Zener Shunt Battery Management System for SLA batteries.

[Leslie]

There has been a few posts about members using SLA batteries here.

I made a few of these to go over each terminal


Things you need.

Updated:

2X12v in parallel 5 Watt Zener Diodes  

2X4 ohm 1 watt resistor in parallel.  2 ohm total.

2 rectifier diodes in series. .8 drop 1 amp.

1 40v schottkey diode .5 drop. 1 amp.

I will do up a circuit picture but this text schematic will do to launch this thread.


Ground    Resistor   12v zener        Rectifier diodes
| |--------WW------>Z---------|S---|<---|<-------+


How this works is that the rectifier diodes drop the SLA terminal voltage by .8v each and the schottkey by .5 , 2.1v total.  

While charging, when the terminal voltage reaches 14.1v the voltage past the rectifier diodes reaches 12v and the zener diodes begin to conduct current.  
The 2 ohm resistor will pad the current to 450ma so as the zeners are not overloaded,  12v 5 watt zeners can do 395ma regulation current so two in parallel should be fine.  

At float voltages or resting voltages the zener blocks the regulating current so it doesn't discharge the pack.  It only works when your 12v SLA is reaching 14.1v

This shunt when the SLA is at 14.1v kicks in and starts to regulate the amount of current that shunts through the zeners and resistors.  So if one series SLA is at 14 volts, and another is at 14.8v, .7 volts is extended past the diodes to the resistor regulating 350ma of charge to the resistor slowing the high voltage battery charging.

The lower SLA 14v gets full charge current and get a chance to catch up to the high charged SLA's  When this has occurred the current is limited to each SLA by 450ma.

As the voltages change on each series SLA, so to does the regulation, the more volts that gets to the 2 ohm resistor the more regulation the BMS string gives to watts.

with a 2.2 amps charger this will allow 2,2 amps normally and 1.75 amps to the whole pack when all BMS strings are fully operational.  This is fine for 12ah batteries.  With smaller amp chargers you may need to increase the resistor to 4 ohm so you are not under currenting the batteries with full charge volts..

You need at least 1 amp for 10 ah SLA.  Any less amps is not to good for charging SLA's in bulk mode.

The resistors will draw 450ma at .9v so .9*.45= 0.405 watts so a single 2 ohm 1 watt resistor should be fine but I use two 4 ohms 1 watt in parallel and 3 amp rectifier/schottkey diodes for lead strength.

Regulation current.

14.1v~0 amps
14.2v~.05 amps


14.5v ~.2 amps

14.8v~ .35 amps

15v~ .45 amps.

As you can see as the voltage gets higher on any SLA the shunt current increases making the lower charged SLA receive more current than the higher charged SLA that is being shunted as opposed to, not as shunted or not at all shunted..  

[Leslie]

Here is a pretty picture for the SLA die hards.

This when used will have your SLA bike jumping out of the box and you should get better balanced charging and more range when you have better balanced SLA's.

[Magneto81]

I was looking at this yesterday as a result of looking for a BMS system... I think it was a link from Kamilion:
http://a4x4kiwi.blogspot.com/2008/06/battery-balancing-cheap-way.html

I was hoping to keep it as simple as possible - is it necessary to have all the parallelism of parts? In the above example, they used 2 zeners and 1 resister per battery. I'm also kind of scared of the whole setup - it looks like they are short circuiting all the batteries, but I suppose that's what the diodes are for?... Sorry, I'm pretty 'newb' when it comes to electronics still, but I can follow instructions

I definately like your setup at 14.1 volts rather than 13.6... I think some of my batteries sit at 14 or less just after a charge. (Usually 13.35 or so, but of course all this depends on temperature.)

