How high charging voltage can the 36V Battery Management System handle?

[Cornelius]

I'm planning on mounting some solar panels on my bike-trailer (See my bike here: http://goldenmotor.com/SMF/index.php?topic=2092.0), and i'm wondering if there's a limit on the charging voltage the BMS in the batterypacks from GM can handle?

I got the 36V/16Ah battery from GM, and i'm planning on using three 12V/20W panels (open voltage at around 20V...) This should add up to around 60V open voltage. Can the BMS in the 36V pack handle that?

I understand from reading this thread: http://goldenmotor.com/SMF/index.php?topic=1115.0 that there's ok to charge the battery while draining it; though, the battery indicator will be more or less useless...

[Leslie]

Hi I working on a simuilar project.

I would use a 3X15v and one 12v 5 watts zener diodes in series (58v) and a very small ohm resistor as a shunt in between the panels wires just to make sure it doesnt go over 60v.  The shunt needed to take away 6v off the open circuit voltage from a 20 watt solar is nothing and the zener resistor shunt wont kick in until the volts reach 58v.  Nothing wasted in your charge field but small regulation gained

From 2 years working with solar panels I know the open circuit voltage on the 20 watt panels is around 20v~ 60v for a 36v pack.  60v will never happen unless the BMS is disconnected from the cells.

Most times I just hook them up just before I leave an allow the battery resistance to beat down the voltage, a good LM317t regulator never hurts.

What I have noticed about solar panels is that they give more watts when the voltse are higher.


For instance.

I can get 1.4 amps out of a 4 series string of 20 watt panels (80 watts)  When the pack is low, and the 85v feed from the panels voltage drops to say 46 volts, at 46v it equals 64 watts.  Get what I mean, Panels struggle in the initial stage of the charge.

My 20 watt panels will give up 1.3a at 17v each.  So to make it more practical we need to find a way to convert the wasted energy into current. Not to mention that 17vx4 panels = 68v and between 46v from discharged pack to 68v to nominal panel voltage there is 22v @ 1.4a (30 watts) adding up to nothing in the beginning stages of the charge,.  And then we have the open circuit voltages above 17v.  


I plan on doing a switcher that will bight into the solar panels voltage and switch it down through an inductor.  

The efficiency should be around 85% which you would loose 12 watts,  but with all things considered the gains should outweight the losses.

I want to get at least 18 watts more out of my 4 series solar panels.

Your 36v system should be fine with a simple fet switch and DPDT relay switch and linear regulator.  Use the fet to trigger the relay to switch over to a linear regulator and back.

[Cornelius]

So what you're saying is that the BMS has a limit of 60V input?

Given a certain amount of sunlight, solar panels can give a certain amount of Amps, regardless of 'clamped' voltage I think... Connecting a 12V battery to a 20W panel with 20V open voltage will give 1,2A @ 12V for example; if we add a 6V battery in series with the 12V total of 18V, the panel will still give 1,2A, but now @ 18V...

That's why MPPT solar regulators has come on the market (Maximum Power Point Tracking)... They use DC-DC converters to utilitize the amps available at the highest possible voltage.

[Leslie]

So what you're saying is that the BMS has a limit of 60V input?

Given a certain amount of sunlight, solar panels can give a certain amount of Amps, regardless of 'clamped' voltage I think... Connecting a 12V battery to a 20W panel with 20V open voltage will give 1,2A @ 12V for example; if we add a 6V battery in series with the 12V total of 18V, the panel will still give 1,2A, but now @ 18V...

That's why MPPT solar regulators has come on the market (Maximum Power Point Tracking)... They use DC-DC converters to utilitize the amps available at the highest possible voltage.

I just asume that the BMS has 63v caps and am suggesting to keep the panels under that in every case situation and keep it safe.

You will see that your panel can out put up to 1.4a I hope in 20 years time its giving 1.2 amps, even a linear voltage regulator should take no current away from your charging and you could set a voltage at least..

The zener BMS should do it fine and would just make sure there are no accidents..

