Battery life

[rm2001]

Hi.
New to site and trying to convert bike to ebike.
But..

Cycle Satiator Lithium Battery Charge Simulator

The Satiator Charge Simulator is a web application for creating custom charge profiles for your lithium battery pack. It allows you to view a simulated charging graph to see the charging time in difference scenarios, produce partial charge profiles to maximize battery cycle life, and download...

ebikes.ca

Above site link (usa etc) has simulators models on expected speed / range (which seemed buggy) ... but also on extending your battery life. As in number of battery charge cycles. So most warranties are for 500 or 1000 charge cycles or x months.

From above site :-

"It is now well known that most lithium chemistries (with the exception of LiFePO4) can see drastic improvements in calendar and cycle life when they are not held at the nominal full charge voltage of 4.2 V/cell but are charged to a lower voltage instead. That’s how electric car manufacturers are able to 5-8 year battery warranties on cells that usually only test to ~500 cycles.

With most ebike chargers, you have no ability to set the full charge voltage and have to accept topping it up to 4.2 V/cell. This gives the most range on a charge, but if you don’t require the full capacity of your battery on most of your trips then you are unnecessarily reducing the battery life every time you charge it. In many cases that means replacing your ~$1000 lithium battery pack every 1-2 years, when with proper management it could be lasting more like 4-5 years. In fact the further from full charge you go, the more pronounced the life cycle improvements".

So is there truth in above ?

A new ebike battery is say £200 so extending pack life double seems worth doing.

 

[matthewslack]

You can achieve that result without the need for a fancy charger simply by understanding how long it takes to charge your battery, and only putting it on for long enough to get to about 80%, and never running it below 20%.

That's half the battle. The other half is down to size and quality of battery, type of motor and riding style. A small, generic chinese battery with a big direct drive hub motor is doomed whatever you do during charging.

A bigger battery of higher quality cells and a gentler motor will give the battery an easier life.

Most batteries don't get used for the quoted number of cycles within the length of the warranty, so don't read too much into that number.

Your best learning comes from getting a bike and starting to ride it. Your second bike will be the one you really want!

 

[soundwave]

my 2014 400w bosch batt made it to 112 charge cycles b4 the capacity was so low it was no good for me anymore as in turbo with a dongle could empty it in about 5 miles.

bosch warranty is 200 charge cycles or 2 years shitmano is 2 years and 1000 cycles neither would make that warranty pulling full amps every day for 2 years as if you want double the speed it will half the life of the batt no matter how you charge it.

they are renewable tho just throw it in the bin and buy a new one

 

[Nealh]

I wouldn't worry about cycle life with most generic stuff as the cell used are medocre to good but not the very best top notch cells, so just use it and enjoy the E feel.
Not many will get 500+ cycles from a small ebike battery.
One would have to ride 25,000 miles minimum to get 500 full cycles from a 50 mile range 15ah battery.

One can mess with 70/80/90% charging but really not worth the faff, it only needs one slightly weak cell group and then the battery will get out of balance unless one has a way on monitoring cell balance. Every 4 -5 years there are cell improvements and something better will be the next best thing, like 21700 cells and there improved mah and current rating.

EV & ebike batteries are chalk and cheese , they are monitored so differently to try and compare cycle life.

 

[WheezyRider]

Charging to 4.2V per cell is right on the limit of what Li-ion cells can handle. It is bad news for the protective interfacial layers at the electrodes, all it takes is a small change in temperature from standard conditions and things can start to break down. That is why they say not to leave Li-ion batteries in a fully charged state for a long time. Manufacturers go for 4.2V/cell as a compromise between reasonable cycle life for maximum cell capacity - under standard conditions.

It is sad to see some of the comments reflecting a use it and throw it away attitude. Lithium batteries are still not recycled in significant numbers, most will end up in landfill and, they are often the most expensive part of a bike conversion in most cases. Not everyone has the time, energy and skill to really care for battery packs, I understand that, but we should be pushing manufacturers to make better battery management systems (BMS), get politicians to set up decent recycling facilities and trying to do what we can to keep our packs going for as long as possible. The lithium for our batteries comes out of holes in the ground and extracting it does a lot of environmental damage and in some cases health & social difficulties for the miners and people living around mines.

