Thundersky LFP activation charge why ?

[evtdriver]

Hi,

first I have to thank this forum for all information I can find here... I like when advices are result of facts and proven.

I just a bought some LFP cells for my EVT4000 from EV Components. They say to charge the cells at 4.2v for the first time. Do you know why the thundersky cells need that ?
Is it really safe to do ? I'm really dummy with batteries and I don't want to do anything wrong with the cells. I suppose I will need to use a balancer charger or BMS as we don't want any cell to go higher than 4.2v...

Thanks,
Jerome

 

[olaf-lampe]

For the EVT4000 you only need 16 cells. It´s easy to charge them individually up to 4.2V with an adjustable power supply.
Once they are initialized, you can bulk charge them with a common LiFePO-BMS. In case you have a BMS with programmable upper cell-voltage limit, you don´t need to charge them individually.

BTW: I have to do this individual charge procedure next week with 100 cells. No fun at all
-Olaf

 

[cell_man]

If that's necessary why isn't it done before shipping and how can they be properly tested? Not knocking, it just seems a bit strange to me.

 

[olaf-lampe]

@cell-man
Isn't it common that they are factory charged to something like 50%?
-Olaf

 

[liveforphysics]

@cell-man
Isn't it common that they are factory charged to something like 50%?
-Olaf

They SHOULD definately have a minimum of 1 full charge cycle from the factory to keep things chemically in there place. The optimal storage voltage for all types of lithium is right at the nominal voltage, or 50% DOD. Higher means less shelf-life, lower means less shelf-life, though anywhere in the 3.1-3.3v area for LiFePO4 should be about the same, it's just when stored at 0% or 100% that it takes a very sharp hit in shelf-life.

That said, the guys at EV components are great guys with a lot of experience dealing with the TS cells, and if they've found this tends to cause better pack building results, then I would trust them. If it really is the first time the cell has ever been fully charged, that seems pretty shady of TS to ship things that way.

To save time, if you've got a strong CC/CV 4.2v power supply laying around, tie them together with some jumper leads and batch charge in groups of 20-30 or whatever. That way at least you don't have to be connecting and disconnecting leads every hour for days. lol

 

[cell_man]

@cell-man
Isn't it common that they are factory charged to something like 50%?
-Olaf

As luke said I'm surprised they haven't been cycled at least once to confirm capacity and various other parameters. I know that when small cells are assembled into a pack by a good pack builder the cells are matched as closely as possible. It just seemed a bit strange to me, but they are very reasonably priced so you can't have everything.

 

[evtdriver]

Thanks so I will do it...

I plan to use 2 DB8 balancers and 2 king power chargers. These chargers are not easy to adjust... So maybe I can start to charge the pack to 3.6V and then continue to charge each cell to 4.2V using a small single cell charger for lipo ? Is it correct ?

 

[AndyH]

I'll bet the initial charge isn't to do anything for the cells, but is so that the pack starts out 'balanced'.

TS ships cells at roughly 50% SOC...very roughly. Scooters made in China and imported have packs built with the still 50% SOC cells and it's a real pain to get that initial full charge without single cell chargers.

It's a whole lot easier to charge them singly and be ready to ride from that point on.

Andy

 

[RemoteContact]

Well if voltage says anything (which I know sometimes it doesn't) all 32 of my 60ah TS cells shipped at exactly 3.31v from EVcomponents. Cells have been great so far, worth the 19 week lead time.

 

[cell_man]

I'll bet the initial charge isn't to do anything for the cells, but is so that the pack starts out 'balanced'.

TS ships cells at roughly 50% SOC...very roughly. Scooters made in China and imported have packs built with the still 50% SOC cells and it's a real pain to get that initial full charge without single cell chargers.

It's a whole lot easier to charge them singly and be ready to ride from that point on.

Andy

That would make more sense and yes agreed it is a real PITA to balance a pack with a typical BMS that has cells that are way out of balance.

 

[olaf-lampe]

Thanks so I will do it...

I plan to use 2 DB8 balancers and 2 king power chargers. These chargers are not easy to adjust... So maybe I can start to charge the pack to 3.6V and then continue to charge each cell to 4.2V using a small single cell charger for lipo ? Is it correct ?

It's pretty useless to charge the pack to 3.6V at first, because this will take forever when the pack is out of balance.
If you charger has a Li-Ion 4.1V charge limit, this would be better than Li-po. Lipo sometimes goes up to 4.25V which IMO is a risk.

If you put them all in parallel to charge them at once, make sure they all get a good low resistance connection. Use a huge copper bar on each pole and connect the charger with several leads of the same length. This spreads the current equally to all cells.
If you do this with a small single cell charger ( doesn't make sense ), make sure the current goes down to 20% of the start current before you disconnect. This can take a few days depending on the charge current.

-Olaf

 

[dak664]

I'll bet the initial charge isn't to do anything for the cells, but is so that the pack starts out 'balanced'.

