Charging pack to 41V experiment

[WheezyRider]

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 forum suggests this may not be such a serious problem, but this has not been proven with hard data.

So I have this pack I built from what are quite likely to be fake HG2 cells:




As I thought the cells may be fake, I added connections to the balance wires so that the voltages of the individual banks can be checked at any time, which makes the pack ideal for this experiment. I have used it for a few cycles so far, but always charged to 41.7V of the SANS charger from Yose Power. I last used it a few weeks ago and it has been in storage at 38.1V

I modified the SANS charger as advised by Nealh to provide an output of 41V

Initial state of banks:

1: 3.82, 2: 3.81, 3: 3.81, 4: 3.81, 5: 3.81, 6: 3.82, 7: 3.82, 8: 3.82 9: 3.82, 10: 3.82

So a maximum bank to bank variation of 0.01V

Pack voltage 38.1V

After charging:

1: 4.1, 2: 4.09, 3: 4.09, 4: 4.09, 5: 4.09, 6: 4.1, 7: 4.1, 8: 4.1 9: 4.1, 10: 4.1

Pack voltage 40.09V

The bank to bank variation followed a similar pattern to that before charging.

I will use the pack and cycle it several times and report back any issues from only charging to 41V

For reference, here are the BMS specs, it is for the 40A version:

 

[Nealh]

Well done for giving it a go, although the BMS balance function won't be used you still get the useful lvc to help protect against discharge, as long as cells don't wander to much then balancing shouldn't be needed 0.01v is acceptable and probably 0.02v as well any more then that then balance.
My pair of 29E's are hard case so I can't keep opening and checking so am relying on intuition and judgement/knowledge of the pack characteristics on keeping an eye out for issues. So far over the year I haven't felt the need to balance yet as the range and discharge voltage is good with no issues occurring not bad as these are well in to year 6.
The guy's on ES are pretty adamant that lowering top charge ekes out the cycle life and capacity, some only charge to 4.0v per cell and claim that cycle life is double and 4.1v about 50% more cycles.
I'm all for trying but monitoring has to be carried out, so is not practical for most users as they don't simply understand li batteries. From my monitoring of voltage the use between 4.2v - 4.1v is paltry and I find that top 0.1v or 1v in the case of a 10s battery is depleted within about 1.5 - 2km of use so holds very little capacity.

 

[cyclebuddy]

The guy's on ES are pretty adamant that lowering top charge ekes out the cycle life and capacity,

I came across this charger recently, which I thought interesting: They call it a "cycle-life charger" as it's already supplied with a fixed voltage of 41.3v (3A current).

Fuyuang Cycle-Life Charger 41.3V 3A for 36V Batteries Li-Ion

Fuyuang FY-4203000 fast charger CCCV 125W (41.3V / 3A charging current) for 36V eBike batteries, powered by Enerpower. Fuyuang Chargers are…

enerprof.de

 

[Nealh]

I prefer my Sans type which is adjustable or the Meanwell HLG's, they don't limit your charging options.

 

[cyclebuddy]

The really nice thing about the Fuyuang chargers is they're Class VI - 90% efficient, so they run cooler (Sans aren't very efficient).

The casings are also much smaller and more robust than Sans, the cables more flexible, and they come in 2/3/4 amp (in regular 42v versions), which makes them ideal for those looking for a fast-charge travel/back-pack charger for pub/cafe/over-lunch charging. But no, you can't open them up to twiddle with output voltage like the Sans (not that I can see anyway).

 

[Nealh]

If you modify the Sans they run quite cool, I drilled about a dozen 8 - 10mm holes in mine for top and thru air flow to allow the heat dissipation (the one thing they lack).

 

[RossG]

I re-built my Sans charger, I fitted a much larger heat sink and put the whole thing in a bigger case I made myself.
In one end of the case fitted inside is a miniature cooling fan the other end has a hole cut in it with a plastic grille covering it to keep out dust, the fan whizzes round sucking in air and it vents out the other end.
In the top of the case I fitted a meter to display SOC/Voltage etc. Bit of overkill but it works very efficiently and doesn't get even slightly warm.

