My Hybrid battery LiMn + LiCo. First test w/data

[CanadianWinters]

Hi everyone.

Some weeks ago I purchased some Makita BL1830 packs from a ES member.

I tested and recovered them and then made a 15s4p (55.5v 6Ah nominal--->60 cells) pack and tried it on my ebike. It worked great!



I then decided to dismantle my 10s12p li-ion pack made with laptop cells and make a hybrid battery with the Makita cells.

I made a 15s3p (55.5v 6Ah nominal -->45 cells) with the laptop cells.

Charged both packs to the exact same voltage (62.0v) and then connected them in parallel.

It took me a lot of time to dismantle the 10s12p pack and rebuild it to fit the battery box but I am pretty happy with the result.



I also added a pre-charge/anti-spark resistor (380ohm 3.75W), so far works great on different setups:

View attachment 4

Bike data:

*Rear Bafang 500W high torque from BMSbattery
*26" wheels
*30 Amps (32Amps in reality) 36v KU123 controller (63v caps inside)
*GT power meter 130A max (read max 60V but I've been lucky, reads my 62v !!)

So I took the bike out for a ride today. Unfortunately I only have 2 Cell-log 8s so I can log the data only from 5 cells of the makita pack and 5 from the laptop pack.

During the ride pedalling was 0%.I tried to go WOT all the time but I was going too fast on the bike path. It was lots of fun, very quick acceleration. I am pretty impressed with the results, after 10km I still hit 40kph+ no problem.

Data:




Notice 45.5kph after 400m (a little downhill though)


Laptop cells log:


Here I seem to have a problem with a weak laptop cell.


Makita cells log:


So far I am pretty impressed with the results. I am not sure about the voltage under load. I read 53.5v during acceleration in the last km of the ride, quickly climbing up to 56-57v (I wish I had a camera to re-view what happened).
What is the safe limit?

 

[dogman dan]

Though the chemistries are different, both charge to a similar voltage and you are using the lower of the two.

I can't see a problem with it, provided you keep an eye out for high charged cell groups in either type. You didn't say exactly how you charge, but fraidy cat that I am, I'd want either good cell level HVC's or at least use of an RC charger that allowed bulk charging with a very postitive ramp down at the end of charge and cutoff. I feel that bulk charging 8 cell groups at a time with a charger that tapers the amps at the end is safer than 15 at a time with a crude bulk charger with lico, unless you have the cell level hvc's. Charging the lico to only 4.1v helps make it safer of course.

The only time I did a hybrid pack is was lead nicad. I charged both with the same nicad charger. So the nicad charged extra slow, and stopped at the time it needed to when the cells warmed up. The lead just went along to the same voltage since it was paralelled. They just both charged away, and finished at the right time for the nicads. It may have very slightly overcharged the sla's but they tolerated it well.

 

[docnjoj]

I use a hybrid pack on my wifes trike. I //6 Fatpacks (also in //) with an old 10AH Ping. Works fine as the fatpack group have an LVC from kfongs switch and of course the Ping has the old BMS. No problems after over a years use. Go for it! The Fatpacks and the Ping are charged with 2 separate chargers.
otherDoc

 

[CanadianWinters]

@ eTrike. You're welcome. I have been reading about hybrid packs in this forum but the info is all spread out in different threads. I was curious about the performance of those packs so I decided to make one myself. I am pretty happy with the result and I hope it stays this way .

@Dogman.
I see what you are saying about charging with a RC charger instead. Well..I tried. I have an Imax B6 and it's the worse balancer..ever! To the point that's better to use the bulk charger instead. I tried several times using the B6 on the laptop pack and on the makitas but I always get one or two cells out of balance. This seems to happen when the charger start to get warm.
The makita seems to tolerate the bulk charge very well. The laptop cell a bit less, I normally balance manually every month or so (that's the average of when they start to get unbalanced), and I keep an eye on the cell voltage while charging.

I am using a 240W ecitypower/BMSbattery charger. I changed the voltage to 62v and current to 2.5A, previously was set to 41V & 5A.

8)

 

[dnmun]

the 62V may be a problem for the charger. the schottky diode on the charger may only be rated to 60V so it could fail. but you can get larger one later if not too much blows up.

you coulda left the current at 5A or even pushed it up to 6A.

 

[CanadianWinters]

the 62V may be a problem for the charger. the schottky diode on the charger may only be rated to 60V so it could fail. but you can get larger one later if not too much blows up.

you coulda left the current at 5A or even pushed it up to 6A.

Good to know dnmun. I think this chargers are good up to 72V:



I have a spare one just in case :wink:


@ docnjoj

That's great! // means parallel right?
I tried to introduce my wife to electric bicycles but she does not like it, she prefers to exercise with a normal bicycle :?

 

[etriker]

Awesome ! ! ! Thanks for sharing !

 

[dogman dan]

I can see a b6 not being an awesome balancer. Mine work tolerably well. My other RC chargers don't balance as accurately, but both do good enough for me.

I just meant that any RC charger, used as a lower cell number bulk charger might be safer than something hacked from a power supply, charging a larger cell number.
No way a b6 is ever fast enough for much more than charging your headlight battery.

Or am wrong? Does a modified meanwell taper down the charge to a lower rate as it reaches full?

In any case, Now I see you use an actual charger to bulk charge, which would taper the charge amps towards the end. Excellent. I try to never forget, that as we have conversations here, a zillion lurking noobs are reading. So you need to spell it out, what is safe, what is riskier.