If looked after properly, there is no reason for a battery pack not to last 5 to 10 years. If you want the latest improvement in battery tech and want to upgrade every 2 years, fine - but pass on your well looked after pack to someone else or sell it on.

Too many manufacturers make things with obsolescence built in, so in a couple of years you will have to buy another. A lot of packs will end up dying not because of how it is charged, but because the BMS is crap and it kills the pack. The phrase "battery murdering system" is often used for BMS I've even seen some BMS with a built in, non repairable fuse, to effectively kill the pack. I'm also sure there is built in firmware on some BMS, programmed to kill the pack after a certain time.

I charge all my packs to 41V (4.1V per cell bank), it is not too difficult to modify a PSU to do this. There is a concern that most battery management systems do not start to balance cells until approx >4.15V, so if only charging to 4.1V per cell there is the potential for cells to go out of balance. However, in my experience, even after 100 cycles or more this has not happened to any significance. I think the ideal would be to do a 4.18V/cell balance charge (still avoiding 4.2V!) every 100 to 200 cycles, or when you are going on a long trip and need maximum range. Having a BMS that will give you individual cell bank voltages I think is a good idea, then you will spot potential issues quickly.

In my experience, charging to 4.1V/cell does help prolong battery life and you don't need to worry so much about a lot of things, like leaving the battery connected to the charger overnight, or using the battery straight after a charge etc. I also avoid deep discharge whenever possible, this is also quite damaging. I would not buy anything smaller than 15Ah for a "250"W motor, so there is a good balance between current requirement per cell and power requirement, while keeping a decent reserve capacity. Also, in winter, assume capacity will fall by a significant amount, and hence the ability to supply maximum current. Furthermore, cycle life is normally given on the basis of having 80% capacity after X number of cycles. If you start with a decent sized pack, even 80% is still useful, whereas 80% of a small pack may be marginal at best, especially in cold weather.

Something that does annoy me is that battery chargers are often set at too high a voltage. I have had a couple of cheap ones at >43V that should have been max 42V. Recently, I bought a new Yose Power Battery with charger and was disappointed to see that it was also reading over 42V. In the past they used to set their chargers to 41.7V. 4.2V/cell is not great, going over that is risking seriously reducing battery life.

One thing to bear in mind when measuring battery voltages you need a decent quality multimeter. Many of the cheaper ones will not have a good enough calibration to measure things accurately.

 

[rm2001]

You can achieve that result without the need for a fancy charger simply by understanding how long it takes to charge your battery, and only putting it on for long enough to get to about 80%, and never running it below 20%.

That's half the battle. The other half is down to size and quality of battery, type of motor and riding style. A small, generic chinese battery with a big direct drive hub motor is doomed whatever you do during charging.

A bigger battery of higher quality cells and a gentler motor will give the battery an easier life.

Most batteries don't get used for the quoted number of cycles within the length of the warranty, so don't read too much into that number.

Your best learning comes from getting a bike and starting to ride it. Your second bike will be the one you really want!

Ok thanks .. so effectively that means buyers should look to getting double Ah battery. Since 80% to 20% is 60% battery capacity.
Which is effectively what lve done ordered 15.5 Ah instead of 12 Ah or less.
Longest of my trips will be 35 miles. And l intend pedal assist. Apparently controller can set maximum current draw .. so l can force controller to be lighter on battery and motor by setting lower amps.
Not got conversion kit yet .. so just going on specs and manuals.
Thanks.

Charging to 4.2V per cell is right on the limit of what Li-ion cells can handle. It is bad news for the protective interfacial layers at the electrodes, all it takes is a small change in temperature from standard conditions and things can start to break down. That is why they say not to leave Li-ion batteries in a fully charged state for a long time. Manufacturers go for 4.2V/cell as a compromise between reasonable cycle life for maximum cell capacity - under standard conditions.

It is sad to see some of the comments reflecting a use it and throw it away attitude. Lithium batteries are still not recycled in significant numbers, most will end up in landfill and, they are often the most expensive part of a bike conversion in most cases. Not everyone has the time, energy and skill to really care for battery packs, I understand that, but we should be pushing manufacturers to make better battery management systems (BMS), get politicians to set up decent recycling facilities and trying to do what we can to keep our packs going for as long as possible. The lithium for our batteries comes out of holes in the ground and extracting it does a lot of environmental damage and in some cases health & social difficulties for the miners and people living around mines.