[Leslie]

I was looking at this yesterday as a result of looking for a BMS system... I think it was a link from Kamilion:
http://a4x4kiwi.blogspot.com/2008/06/battery-balancing-cheap-way.html

I was hoping to keep it as simple as possible - is it necessary to have all the parallelism of parts? In the above example, they used 2 zeners and 1 resister per battery. I'm also kind of scared of the whole setup - it looks like they are short circuiting all the batteries, but I suppose that's what the diodes are for?... Sorry, I'm pretty 'newb' when it comes to electronics still, but I can follow instructions

I definately like your setup at 14.1 volts rather than 13.6... I think some of my batteries sit at 14 or less just after a charge. (Usually 13.35 or so, but of course all this depends on temperature.)

Yea I think those 1w zeners wont pass 347ma reversed for too long.

Zener

15.2v-13.6v =1.6v

1.6v/4.6 ohm= 347ma

1N4735A 6.8v zener constant current ^IZM (current zener max) is 146ma.  Ooops.  someone didnt do their math.    Thats surviving on the surge current rating.


Resistors

1.6v*347ma=.55 watts.  

LOL

Hell why the massive big 5 watt resistors and the skinny ass zeners?  Its backward designed to fail and melt SLA's.  Nice web design though.


You need all shunt zeners to operate perfectly and one can not fail or else you will end up in deep doo.

Another reason why I use 2x5 watt zeners and two one watt resistors is that the zeners get freakin hot so using two can space out the heat and the leads are so much better than the 1 watters.

The 5 watt resistors are fine a little bit overkill but they are not near as strong and tough as two one watters. A single one watter can get hot too.  Best to allow more air to the watts.

[Leslie]

And

13.6 volts is cutting very close to the float mode voltages. This is cool in some ways.  It slower charging but can make balance of the batts in float mode.

On a 36v pack

13.6v would agree with this voltage but if one fails there can be too much voltage to the failing bms string and things can get nasty.  Like two batts hanging at 14.2v and the failed BMS sitting on 16.6v boiling away.

on a 48v pack you need it fail proof regardless.  On a 144v car SLA beast with 12 12v batts in series, if one fails, you have smoke and flames.  

I do try to balance as close to finish of charge cycle.  If one fails at 14.1v you may get close to two batts at 14.6v and one at 15.8v is a little better.  Ball park guesses.

Over the years Ive used the zenners they never failed elecrically just physically and this is why I stress the types of components used in the diagram I have provided.

I used 14.4 volts on some good branded SLA's with a 1 ohm resistor and used 3 rectifiers, and that was fine.  My charger was set at 14.8v per SLA on a 48v pack and if one failed I still got only about 16v on the failing BMS stirng when the charge would cut.

If you want to use a lower volt zenner shunt I will work on one  tomorrow and post it here.  

[Leslie]

I can get the 12v 5 watt zeners at jaycar.  They are a local electronics store here in Australia.  If you can get 13v 5 watt zeners they would be nice.


The 1N5350B 13v 5 watt zener is good for 365ma. 3 ohms will draw 400ma through the shunt.  Id still use one 3 amp rectifier why?  It only happens once and it will piss you off.  One can fiddle around getting them right.  The rectifiers work very well as reversed polarity protection.  Some times the little white line on the zeners gets worn off or is hard to see.  You make up these little BMS strings with a rectifier. Electrically, They will "always" work.


New parts list for a 13.8v SLA BMS.

2x 13v zeners 1N5350B.

1 rectifier diode 3 amp.

2x six ohm resistors.

Or

2x 12v zeners 1N5349B parallel

1 rectifier diode

2x schottkey diodes

2x six ohm resistors

or

The best config I can come up with is this one at 13.7 v

3 x 4.3v zener  series 1N5336B 1010ma ^IZM (12.9v)

1x 3 amp rectifier diode.

2x 6 ohm resistors in parallel (1 watt)

The lower volt zeners can provide more regulating current than the higher volt zeners.  Still the issue with lead strength, zeners are a little on the malleable side I still like to parallel them..

If you cants get the 5 watt zeners.  No reason you cant use 4x 1 watters.  

