Do headway cells really need a BMS?

[jondoh]

Do headway cells really need a BMS? I'm thinking no.

Here's why:

1. EXPERIENCE: I've used thundersky 100 ah batteries (which is LiFePo like the Headway) in my car for the past year without a bms and I haven't had a problem yet. Actually, i did replace one cell that was killed by a shunt resistor type bms and that's when i took out the bms.

2. LiFePo CHEMISTRY: the Lifepo chemistry is good at masking their true condition. When at rest, they're all just about the same voltage +/- 0.02 volts. The only way to tell if a cell has a problem is install a cell log or similar data collection device and see how it performs under load.

3. WEAK CELLS STILL WORK: even if you do have a weak cell, it will still operate-- you just don't have full pack voltage under load since weak cells sag. Check out this battery test here http://s2kev.blogspot.com/

4. WHEN CELLS DIE, THEY BECOME A SHORT: The one cell that died on my car became a short circuit. I was never stuck-- though i was freaking out over the lower pack voltage. I replaced the dead cell with a brand new one and it integrated itself nicely into my pack without any fuss.

So, I'm asking people out there that have had experience with LiFePo4, if you are using a BMS, is it a worth while thing to have? WHY? I understand that some cells like early ping batteries and other 2C rated Lifepo batteries might have issues that a BMS may be useful for but for a 5C continuous (15C burst) cell like the Headway, how does having a BMS help?

- jon

 

[Ypedal]

Depends on a few factors...

- If you get good, well matched, fresh cells, take the time to cycle them on the bench before assembly to check end of charge voltage and capacity, run a capacity meter on your ride, monitor voltage actively while riding.. and never.. NEVER.. use 100% of the rated capacity..... you might ok.

- You need a proper charger, one that is set for proper end of charge voltage with a bit of breathing room ( at the cost of less than an absolute full charge ) ..

- Don't mind repairing a pack once in a while, buy a few extra cells !.

I have run almost all my packs with no BMS.. lipo, limn, lifepo4( PSI and Headway and Vpower ) .. never drain more than 75% DOD, use single cell chargers once in a while to top off all the cells.. fine for me.

 

[Russell]

My 10Ah pack is made from 5Ah pouch type cells like Ping uses. After 38 cycles the last cell group died, I believe from mechanical stress on the fragile tab, so I removed them along with the BMS. I then had a 15 cell group pack which I have run sans a permanent BMS for more than 100 cycles. I charge the pack with a 54.6V/2A charger and once in a while reconnect the BMS to balance the pack (I use a different 55.5V/4A charger). I did check the individual cell voltages for about 10 cycles after removing the BMS but haven't bothered for the past 90+ cycles. I do connect a wattmeter and keep a record of the data on a log sheet. I try to keep the DOD to less than 80% and have only exceeded that a few times. I would still be using the BMS if the failure had not occurred however the pack is performing quite well without it.


-R

 

[jondoh]

Interesting. Two people operating without a BMS (three if you include me).

Sounds like the occasional monitoring with a volt meter and a LVC are what's most important.

Still, I would like to hear someone say that their BMS saved them. I'd like to know how.

 

[Battboy]

The BMS won't save you. What saves you an an LVC circuit board on the battery pack. Unless you control it by another device?

 

[999zip999]

So my bms duty is to balance the cells after the chargher light and 16led lights turn on. And limit the top end cutoff at 20-30-40-60 amp. bms ect. through the shunts, that can be soldier to raise the output amps. That sounds trickie. Plus cutoff power when one of my 16 cells gets lower than ? 3.28v on a 48v20ah ping ? Is this true ? what else can it do ? My 60 amp. Ping bms has 10 fets as oppose to a 40amp. ping that has 5 fets (the black square things with two pins and the ass end soldier on the board). What is the advantice of extra fets ? Remeber this is the purposes not just to ask dumb question, But to get smart answers. Thanks My brain must know......

 

[Russell]

Plus cutoff power when one of my 16 cells gets lower than ? 3.28v on a 48v20ah ping ? Is this true ? what else can it do ? My 60 amp. Ping bms has 10 fets as oppose to a 40amp. ping that has 5 fets (the black square things with two pins and the ass end soldier on the board)

My LiFePO4 BMS cut output power if any cell reached 2.0V or opened the charging circuit if any cell went over 3.90V (these limits can vary a bit).

More FETs means higher power handling capability.

-R

 

[jondoh]

So a BMS is supposed to:

1. cut charging power on the high side (3.65v per cell) aka HVC
2. cut discharge power on the low side (2.0v per cell) aka LVC
3. balance cells

My experience (which has also been echoed here) is that these cells don't fall out of balance-- actually they tend to group closer together after a few cycles. This takes care of #3. The charger should cut power after voltage hits 3.65v per cell-- this is normal (#1, check!). I think #2 can be taken care of by either LVC setting on the controller or a cycle analyst.