If looked after properly, there is no reason for a battery pack not to last 5 to 10 years. If you want the latest improvement in battery tech and want to upgrade every 2 years, fine - but pass on your well looked after pack to someone else or sell it on.

Too many manufacturers make things with obsolescence built in, so in a couple of years you will have to buy another. A lot of packs will end up dying not because of how it is charged, but because the BMS is crap and it kills the pack. The phrase "battery murdering system" is often used for BMS I've even seen some BMS with a built in, non repairable fuse, to effectively kill the pack. I'm also sure there is built in firmware on some BMS, programmed to kill the pack after a certain time.

I charge all my packs to 41V (4.1V per cell bank), it is not too difficult to modify a PSU to do this. There is a concern that most battery management systems do not start to balance cells until approx >4.15V, so if only charging to 4.1V per cell there is the potential for cells to go out of balance. However, in my experience, even after 100 cycles or more this has not happened to any significance. I think the ideal would be to do a 4.18V/cell balance charge (still avoiding 4.2V!) every 100 to 200 cycles, or when you are going on a long trip and need maximum range. Having a BMS that will give you individual cell bank voltages I think is a good idea, then you will spot potential issues quickly.

In my experience, charging to 4.1V/cell does help prolong battery life and you don't need to worry so much about a lot of things, like leaving the battery connected to the charger overnight, or using the battery straight after a charge etc. I also avoid deep discharge whenever possible, this is also quite damaging. I would not buy anything smaller than 15Ah for a "250"W motor, so there is a good balance between current requirement per cell and power requirement, while keeping a decent reserve capacity. Also, in winter, assume capacity will fall by a significant amount, and hence the ability to supply maximum current. Furthermore, cycle life is normally given on the basis of having 80% capacity after X number of cycles. If you start with a decent sized pack, even 80% is still useful, whereas 80% of a small pack may be marginal at best, especially in cold weather.

Something that does annoy me is that battery chargers are often set at too high a voltage. I have had a couple of cheap ones at >43V that should have been max 42V. Recently, I bought a new Yose Power Battery with charger and was disappointed to see that it was also reading over 42V. In the past they used to set their chargers to 41.7V. 4.2V/cell is not great, going over that is risking seriously reducing battery life.

One thing to bear in mind when measuring battery voltages you need a decent quality multimeter. Many of the cheaper ones will not have a good enough calibration to measure things accurately.

Ok great.
Are there any high quality chargers out there to buy or modification tips ?
Im not electronics engineer but from past a simple (high current) diode requires 0.6 v to forward bias it. So a simple (switchable) way to force output voltage down 0.6v. from nominated 42v ?

 

[rm2001]

Ok thanks .. so effectively that means buyers should look to getting double Ah battery. Since 80% to 20% is 60% battery capacity.
Which is effectively what lve done ordered 15.5 Ah instead of 12 Ah or less.
Longest of my trips will be 35 miles. And l intend pedal assist. Apparently controller can set maximum current draw .. so l can force controller to be lighter on battery and motor by setting lower amps.
Not got conversion kit yet .. so just going on specs and manuals.
Thanks.

Ok great.
Are there any high quality chargers out there to buy or modification tips ?
Im not electronics engineer but from past a simple (high current) diode requires 0.6 v to forward bias it. So a simple (switchable) way to force output voltage down 0.6v. from nominated 42v ?

As well .. were reports of fires caused by eScooter (lower quality ?) Batteries.
Is this forcing of a lower charging voltage .. accepting the range reduction .. both a battery life cycles and safety benefit ?

 

[WheezyRider]

It's hard to recommend anything these days. SANS chargers used to be good as they had a pot that allowed you to tweak the output voltage. Vruzend is a higher end brand, but it looks like they just rebadge generic stuff. In both SANS and Vruzend chargers I've had to touch up dodgy soldering due to dry/poor joints. These days, any charger I buy I put it on a decent multimeter I trust to check the output and if possible open it up to check the soldering.