 

[etriker]

I can see a b6 not being an awesome balancer. Mine work tolerably well. My other RC chargers don't balance as accurately, but both do good enough for me.

I just meant that any RC charger, used as a lower cell number bulk charger might be safer than something hacked from a power supply, charging a larger cell number.
No way a b6 is ever fast enough for much more than charging your headlight battery.

Or am wrong? Does a modified meanwell taper down the charge to a lower rate as it reaches full?

In any case, Now I see you use an actual charger to bulk charge, which would taper the charge amps towards the end. Excellent. I try to never forget, that as we have conversations here, a zillion lurking noobs are reading. So you need to spell it out, what is safe, what is riskier.

The power supply has to have a higher voltage than the cells for current to flow to charge the cells.

If the voltage on the power supply is the same as the voltage on the cells no current will flow.

Current tapers off because the cells become charged to the Mean Well max output voltage.

I use a Hyperion charger between my Mean Well and battery pack.

 

[CanadianWinters]

Thank you guys for the feedback

In any case, Now I see you use an actual charger to bulk charge, which would taper the charge amps towards the end. Excellent. I try to never forget, that as we have conversations here, a zillion lurking noobs are reading. So you need to spell it out, what is safe, what is riskier.

I think etriker replied to you about the MeanWell. I have never used them so I don't know.

I should have specified about the charger in the first post . Anyway I am going to do another test today, let's see what happens .

BTW I am using this http://www.ebay.ca/itm/1-x-PCB-3-6V-16340-18650-Li-ion-Lipo-Battery-Pack-/271025870874?pt=LH_DefaultDomain_0&hash=item3f1a66981a#ht_1891wt_1016 + old phone charger (5v out 400mA) to balance single laptop cells, I've done it a few times and seems to work pretty well, at the end the voltage was always 4.20v. Better than my Imax b6. If I want a lower voltage I normally discharge with a 3.6v 0.7A light bulb.

8)

 

[dnmun]

i use a small 5 ohm cement resistor, 5 watts and it sucks the excess voltage off the high cells in a little while, sometimes too much and then i have the others cells all balanced and waiting for the one i drained to fill back up.

i doubt if the diode is 60V either because it would already be leaking if you pushed it over 60V. they are pretty sharp on the cutoff. but my 36V and 48V kingpans all use the 60V schottky. my meanwell clone uses a high voltage, 200V, schottky, but 30A!!!!!!!!!!!!

but you can turn the current back up, those chargers are overbuilt that way and it will even go higher than 6A i bet.

 

[CanadianWinters]

i use a small 5 ohm cement resistor, 5 watts and it sucks the excess voltage off the high cells in a little while, sometimes too much and then i have the others cells all balanced and waiting for the one i drained to fill back up.

i doubt if the diode is 60V either because it would already be leaking if you pushed it over 60V. they are pretty sharp on the cutoff. but my 36V and 48V kingpans all use the 60V schottky. my meanwell clone uses a high voltage, 200V, schottky, but 30A!!!!!!!!!!!!

but you can turn the current back up, those chargers are overbuilt that way and it will even go higher than 6A i bet.

Thanks for all the info dnmun, I just wanted to be safe for the first few charges and not take any risk to blow up the charger or overheat the batteries.


Today I went out for another ride.A long one this time, about 25km with plenty of stop & go. Unfortunately the cell-log connected to the makita cells turned off and on a couple of times (the balance lead connector was loose I found out) and it did not record much (2 minutes ).

Here the data:







Peace 8)

 

[CanadianWinters]

Here are some data from another ride. This time I went uphill (incline 4-12%) and downhill. Uphill I could go 35kph (i had to slow down because of traffic), current draw was about 20-25Amps (I read 23Amps for some time). After this "abuse" I was still able to hit 40kph + at the end of the ride (17.4km)

The makita cells seem to like the "aid" provided by the laptop cells. I still have to change a couple of weak laptop cells as you will see from the data.

I am now charging at 63.0V and the system seems to be happy with it.Also the charging current is now 4.0 A.

Data:




Just consider the "3b" row:



Makita perfectly balanced during ride:


Makita at the end of the ride:


Laptop during the ride:


Laptop at the end of the ride:

Still problems with cell number 2. I am going to change it as soon as I have some time.

Has anybody tried to make parallels at cell level?

8)

 

[etriker]

http://www.orbtronic.com/batteries-chargers/imr-panasonic-cgr18650ch-10a-18650-rechargeable-battery-cell-high-drain

Panasonic did it at the cell level ?

 

[docnjoj]

Bosch Fatpacks are 10S2P with similar cells to Makita and work fine.
otherDoc

 

[CanadianWinters]

Has anybody tried to make parallels at cell level?

Nobody seem to have bothered with this test so I did.

What I did:

I tested a 1s2p 4.4Ah LiCo Panasonic (Laptop) battery @ 1C to check its real capacity.(4.12Ah)
I tested a 1s2p 3.0Ah LiMn Sony Konion battery @ 1C to check its real capacity.(2.85Ah)
I made a 1s4p battery paralleling the above cells. Real capacity of this battery is about 6.95 Ah

Then I tested the battery at 1C (6.95A costant load). I extracted 100% of the capacity.

I expected far worse!

The graph:



The cut off for all the tests was set to 2.80v.