Your right about the panels outputing only one amount of current through a resistive load power source. But when you put it through an inductor it become an inductive power source and both volts and amps are interchangable.

On a full discharged pack the volts are low and so are the panels.  Like you say they can only output so many amps however you will see the currnet consistantly until the pack is full charged.


I think there is more current to be drained from everything above the 17v nominal voltage rating and above the lowest charge limits as long as you place the load thriough an inductor,

[Leslie]

So what you're saying is that the BMS has a limit of 60V input?
 Connecting a 12V battery to a 20W panel with 20V open voltage will give 1,2A @ 12V for example; if we add a 6V battery in series with the 12V total of 18V, the panel will still give 1,2A, but now @ 18V...

The 20 watt panel on the six volt battery will charge no faster at 18v @ 1.2a as it would on a 7.5v panel @ 1.2a.  

On a 6v battery

6.5v X 1.4 amps

Your 20 watt panles have a total of 9.1 watts.  

So sad and watt a waste.  

You can convert this excess voltage with a switcher through an inductor and diode and put 1.4 amps into the switcher and get 2 amps out of the inductor at 6v.  

This area should be exploited and turn the ,

[Leslie]

I hooked up one panel in place of a single series SLA battery, (dont do this with all panels as you may fry your controller). to a brushed motor and controlller with LVC. And the bike pedaled and went along very slow jogging pace.

Which leads me to thinj you could do an inductor switcher through a brushed controller set at all the right volts.  Buck or boost and 50ma running cost with PWM to boot. LVC shut off programming.  Wow. what an idea.

My next experiment is to use brushed controller as switcher. I needs to set a referance voltage over the throttle.  

[Cornelius]

So what you're saying is that the BMS has a limit of 60V input?
 Connecting a 12V battery to a 20W panel with 20V open voltage will give 1,2A @ 12V for example; if we add a 6V battery in series with the 12V total of 18V, the panel will still give 1,2A, but now @ 18V...

The 20 watt panel on the six volt battery will charge no faster at 18v @ 1.2a as it would on a 7.5v panel @ 1.2a. 

On a 6v battery

6.5v X 1.4 amps

Your 20 watt panles have a total of 9.1 watts. 

If you clean your skull-glasses, you'll se that I said a 6V battery IN SERIES with the 12V. My point was that then we would exploit 1.2A x 18V = 21,6W instead of 1.2A x 12V = 14.4W out of the panel. This was just an example on how much watt varies with the voltage load... Oh, never mind; we're actually saying the same thing... (Besides, we'd need 21.6V to really top up that 18V battery, but assuming a flat battery... )

I'll be on the lookout for some 32cells panels; that would suit me better. But it might be hard to find here in Norway; i've only seen 36cells panels here...

[Leslie]

So what you're saying is that the BMS has a limit of 60V input?
 Connecting a 12V battery to a 20W panel with 20V open voltage will give 1,2A @ 12V for example; if we add a 6V battery in series with the 12V total of 18V, the panel will still give 1,2A, but now @ 18V...

The 20 watt panel on the six volt battery will charge no faster at 18v @ 1.2a as it would on a 7.5v panel @ 1.2a.  

On a 6v battery

6.5v X 1.4 amps

Your 20 watt panles have a total of 9.1 watts.  

If you clean your skull-glasses, you'll se that I said a 6V battery IN SERIES with the 12V. My point was that then we would exploit 1.2A x 18V = 21,6W instead of 1.2A x 12V = 14.4W out of the panel. This was just an example on how much watt varies with the voltage load... Oh, never mind; we're actually saying the same thing... (Besides, we'd need 21.6V to really top up that 18V battery, but assuming a flat battery... )

I'll be on the lookout for some 32cells panels; that would suit me better. But it might be hard to find here in Norway; i've only seen 36cells panels here...

Yes sorry I miss read your post, alot goes on here some days. Yes its the usefull energy I am trying to point out to you. If you could keep your batts topped up it would be better all round.

Or better still the energy can be used to charge a jump battery.  Another way to convert these volts into current useable power is to use a set of cells closer to nominal voltage then allow it to trickle into a pack of a lower voltage source.