If you are lucky, as I said, it can be just a case of tweaking an output pot to get the voltage you want, but often these days, it's not so easy. The manufacturer may just not care about the exact output value and slap a generic resistor in instead, or sometimes someone will go to the trouble of fitting a charger specific resistor to get the right voltage.

If there is no pot, you have to track down the voltage regulating circuit and adjust the resistors yourself:

Tuning cheap 42V charger

A while back I bought a couple of cheap 42V chargers as spares, so I could have one at work, at home etc. I got two for just under £10 each. They are quite a lot smaller than the SANS charger and looked good: However, when I checked the output, I was disappointed to see that it was nearly...

www.pedelecs.co.uk

Mikes eBikes 3A charger

I picked up one of these recently and thought I'd give it a once over. It looked quite compact, came with a fan, already had an XT60 connector and I could do with something that charges a bit faster than my SANS chargers. It uses a PSU PCB company that also makes the PSU boards for the Vruzend...

www.pedelecs.co.uk

Vruzend 58.8V Charger - a look inside the box

I'm in the process of building a new 14s battery pack and I've bought a Vruzend 3A 58.8V charger for this project. I was interested to see what was inside, in particular to see if it would be possible to tweak the output voltage to charge to 4.1V per cell. It cost £36. First a look at the...

www.pedelecs.co.uk

 

[WheezyRider]

Sometimes you may find a charger running at >42V for a 36V nominal battery. This may be because there is some additional electronics in the battery controller that drops it to 42V (eg a silicon p-n junction will typically cause 0.6V voltage drop):

Dangerous e-bike chargers found on Amazon and eBay

https://www.bbc.co.uk/news/technology-62751274

www.pedelecs.co.uk

However, it is not ideal to assume that a p-n junction will always give exactly 0.6V drop. The drop depends on temperature and also the amount of current flow through the junction. Hence, if you want to accurately and reliably change the output voltage, it's best to mod the voltage divider on the TL431 temperature compensated precision regulator on the PCB.

 

[Bikes4two]

----------------------
However, it is not ideal to assume that a p-n junction will always give exactly 0.6V drop. The drop depends on temperature and also the amount of current flow through the junction.
------------------------

• Do you know by how much the junction voltage varies by temperature please?
• My 4 amp charger outputs 41.8v off load and 41.0v when charging and I have at various times considered 'splicing in' a suitably rate diode in series with the charger output but a bitlike @Nealh I not sure I can go with the faffing around.

 

[StuartsProjects]

Do you know by how much the junction voltage varies by temperature please?

Not a lot I guess, but it would likely vary depending on the particular diode, and the amount of current, which in turn would affect the temperature. Its data that just might be in the particular diodes data sheet.

It does vary quite a bit with the amount of current, higher voltage drop for more current.

If you want an accurate answer, then you would have to measure it.

 

[WheezyRider]

• Do you know by how much the junction voltage varies by temperature please?
• My 4 amp charger outputs 41.8v off load and 41.0v when charging and I have at various times considered 'splicing in' a suitably rate diode in series with the charger output but a bitlike @Nealh I not sure I can go with the faffing around.

4.18V output OCV sounds good for a charger, although it would be good to know that the calibration of the multimeter you are using is correct. It is 0.02V away from 42V and 0.02V in the 200V scale is far less than the error you expect from low to mid quality multimeters.

As has been said, you need to check what the drop would be with temp and current flow for a given diode from the data sheet. If you do use a diode to drop the voltage, make sure it is rated for more than the current you expect to pass through it. A standard 1A diode on it's own would not be good. So get a 5A type for a 2 A charger, or put several 1A diodes in parallel. Also, it will probably get warm, so make sure the heat can escape.

 

[saneagle]

So is there truth in above ?

No, its complete bunk. If you want to charge an ebike battery to anything other than 100% (4.2v per cell), you must first dismantle it and replace the management PCB with a different and more expensive one, as the one in it is designed to manage at 4.2v per cell and will not work properly at less. A car battery management system is completely different, as it's designed to manage the cells at a lower voltage. You cannot compare the two.

 

[WheezyRider]

No, its complete bunk. If you want to charge an ebike battery to anything other than 100% (4.2v per cell), you must first dismantle it and replace the management PCB with a different and more expensive one, as the one in it is designed to manage at 4.2v per cell and will not work properly at less.