Holy moly the 3.3v 1 watt zeners are good for 267ma these can make easy LiFePo4 bms strings.  2x 3.3v 1 watt zeners parallel  1x 1 ohm 1 watt resistors.  Good for 300ma, around 13.6% regulation for a 2.2A.

Suggestion for LiFePo4 bms states you need at least 10% regulation current off your charger current.  Id say the same goes for SLA too.

[Leslie]

After all that.

You know what I use?.

3x 12v 1 amp chargers in series and connect with a 6 pin plug.  Only use 4 pins ATM.  2 pins for solar charging.

This isn't the best solution but its off the shelf.  

Three of these might do.  Depends on isolation.

Vector 2,4,6 Amp 12 Volt Car Battery Rapid Charger

http://cgi.ebay.com.au/Vector-2-4-6-Amp-12-Volt-Car-Battery-Rapid-Charger-/280536106850?cmd=ViewItem&pt=LH_DefaultDomain_2&hash=item4151414762

Not as cheap as zener regs and a Ecrazyman SLA charger.

[Bikemad]

Leslie, I was just looking at this diagram and something caught my eye. I remembered seeing something regarding connecting zener diodes in parallel, so I did a quick Google search and found this:

Q-1
I would like to connect several Zener diodes in parallel to achieve loss dispersion, but can this be done without problems?

A-1
Do not connect Zener diodes in parallel.
If several Zener diodes are connected in parallel and a voltage that exceeds the Zener voltage is applied, the Zener current becomes extremely unbalanced if there are even slight fluctuations in the Zener voltage, causing topical load concentrations In the case of combinations with large fluctuations in Zener voltage, sometimes only one of the Zener diodes switches on.

Just thought it might be worth mentioning.

Alan
  

[Leslie]

Leslie, I was just looking at this diagram and something caught my eye. I remembered seeing something regarding connecting zener diodes in parallel, so I did a quick Google search and found this:

Q-1
I would like to connect several Zener diodes in parallel to achieve loss dispersion, but can this be done without problems?

A-1
Do not connect Zener diodes in parallel.
If several Zener diodes are connected in parallel and a voltage that exceeds the Zener voltage is applied, the Zener current becomes extremely unbalanced if there are even slight fluctuations in the Zener voltage, causing topical load concentrations In the case of combinations with large fluctuations in Zener voltage, sometimes only one of the Zener diodes switches on.

Just thought it might be worth mentioning.

Alan
  

Nope.

Ive used 5 zeners in parallel in a similar application and it only makes the voltage drop more stable by clamping each other.  Let it be said that the zeners I chose have a peak current well in the capability of the current draw of the resistors and we talking 1.2 at the other end of them not 30v.


Switching, with high volts past the zener I would agree with what you are saying.  But this is a slug of a shunt and we are talking 1.2v.

When they first start conducting it maybe 20 minutes before we start putting stress on any individual zener.  Plenty of time to allow another in parallel  zener break down. 


Maybe why this has worked for me without ever failing.


Edit:
14.1v~0 amps

5 minutes later.

14.2v~.05 amps

No problem here 50ma is not an issue

[Leslie]

The answer to why it works fail proof is that zeners are not 100% strict to the voltage drop. Which is why there is an issue in some applications, gladly not with this circuit.


The more zeners conduct they drop just a little more voltage ..  So if one is a little slow to zener reverse its zener voltage is higher. The slew of the one that has reversed will be forgiving. As the battery voltage slowly rises this slew will allow voltages over the zener still blocking to rise enough to conduct.

There will be a couple of uv and ua between the output and the input of the zeners creating the clamp effect I observed.  The lower drop zener will clamp the higher drop zener which will always remain in reversed as it was the first to switch on.

Providing you allow enough head room with you choice of component to allow for the clamping to occur and you are not trying to reverse close to the max current of a single zener in the parallel circuit with PWM, its a win win situation.  Better regulation.

This has been tested through and through.  I could sleep at night knowing my charging batts were with in .05v of each other in the morning.