 

[WheezyRider]

The really nice thing about the Fuyuang chargers is they're Class VI - 90% efficient, so they run cooler (Sans aren't very efficient).

The casings are also much smaller and more robust than Sans, the cables more flexible, and they come in 2/3/4 amp (in regular 42v versions), which makes them ideal for those looking for a fast-charge travel/back-pack charger for pub/cafe/over-lunch charging. But no, you can't open them up to twiddle with output voltage like the Sans (not that I can see anyway).

What is the efficiency of a Sans charger? Do you have a data sheet?

There isn't a huge size difference. The Sans is about 12mm wider and about 17 mm longer. The Sans weighs about 345g (not inc. mains cable). What is the weight of the Fuyuang charger?

I have cut the cable from my Sans quite close the box and soldered on an XT60 connector, so the cable isn't an issue for me.

The Sans does still use significant power when it is not connected to the battery which is not ideal. Is the Fuyuang better in that respect?

For now I'll stick with the Sans, a 3A charger would be nice, and if it's more efficient great, but hard to justify the extra expense at the moment when the Sans does the job and I've got 2 of them already.

 

[cyclebuddy]

What is the efficiency of a Sans charger?

You need to measure power factor to determine efficiency. You can do that easily/cheaply with a power meter (ebay). The last time I did, my Sans measured >68%, a Bosch measured better at >80%, and the Fuyuang measured at >89%. There are quite a few different 2A Sans models internally though that all look the same externally (same casing). Yours may measure differently.

There isn't a huge size difference. The Sans is about 12mm wider and about 17 mm longer. The Sans weighs about 345g (not inc. mains cable). What is the weight of the Fuyuang charger?

My 3A Fuyuang weighs 500gms and my 2A Sans weighs 700gms (very roughly) with all leads and plugs needed (metal 5-pin XLR's). But you may be comparing apples with pears for size: This pic shows a 3A Fuyuang against a 2A Sans. The 2A Fuyuang is smaller still. It's not a huge size difference, but the thinner/ more flexible cables make it easier/less wieldy and the more robust casing helps for stuffing in my back-pack.



I have three 2A Sans chargers, which I've uprated the caps on, and which all work fine. The Fuyuang I bought specifically for day trips: It's quicker for free power in pubs and when working off the inverter in the van. My three e-bike batteries are only 11A, so extra top-ups are sometimes needed on all-day rides. The specs say the batteries will take up to 4A charge current which Fuyuang also do, but I thought that might be pushing my luck.

Full Fuyuang spec sheets (inc efficiency and sizes - you'll see the 2A is smaller still than the 3A in the photo).

Chargers - Fuyuang Germany

Fuyuang products can be used in a variety of industries including electric vehicles, LED lights, medical devices, robots, audio and power tools.

fuyuang-germany.de

 

[WheezyRider]

Just a quick update.

I'm up to 6 cycles now and this evening when I got home, all banks were 3.66V before I put the pack on charge. I had hoped to do more cycling last week and get more charge/discharge cycles in, but circumstances conspired against me. Should be able to get a decent amount of cycling in this week.

 

[Nealh]

Just a quick update.

I'm up to 6 cycles now and this evening when I got home, all banks were 3.66V before I put the pack on charge. I had hoped to do more cycling last week and get more charge/discharge cycles in, but circumstances conspired against me. Should be able to get a decent amount of cycling in this week.

You have v,g matched cells there Wheezy.
My last charge to 41v on my 6 yo 29E battery accrued 42 miles down to 35.5v straight off the bike so would have still got a good few more miles if I rode it down to LVC of about 33v. Resting voltage 35.7v next day so holding up v,well, £ for £ the 29E is regarded on ES as the top 18650 cell. The cycle ageing test also revealed it to be a top cell.