As long as you are aware that a BMS designed for 4.2V per cell charging may not balance cells below approx 4.15V per cell and hence do a balancing charge every now and then, there is no need to replace the BMS for the battery to function perfectly satisfactorily.

 

[saneagle]

there is no need to replace the BMS for the battery to function perfectly satisfactorily.

There is.

 

[WheezyRider]

There is.

Because.....?

 

[WheezyRider]

Here is a dismantled Yose power pack, generic Chinese cells, crappy 20A BMS. Bought in 2019, charged only to 41V since June 2020, so 2.5 years of 41V charging, hundreds of cycles and thrashed quite hard, up to 28A cont for many of those cycles. In that time it has never had a charge cycle above 41V.



The BMS has failed after the abuse it has had and I have been planning to replace it for a long time with a better one. So what do the cell bank voltages look like?

Cell Bank / Voltage (V)
1 ............ 3.677
2 ............ 3.678
3 ............ 3.678
4 ............ 3.678
5 ............ 3.677
6 ............ 3.661
7 ............ 3.676
8 ............ 3.678
9 ............ 3.679
10 .......... 3.678

Overall pack voltage 36.762V

So after all that abuse and hundreds of cycles, the maximum variation is 17 mV, mostly because just one bank is slightly lower, the rest are within a couple of mV. I expect the variation to increase slightly once the pack is charged - this is what I find with my homebuilt pack I added monitoring ports to:

Charging pack to 41V experiment

Some people think it's a good idea for cell longevity to only charge to 4.1V instead of the normal 4.2V. However, when you have a pack of cells, only charging to 41V may cause balance issues as the BMS in many cases does not start balancing until about 41.5V. Anecdotal evidence from some on this...

www.pedelecs.co.uk

I will charge it up with a new BMS and see. However, this variation is unlikely to exceed 20 to 30mV. With this pack I never knew what the bank voltages were from new, so it is quite possible bank 6 has always been like this. I will charge to 41V initially and then to 42V with the new BMS and see what happens to bank 6, will it balance? We'll see.

This quite generic pack was abused and not balanced for 2.5 years and hundreds of cycles and yet it hasn't gone seriously out of balance. This shows that for normal use, charging to 41V for a significant number of cycles, and then balancing at 42V every now and then does not mean a pack's BMS has to be replaced with one designed for 41V charging.

 

[saneagle]

Here is a dismantled Yose power pack, generic Chinese cells, crappy 20A BMS. Bought in 2019, charged only to 41V since June 2020, so 2.5 years of 41V charging, hundreds of cycles and thrashed quite hard, up to 28A cont for many of those cycles. In that time it has never had a charge cycle above 41V.

View attachment 49657

The BMS has failed after the abuse it has had and I have been planning to replace it for a long time with a better one. So what do the cell bank voltages look like?

Cell Bank / Voltage (V)
1 ............ 3.677
2 ............ 3.678
3 ............ 3.678
4 ............ 3.678
5 ............ 3.677
6 ............ 3.661
7 ............ 3.676
8 ............ 3.678
9 ............ 3.679
10 .......... 3.678

Overall pack voltage 36.762V

So after all that abuse and hundreds of cycles, the maximum variation is 17 mV, mostly because just one bank is slightly lower, the rest are within a couple of mV. I expect the variation to increase slightly once the pack is charged - this is what I find with my homebuilt pack I added monitoring ports to:

Charging pack to 41V experiment

Some people think it's a good idea for cell longevity to only charge to 4.1V instead of the normal 4.2V. However, when you have a pack of cells, only charging to 41V may cause balance issues as the BMS in many cases does not start balancing until about 41.5V. Anecdotal evidence from some on this...

www.pedelecs.co.uk

I will charge it up with a new BMS and see. However, this variation is unlikely to exceed 20 to 30mV. With this pack I never knew what the bank voltages were from new, so it is quite possible bank 6 has always been like this. I will charge to 41V initially and then to 42V with the new BMS and see what happens to bank 6, will it balance? We'll see.

This quite generic pack was abused and not balanced for 2.5 years and hundreds of cycles and yet it hasn't gone seriously out of balance. This shows that for normal use, charging to 41V for a significant number of cycles, and then balancing at 42V every now and then does not mean a pack's BMS has to be replaced with one designed for 41V charging.