Those panels could charge some smaller 4ah cells really fast and when fully charged dump there capacity into your main pack with as many amps as the BMS allows. But I would try keep it to as close to your panel output.

[Leslie]

I did some tests on a 36v SLA pack and 4 series X 12v 20watt panel system and it turns out although the current output was equal to the same pack configuration on all tests the voltage increase over the terminals was over .5v with 4 series panels higher, than on 3X12v series panels.


How much of that energy is being taken into by the batteries, I don’t know and the effect of having a higher than rated load on panels & batteries, I don’t know either.

I was under the impression that solar panels could take a heavy load indefinitely.


Some more tests and research over the week will reveal some interesting results.

I used to charge my 48v pack over a 52hz 73v voltage doubler off a big tranformer.  This worked very well but in this situation the battery resistance induced the transformer to produce more current.

[Leslie]

My bad.  

It causes a different balance in between the 12v batteries.  The series pack actually dropped  from 3.84v to 83.2 when using four panels but the batteries are different types and ages and I saw a wider difference between each battery. This happening could be exploited to our advantage in a BMS setup.

And the current with 4X12v panels over time dropped by about 50ma. Showing there might be heat generated in the solar cells.

Back on to 3 X panels to a 36v pack the current picked up to 1.22 amps and through an LM338 regulator limited at 60v 1.22 amps.  No difference and no heat in the package as there needs to be no regulation yet..

I am using this LM338 regulator and set it at any voltage I want.

I set my lLM338 at 60v for the 60v max 3 amp limit switcher I will build.  . I may be able to put a barrier diode, cap and inductor before the  input of the switcher to see if I can induct the voltage above the LM338 on an off switch. I am going to try scavenge for any useful power for my output I can draw 20v @ 1.2 amps top the top of my switcher which could equate to 300 ma current

[Cornelius]

The difference you saw in voltage are probably due to both different condition and types of batteries in your pack, and amp output drop in the solar panels happens when they are hot. I got a 50W panel here which i've seen around 53W into a 12V battery during wintertime with -10deg. C. outside, while the same panel never exceeds 45W during summertime into the same battery...

[Leslie]

Yes but the luminary at summer extends and if the panel is cooler in mornings in summer or afternoon, maybe face the panel where the spots are sweet.

On the bike in motion and the panel lie flat, I discovered I will be lucky to get 1 amp during mid day and about 800ma in before lunch or afternoon, Heat wont be the issue here.

I rode the SLA's near flat last night, the bike is an old Currie Tech Dyno KMX brushed POS with 36v SLA runs about 250 watts max, I rode about 26kms, not much hill, no pedals, and heavy handed throttle.  And I did nothing but use 3 series panels to charge my 36v 2X12ah and older 24ah batteries.

In the morning I set up my prototype and tried a few things.  I did manage to get a full charge today.

With a number of interruptions because of tests, by the afternoon about 4.30pm, none of the SLA's could of went over 13.6v and the trees were eclipsing the panels faster than it was worth moving the rig.  

So I pulled the bike out of the its spot and put each SLA on 1 amp wall chargers, two 1 amp chargers to both 12ah batts went green with within a few seconds of being connected, pretty good for a 1 amp input current, and the 24ah SLA took its time, but was reading 14.6 when I decided to take my brushed bike for a spin. With two 1 amp 12v wall chargers in parallel (2 amps total) the 24ah battery was indicating full charge. Despite not making it to 15v per SLA all batteries had acquired adequate capacity to indicate full charge to the wall chargers.  So what may look as no hope when looking at a multimeter in essence the panels actually works quite well.

So today I rode around from my solar panels.  The bike felt as it normally does, but it felt better knowing the energy came from the sun.

[Leslie]

Here are my regulators.  All surface mounted,   the switcher is waiting for a higher volt diode than the one I have.



One is an LM338 and the other is a simple switcher LM2576HV.

If I can track down a fast diode that can do +45v I will see if I can make my 4 X solar through a hand would inductor give up some better current to my 36v pack.