 

[Russell]

So a BMS is supposed to:

1. cut charging power on the high side (3.65v per cell) aka HVC
2. cut discharge power on the low side (2.0v per cell) aka LVC
3. balance cells

My experience (which has also been echoed here) is that these cells don't fall out of balance-- actually they tend to group closer together after a few cycles. This takes care of #3. The charger should cut power after voltage hits 3.65v per cell-- this is normal (#1, check!). I think #2 can be taken care of by either LVC setting on the controller or a cycle analyst.

Most LiFePO4 BMS boards do not open the charging circuit unless a cell hits the over-voltage setpoint which is usually around 3.90V for LiFePO4.

Bulk chargers also don't cut power at a specific cell voltage, most are simply set to a final overall value and this can vary; for example on a 16 cell LiFePO4 charger it can be from 57.6V to 61V depending on the supplier.

Once the charger voltage reaches its final voltage and the current falls to the charger's setpoint it will indicate a "green" or fully charged state. If the charger is of the type where the voltage remains high at the end of a cycle (most I have do) I will generally leave it connected until the current drops even lower to a few mA (on my 10Ah pack). If everything is working properly the shunts of the commonly used "bleeding" type BMS will all be active at this point. It's then time to remove the charger and let the shunts lower the individual cell voltages to the BMS designed value, normally 3.60 to 3.70V for LiFePO4.

-R

 

[jondoh]

Gotta watch those "bleeder" type bms don't bleed your batteries to death. This is what happened to one of my cells when the transistors died. The bms became a resistor across the cell.

 

[999zip999]

My Ping 48v20ah has a 60amp bms, I thought it would be good for a high c headway. My battery is at it's limit's with my love for speed 37mph which is fine, but hills are hard on the battery. I like what I hear about headway, but if had more amps. I would be using more watts hrs. And my battery would run out fast ? Yes. If I understand jondoh, I could use the ping chargher 61.1v for top leve and use a ca for cutoff with a cheap alarm for lvc. then i'm good to GO. My batt. comes hot off the chargher 61.1v at 3.9v to 3.85v per cell let the bms eat it down to 3.6v-3.59v per cell 56.9v for the batt., but after short distant it is realy 53.9v. Now this are proper numbers for headway or thundersky. Plus my bms cutout two days ago at 50.9v ? I guess just one cell mite have gone low ? for bms to cutoff.

 

[magudaman]

I think headways need balancing. I run 32 of them in my 16s2p pack and find that if I don't let them balance out I will hit my LVC earlier on rides than would normally be expected. I do enjoy running them occasionally to 100% dod

A BMS is a good fail safe protection so you don't have to baby your batteries all the time. Worrying while discharging, worrying while charging, worrying while shorting that wrench out as it blows molten metal in your face before that slow blow fuse you had goes out. That just get old. I enjoy being able to War charge anywhere and everywhere at 20 amps 100% in about an hour which no single cell setup I know of can do. Again not having to bring a volt meter with me and constantly watch it. It gives me a piece of mind and I feel will result in a longer life of my batteries.

I think #2 can be taken care of by either LVC setting on the controller or a cycle analyst.

If you are going anywhere near your 100% mark this doesn't always provide an accurate measurement. One single cell could be lower than all the others and could be damaged before you reach your LVC or see that voltage drop on your CA. For example if all your cells are at 2.5v and one is at 1.5v that is only a 1v difference in a series pack, not too easy to catch.

 

[999zip999]

Magudaman, What bms do you run, what's the cutoff with the headway? So i can use the ping chargher at 61.1v it's cutoff without a bms.Or if one cell is low it mite charges one other cell past 4.0v and may cause damage ? I guess it has no way of keeping up with the level of each cell during dischargh ? Only after the chargher hits 61.1v and the chargher unpluged does the bms balance and not during charghing ? Is this correct ?

 

[dogman dan]

A bms is a way to avoid having to do a lot of manual monitoring of what is going on. But trusting it can be risky.

A lot depends on your use pattern, but backing up any bms with some kind of monitoring is not a bad idea. A smart use pattern should leave the pack pretty balanced all the time, and never get close to overdischarge. My use pattern is theoretically that way, till I get a good headwind on the uphill ride home. Then I need every watt and the bms better work.


I rode around clueless to my packs voltage for years and got lucky I guess. No problems in 600 cycles. But one bad part in that bms, and at least one string of cells would have gotten ruined. How you know you can trust your bms without risking the pack I don't know.

 

[BikeFanatic]

Plus my bms cutout two days ago at 50.9v ? I guess just one cell mite have gone low ? for bms to cutoff.