TS ships cells at roughly 50% SOC...very roughly. Scooters made in China and imported have packs built with the still 50% SOC cells and it's a real pain to get that initial full charge without single cell chargers.

It's a whole lot easier to charge them singly and be ready to ride from that point on.

Andy

That would make more sense and yes agreed it is a real PITA to balance a pack with a typical BMS that has cells that are way out of balance.

My 36 headway 10AH cells from the group purchase all read about 3.3 volts but were ~3AH out of balance. Balancing at 100ma using the headway BMS did take a couple of days. A faster way, If you can monitor each cell voltage, is to charge at 300-400 ma and clip a 1 watt 10 ohm resistor across each cell as it comes up. When the last cell fills remove all resistors and finish at 100 ma.

 

[ahambone]

I suspect the "charge cells to 4.2V to break in" procedure is just as was said earlier - to make sure users balance their cells before making packs out of them. An implicit top-balance if you will.

When I got my 125 Headway 38120S cells I placed them in parallel groups of 5 with short runs of wire and then charged the parallel groups with a VoltPhreaks charger. That brought each set of cells of to ~3.7 volts and balanced them along the way. The VP chargers are 2 amps max (observed to be ~1.73 to ~1.75 amps in reality) so balancing/charging a 50 Amp-Hour group usually took around 24 hours. I had 3 of these rigs running in parallel and got the initial balancing charge done in about a week.

I've had vary few balance issues since then and the G/F BMS has worked very well with the pack since it started out balanced. I'd bet the 1/2 amp shunting of the BMS would have taken several hours to days to balance the pack without the initial balancing charge.

Cheers,
--Adam

 

[evtdriver]

Ok and what do you think using a RC chargers like Turnigy 8150 to charge-balance the LFP packs. They are damn cheap and offer many features : monitoring, store charge, discharge, many chemistry supported.
Downside is they can't output more than 5A with 8s cells. (150w max) Is this charger really able to work with large cells ?

 

[evtdriver]

Well if voltage says anything (which I know sometimes it doesn't) all 32 of my 60ah TS cells shipped at exactly 3.31v from EVcomponents. Cells have been great so far, worth the 19 week lead time.

Mine were at exactly 3.30v. Did you balance them at 4.2v before to use them ?

 

[GGoodrum]

I agree with Andy that the real reason they say to charge them to 4.2V is they are trying to get the cells as closely balanced as possible, but this won't do it, in my opinion, and it can't be good for the cells, I don't give a rat's ass what Jack Ricard says.

Here's what I would do with new cells. First, if the cells are roughly at the same voltage, just connect them all in parallel, and leave them that way for an hour or so, That will definitely balance them to the exact same voltage. Then, connect them in series, and use your CC/CV charger/supply which should be set to 3.60-3.635V per cell, so if you have 16 in series, it would be around 57.5-58V.

-- Gary

 

[liveforphysics]

I agree with Andy that the real reason they say to charge them to 4.2V is they are trying to get the cells as closely balanced as possible, but this won't do it, in my opinion, and it can't be good for the cells, I don't give a rat's ass what Jack Ricard says.

Here's what I would do with new cells. First, if the cells are roughly at the same voltage, just connect them all in parallel, and leave them that way for an hour or so, That will definitely balance them to the exact same voltage. Then, connect them in series, and use your CC/CV charger/supply which should be set to 3.60-3.635V per cell, so if you have 16 in series, it would be around 57.5-58V.

-- Gary

Winner!
Simple, and perfect.

 

[olaf-lampe]

Ok and what do you think using a RC chargers like Turnigy 8150 to charge-balance the LFP packs. They are damn cheap and offer many features : monitoring, store charge, discharge, many chemistry supported.
Downside is they can't output more than 5A with 8s cells. (150w max) Is this charger really able to work with large cells ?

From what i know, these RC chargers have a timeout cutoff. After 1hour10min they must be restartet to charge bigger packs. Beside that, they are fine. Be sure, you have a powerful Ac/Dc adapter at hand. Usually they work with a 12V starter battery. But then again you could also buy a 5V Ac/Dc adapter that can be regulated to 4.1V and charge directly :lol:

Having to many options doesn't make life easier
-Olaf

 

[evtdriver]

Having to many options doesn't make life easier

Agreed

I read the timer can be changed and disabled. So maybe it worths to have a look at this.

 

[MikeB]

From a posting about Thundersky LFP cells over at the V:
http://visforvoltage.org/forum/6691-xm-3000-lifepo4-upgrade

When new, the cells have free liquid electrolyte that can be heard sloshing around when shaken. This free liquid electrolyte goes away after the first one or two charge/discharge cycles. I have since learned that this is due to a portion of the organic solvent electrolyte breaking down electrolytically during the first few charge cycles and forming a "solid electrolyte interphase" on the anodes. This interphase then prevents the remaining electrolyte from breaking down.

