:51pm


newb123 wrote:is 1 volt an acceptable number for out of balance tolerance? for example most cells at 4.0 and one at 3.2?
No. Are these scrapped secondhand cells?

Hillhater   100 GW

100 GW
Posts: 10671
Joined: Aug 03 2010 10:33pm
Location: Sydney ..(Hilly part !) .. Australia/ Down under !

Re: Acceptable Battery Sag

Post by Hillhater » Nov 19 2019 2:52am

st35326 wrote:
Nov 18 2019 9:33pm

I've got a 72 volt 20s 6p pack made with Samsung 40T cell that Sags exactly 10 volts when FLux weakening kicks in (at about 6kw or anything above 45mph). Its been that way for some time now, it likely could use one more parallel groups but the pack was bought and paid for before the more powerful controller was installed. Seems to handling just fine though.
Yours is a very different situation
Most of your 10v sag, is simply due to yhe very high current load each cell sees (approx 14A per cell, = 0,3v sag per cell). With the high currents , you will likely “lose” voltage through connections and cables also.
This forum owes its existence to Justin of ebikes.ca

User avatar
st35326   100 W

100 W
Posts: 249
Joined: Jan 26 2016 9:46am
Location: Modesto, CA

Re: Acceptable Battery Sag

Post by st35326 » Nov 19 2019 5:36am

Hillhater wrote:
Nov 19 2019 2:52am
st35326 wrote:
Nov 18 2019 9:33pm

I've got a 72 volt 20s 6p pack made with Samsung 40T cell that Sags exactly 10 volts when FLux weakening kicks in (at about 6kw or anything above 45mph). Its been that way for some time now, it likely could use one more parallel groups but the pack was bought and paid for before the more powerful controller was installed. Seems to handling just fine though.
Yours is a very different situation
Most of your 10v sag, is simply due to yhe very high current load each cell sees (approx 14A per cell, = 0,3v sag per cell). With the high currents , you will likely “lose” voltage through connections and cables also.
Right on.

User avatar
dogman dan   100 GW

100 GW
Posts: 35406
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: Acceptable Battery Sag

Post by dogman dan » Nov 19 2019 6:50am

1v is way out of balance, as measured on one 4.2v cell group. .1v repeat, .1v out of balance is still out of balance, but tolerable if your are not discharging 100%. .05v should be tolerable, provided it does not continue to drift farther than that. So if your cell groups when fully charged are half of them at 4.15v, and the others at 4.18v, this is typical for a "balanced" bike battery when fully charged. All of them at 4.2v exactly would be great, but in the real world, cells don't act like that for all that long. After use, sometimes even when new, a few cells will charge faster, causing the bms to stop the charge. So those cells at 4.15v never quite got 100% full, or they did, and then when stopped at 4.2v, they discharged just a bit because even new, they simply only hold 4.15v. This isn't the test lab, your pack has some great cells in it, and some that are only average. But if all your cells are at 4.15 or more when charged, its all good. Good enough to ride, and use the battery for years.

But,,, I'm not sure what you are seeing. Are you able to check the voltage of each group of cells? Some batteries make the bms and its plug inaccessible.

3v of sag under a 5 amps load is a lot, but alas, typical of cheap ass cell packs. Sounds like you may have one. Its just a lot of load for cheap cells.

You say your pack is 4p. 4p of what? Typical laptop cells found in really cheap batteries can handle about .5c of discharge rate. Lets say your cells hold 2 ah each. that would mean 8 ah of size. a .5c discharge rate of that pack would be 4 amps. So if your cells are that type, you would be seeing a lot of sag at 5 amps, and be killing them at 10 amps.

We may also be talking about two different things. Voltage sag is the voltage drop when load is applied.

But as you ride, the voltage will drop as you ride. This is not sag, its just discharging it. The more you ride, the lower the pack voltage goes,, and,, it still sags under load. At some point, sag alone can shut off the pack before its absolutely empty.. Sag tends to get worse as you get into the last 10% of the charge. So the lower your voltage when stopped gets, the less amps you can give it, especially at the very end.

newb123   10 W

10 W
Posts: 93
Joined: Jun 07 2019 4:12am

Re: Acceptable Battery Sag

Post by newb123 » Nov 20 2019 2:34am

No. Are these scrapped secondhand cells?

yes

They are not Chinese though. I have the data sheet

it seems the discharge rate is 2.15 A, so 4 of these 18650 Sanyo in parallel, would handle 5 amps,

I measure them off the charger with BMS, all balanced at 4.2

I will do another ride, and check the balance again
SANYO 18650.jpg
SANYO 18650.jpg (132.58 KiB) Viewed 1155 times

Jan-Erik-86   10 W

10 W
Posts: 91
Joined: Jan 23 2019 4:57am

Re: Acceptable Battery Sag

Post by Jan-Erik-86 » Nov 20 2019 3:17am

You don't tell exactly which cell it is, but by the specs it looks like it's Sanyo UR18650A?
Since these are used cells, did you test each cell for capacity, DCIR, and self discharge before putting them together?

I'd charge full and check that all are at 4.2V, then discharge and check again if all cells are still nearly the same voltage.

newb123   10 W

10 W
Posts: 93
Joined: Jun 07 2019 4:12am

Re: Acceptable Battery Sag

Post by newb123 » Nov 20 2019 5:39am

I got R1112 Sanyo off the cell wrapping

I ordered the li ion cell discharger from ebay, awaiting its arrival

best I can do now is charge ride check

Jan-Erik-86   10 W

10 W
Posts: 91
Joined: Jan 23 2019 4:57am

Re: Acceptable Battery Sag

Post by Jan-Erik-86 » Nov 20 2019 6:21am

R1112 is just the date code, meaning march 2013. The actuall modell number should also be written on it, but its not easy to read the text on Sanyo cells...

docware   100 W

100 W
Posts: 293
Joined: Sep 16 2018 5:36am

Re: Acceptable Battery Sag

Post by docware » Nov 20 2019 7:47am

newb123 wrote:
Nov 20 2019 5:39am
I ordered the li ion cell discharger from ebay, awaiting its arrival

best I can do now is charge ride check
Second hand cells, production date March 2013, high voltage sag, …..
Looking at the facts, I would say best you can do is to start selection of new cells suitable to your requirements.

Post Reply