I would bet that one cell group dropped off and that is why the Bms shut off early.
However as was mentioned earlier in the thread the pack will jump up to normal voltage at rest. I may have recommended a Turnigy watt meter so you can
monitor your pack while riding, they are cheap. Then maybe you can observe your pack sagging under load ( I have a 48 volt 15 AH ping I watch it sag from 53 to 50 or from 51 to 48 under load.)You can get a watt meter for about 30 $ at hobby king.

 

[magudaman]

Magudaman, What bms do you run, what's the cutoff with the headway? So i can use the ping chargher at 61.1v it's cutoff without a bms.Or if one cell is low it mite charges one other cell past 4.0v and may cause damage ? I guess it has no way of keeping up with the level of each cell during dischargh ? Only after the chargher hits 61.1v and the chargher unpluged does the bms balance and not during charghing ? Is this correct ?

I am running a BMS I purchased from Ecitypower.com. It was modified by them to allow for 150 amp intermittent and 100 amp continuous, cost around $80 . I'm really not sure what the pack level cutoff is but it will shut down the charger if one cell goes over 3.9v. When the pack was new and out of balance this actually saved me on that first charge for overcharging one cell.

It is possible with out careful monitoring and balance that you could over charge cells when using your ping charger.

Balancing can occur while your charger is connected. I think mine starts to trim cells once they reach 2.7v and bleeds them down to 2.65v .

 

[jondoh]

I think mine starts to trim cells once they reach 2.7v and bleeds them down to 2.65v

probably meant "3.7v" and "3.65v".

 

[thedarlington]

so what lvc, charger and battery full cutoff, are you all using? how about to monitor discharge rate? i know a BMS is a good thing but it like driving auto when you could be driving stick, and its always nice to be aware of just whats going on with ur bike especially if it can help advert failure by a bad bms

 

[jdcburg]

Hi thedarlington - I answered a similar question in a previous post: "I bought my LiFePO4 packs without a formal BMS, as some riders propose that LiFePO4 are the most forgiving and that, with care, a rider can be his own BMS. While that may be true, I use a pair of relatively inexpensive devices to be my BMS and they seem to work well. The first is a Celllog 8M, which is essentially an 8 channel voltmeter with fully adjustable alarms for high voltage (pack/individual cell), low voltage (pack/cell) and differences in voltage between cells. The alarm beeps and the affected cell indicators flash when the alarm is triggered. The Celllog also has an external alarm circuit that will trigger a louder alarm or LED. The alarm circuit can also be used to turn off the charger when any cell or the whole pack reaches a preset limit or turn off the controller if a cell or the pack gets too low. As I use an 8 cell (24V) pack, I use one of these. I don’t bother with the external alarms/shutoffs at this point. If I know I might be running low, I keep a closer eye (and ear) on it while riding. They are under $15 from HobbyKing Online R/C Hobby Store: http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=10952. It requires a 9 pin JST-XH connector. I got 5 connectors for $10 (inc s&h) on ebay.

"The other device I use is a Turnigy Accucel 8150, $42.44 from Hobby King as well - http://www.hobbyking.com/hobbyking/store/uh_viewitem.asp?idproduct=7523. This is a balancing charger that uses the same JST connector as the Celllog to monitor the battery while charging. It is fully programmable for charging all kinds of Li batts as well as lead, NiCad and NiMH. It monitors each cell (up to 8 ) while charging and brings them all up to the proper voltage (3.60V for LiFePO4) without overcharging. It requires an external 11-18V power source. I originally used an SLA and charger for its power but since have adapted an old 300W computer power supply. It charges at up to 7A, so if I use 16 ah I can get it mostly charged in a little over 2 hours, although a full charge with balancing takes longer.

"Hobby King is in Hong Kong and shipping is slow and relatively expensive, or faster and more expensive. I ordered them separately and they took 2-3 weeks each to eastern US. S&h for the Celllog was about $5 and the charger was about $19 s&h. Add a few bucks for standard connectors and wire and you’re all in for under $100. If you use 36 or 48 volts, buy 2 Celllogs and divide the pack for charging." Only you can decide if you would be comfortable with this type of system but it does give you some flexibility - jd

 

[wookey]

I've got a 16S3P headway pack and have been running it for 9 months without a BMS (cos about 9 months ago the old G/F BMS boards went out of stock and the replacement would be ready 'next week' . It needs weekly balancing as some of the cells go out of balance and would soon get overcharged even using a charger that cuts off on a very conservative overall voltage. I'm just using a voltmeter and a 10W resistor to bleed off the excess and it takes about 40 mins each time. I've just bought some celllogs as I'm getting bored of the totally-manual version, and hopefully these will go into the BMS when I finally get one.

I can (fortunately) not worry about LVC because I have loads of spare capacity so never get more than about 50%-60% discharged. Anyone getting anywhere near LVC realy does need something automatic IMHO.

So whilst one can run without a BMS, that's only because you can put off doing the BMS job for a few days, but something/someone does have to do it reasonably often.