The above has two important implications:

1. The proper formation of this electrolyte interphase is important for good cell life and keeping the internal resistance low. This may be the purpose for the Thundersky-recommended procedure (page 31 of the manual) of initially "conditioning" each cell, by charging it at 1/3C, discharging it at 1/3C (to a minimum of 2.5 V but I'd stay higher), then recharging at 1/3C. Unfortunately, due to the confusing bad English in the manual, many people miss or mis-interpret this recommendation and think it is referring to kind of optimum procedure for regular use.

2. The cells should be mounted upright, for at least the first few charge cycles, (in my opinion, all the time) or some of the anode and cathode material may not be submerged in the electrolyte, inhibiting proper interphase formation.

I would guess the purpose of the activation charge is part of the "conditioning" process, which seems to solidify the free electrolyte.

 

[evtdriver]

That's true I can hear the electrolyte when shaking the cells... I guess 1/3C is a maximum value so using lower current won't hurt ? A RC charger could be very interesting to process and monitor the 2 first cycles.

 

[AndyH]

From a posting about Thundersky LFP cells over at the V:
http://visforvoltage.org/forum/6691-xm-3000-lifepo4-upgrade
<snip>
I would guess the purpose of the activation charge is part of the "conditioning" process, which seems to solidify the free electrolyte.

Hi Mike,

I have no problem with the information on the initial formation charges. PSI performs those initial charges at the factory, the leaves the cells at about 50% SOC for shipping. I expect, but don't know for certain, that TS would do the same.

The info on page 31 of the TS manual (available hereand attached) says:

"Please do not use the new cells directly because they load half power when they are produced. You need to charge each cell following its charging standard with suited charger."

In other words - charge first with an appropriate charger before first use.

There's no magic to the 1/3C charge for first use - this is the standard TS charge and discharge rate.

Andy

 

[MikeB]

Thanks for the manual link, I hadn't looked at it directly before, was just repeating what I'd read elsewhere.

In other words - charge first with an appropriate charger before first use.

There's no magic to the 1/3C charge for first use - this is the standard TS charge and discharge rate.

Looking at page 31, I'm not sure I agree with this statement. The yellow box at the bottom of the page reads "After first charge and discharge please setup the charging and discharging voltage of different types of battery according to it's standard".

This statement seems to imply that the 'first charge and discharge' are to be done at something other than 'standard' voltage, and that 'first charge' voltage is listed in the table in the middle of the page. I think the interpretation of an initial 'conditioning' cycle, as described by the original poster at the V, seems more likely. The first (conditioning) charge cycle is done at one standard, then regular usage follows a different standard.

 

[liveforphysics]

Ok and what do you think using a RC chargers like Turnigy 8150 to charge-balance the LFP packs. They are damn cheap and offer many features : monitoring, store charge, discharge, many chemistry supported.
Downside is they can't output more than 5A with 8s cells. (150w max) Is this charger really able to work with large cells ?

From what i know, these RC chargers have a timeout cutoff. After 1hour10min they must be restartet to charge bigger packs. Beside that, they are fine. Be sure, you have a powerful Ac/Dc adapter at hand. Usually they work with a 12V starter battery. But then again you could also buy a 5V Ac/Dc adapter that can be regulated to 4.1V and charge directly :lol:

Having to many options doesn't make life easier
-Olaf

For every RC charger I've got, it's as simple as Menu/setting/time out/OFF

 

[AndyH]

Thanks for the manual link, I hadn't looked at it directly before, was just repeating what I'd read elsewhere.

In other words - charge first with an appropriate charger before first use.

There's no magic to the 1/3C charge for first use - this is the standard TS charge and discharge rate.

Looking at page 31, I'm not sure I agree with this statement. The yellow box at the bottom of the page reads "After first charge and discharge please setup the charging and discharging voltage of different types of battery according to it's standard".

This statement seems to imply that the 'first charge and discharge' are to be done at something other than 'standard' voltage, and that 'first charge' voltage is listed in the table in the middle of the page. I think the interpretation of an initial 'conditioning' cycle, as described by the original poster at the V, seems more likely. The first (conditioning) charge cycle is done at one standard, then regular usage follows a different standard.

Maybe. The top of the page also says to use the charging standard for the initial charge. Notice that the text mid-page lists 4.25V and 2.5V for the volgage limits and 0.3C for change and discharge.

Earlier in the doc they say max charge is 4.30. One might think, then, that the initial charge should only be to 4.25V - lower than the 4.30V max.

The initial 'formation' charges should have been done at the factory, so while the V posting might be technically correct, it should be outside the control of the end consumer.

The closest parallel I can come up with off the top of my head would be a note in a car owners manual outlining the parameters for the first engine run on the test stand. Nice to know, but that'll already be done when I get the car...

I understand that TS answers email - that might be the best place to ask...

Andy