I have batteries like that, which are ten years old and have always been charged to 42v, so what does that prove about all batteries? I also have a garage full of batteries that became unserviceable because of all sorts of reasons, including their previous owners using charging policies recommended by some guy on the internet.

It's possible to completely remove the BMS from a battery and wire the pack power wires directly to the output, then use it like any other battery and get years of use out of it with no problems. On the other hand, it might go out of balance and become useless or catch fire. There's no way to determine which of the two you'd get other than by doing it. There isn't even any data on probability of which you'd get. If someone was successful with a BMSless battery, would that be justification to suggest removing the BMS as acceptable practice?

 

[guerney]

This chap made his 11kg 48V ebike battery using large prismatic cells - might battery packs made with fewer cells, lower the odds of pack failure?

Another prismatic cell build (for electric motorbike)

 

[WheezyRider]

I have batteries like that, which are ten years old and have always been charged to 42v, so what does that prove about all batteries? I also have a garage full of batteries that became unserviceable because of all sorts of reasons, including their previous owners using charging policies recommended by some guy on the internet.

It's possible to completely remove the BMS from a battery and wire the pack power wires directly to the output, then use it like any other battery and get years of use out of it with no problems. On the other hand, it might go out of balance and become useless or catch fire. There's no way to determine which of the two you'd get other than by doing it. There isn't even any data on probability of which you'd get. If someone was successful with a BMSless battery, would that be justification to suggest removing the BMS as acceptable practice?

If you have some batteries 10 years old that were always charged to 42V, great...but how many cycles have they had? Have they really been charged to 42V? Did you check the chargers with properly calibrated DVMs? Could it be that they have actually been charged to lower than 42V in that time? Until recently, Yose power used to supply chargers set to 41.7V. I m trying to offer some data from personal experience. You offer anecdotes, but no data and then raise the option of running a battery without a BMS. Although many BMS supplied with packs are not great and can sometimes contribute to battery failure, I would never recommend running a pack without one. I know some people do it, but it is far too dangerous. The BMS has built in short circuit protection, limiting the current if there is a short. Without it, you could end up with hundreds of amps going through a short, batteries bursting and serious fires. Ok, this could be mitigated by fitting a fuse, but the other thing the BMS does is to cut power if one cell bank goes below a critical voltage. Without it, the other cells end up dumping power into the weak bank, it gets very hot and you soon have a fire on your hands. So no, do not operate a battery without a BMS.

I am not recommending everyone to charge their batteries to 4.1V/cell, this is only going to appeal to a certain niche of people who have the technical understanding of what they are doing and in the first place, have the ability to modify their charger to the desired voltage - using a decent multimeter to make sure they have things set up properly. Then there is the need to remember to perform balance charges now and then. The vast majority would not even be capable of doing this, or be bothered with the extra time overhead of monitoring things. These people should carry on charging to 4.2V/cell, BUT, I think where possible it is a good idea for them to check that their charger is not actually giving more than 4.2V/cell. Also, they should do what they can lobby for better recycling for their packs and better chargers/BMS to prolong battery life, over maximum capacity for a relatively short number of cycles.

What I post also only relates to 18650 based packs. It is a relatively mature technology now and cell to cell variation from quality brands is very small to negligible and even unbranded Chinese cells are getting quite good. I am not convinced things are as mature with other cell sizes, so the risk of things going out of balance is greater. That is something I don't have data for, but I have a hunch about, from what I've read.

Charging to 4.1V/cell to prolong battery life is well known, understood and documented:

BU-409: Charging Lithium-ion

BU meta description needed...

batteryuniversity.com

Li-ion does not need to be fully charged as is the case with lead acid, nor is it desirable to do so. In fact, it is better not to fully charge because a high voltage stresses the battery. Choosing a lower voltage threshold or eliminating the saturation charge altogether, prolongs battery life but this reduces the runtime. Chargers for consumer products go for maximum capacity and cannot be adjusted; extended service life is perceived less important.

Ultimately, it's up to people to decide what they do with their packs. They have to consider their technical ability and their time commitments, but everyone should know the facts and be able to make up their own mind. Also, when armed with the facts, everyone has the power do something to reduce what goes into landfill.