 

[WheezyRider]

You have v,g matched cells there Wheezy.
My last charge to 41v on my 6 yo 29E battery accrued 42 miles down to 35.5v straight off the bike so would have still got a good few more miles if I rode it down to LVC of about 33v. Resting voltage 35.7v next day so holding up v,well, £ for £ the 29E is regarded on ES as the top 18650 cell. The cycle ageing test also revealed it to be a top cell.

Yes, I'm really surprised. These have got to be fake cells, they can't be genuine HG2s, they certainly don't have the 3000mAh capacity of HG2 cells. However, crappy as they are, they seem to be equally crap!

This morning after recharging they were all at 4.1 V except bank 6 which was at 4.11. Then I ran the pack down to 37 V on my way to work. All banks were at 3.7V. It's charging again now.

 

[Nealh]

0.01v isn't too much to worry about and to be that good they must still be a good faked cell.
Although you think they are fake wrapped HG2's they may still be genuine real McCoy LG's, some rewraps have been known to be LG HE4's or HE2's 2500mah cells although the easiest cheapest way to clone a HG2 they are still 20a rated cells but at half the cost.

 

[WheezyRider]

0.01v isn't too much to worry about and to be that good they must still be a good faked cell.
Although you think they are fake wrapped HG2's they may still be genuine real McCoy LG's, some rewraps have been known to be LG HE4's or HE2's 2500mah cells although the easiest cheapest way to clone a HG2 they are still 20a rated cells but at half the cost.

I think you are probably right, re-wrapped HE4s or HE2s. The capacity of this pack isn't much different it seems to my two Yose Power 15Ah packs with Chinese cells, maybe even slightly less. No way are these 3000 mAh. But at least they seem consistent. I have been hammering the pack quite hard, up to 28 Amps or more at times, but the cells still seem to be balanced so far.

 

[WheezyRider]

Ok at more than 10 cycles now.

After overnight charge, all banks this morning at 4.1V, except banks 6 and 10 at 4.11V

Got to work this morning, took 6.4 Ah out, thrashing at up to 28A peak, all banks at 3.67V.

Getting boring, I was hoping to see more happening by now...

 

[Nealh]

Be pleased you haven't a total lemon, as I mentioned before you have probably v.g HE2/4 cells rewrapped/faked as the HG2. 0.01v with no balance is excellent imho.

 

[WheezyRider]

Be pleased you haven't a total lemon, as I mentioned before you have probably v.g HE2/4 cells rewrapped/faked as the HG2. 0.01v with no balance is excellent imho.

Getting bored in a good way! Relieved it wasn't money down the drain lol

 

[WheezyRider]

Just taken the pack off charge. All banks 4.1V, except Banks 6 and 10 at 4.11V

My second Yose Power pack is having trouble with the key switch, so it looks like I am going to have to dismantle that soon and repackage it. It'll be interesting to do the same mod so I can access the voltages of the cell banks externally and see if that pack is as well behaved when charging to 41V

 

[WheezyRider]

My testing has ground to a halt recently as due to health issues, I can't ride at the moment.

However, I found some interesting info on the Grin web page. They claim that only charging to 4.05 V can extend the life of lithium batteries to approx 1500 cycles. Also they claim that good quality cells normally stay balanced for > 100 cycles.

What I wonder though with these cycle life plots is the extent of depth of discharge. How damaging is this in comparison with charging to >4.1V? Would you really get 1500 cycles discharging down to LVC, or even 20% of capacity? This is something they don't go into.





(from: https://www.ebikes.ca/product-info/cycle-satiator.html )





(from: https://www.ebikes.ca/tools/charge-simulator.html )

 

[WheezyRider]

After my last ride back at the beginning of August, I left the pack at about 37V, while I rested up.

Last week was my first tentative ride. I charged to 41V and all cells were still within 0.01V of each other. I rested again for a week. Since then I've had another 5 charging cycles from 41 to 37V and back to 41V. All banks are still within 0.01V

So, looks like the pack is doing better than I am