Home grown BMS ideas !

How do you store and manage your electricity?
Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Sun, 18 Jan 2009, 21:34

I found an article in EE Times-India that describes an active charge balancing system for Li-ion cells that looks very good. I can't put in a link as you have to log on to read it, but I have it as a .pdf file if anyone wants the full text. It describes "bottom balancing" where a cell lower than the average is given extra charge and "top balancing" where a cell that is at the highest voltage is discharged, the energy going back into all the other cells. The system also allows a voltage scan to monitor all the cell voltages and the way it does this is pretty clever.
It uses a simple multi-winding transformer (20 kHz ferrite). There is a single primary plus a secondary for each cell ie 12 cells need 13 windings. A transistor switch is required for each winding. A microcontroller manages all the switches.
That's all you need for a block of 12 cells, parts cost under $20.
Their prototype is claimed to achieve balancing currents of up to 5 amps with a dissipation in the complete block of only 2 watts.
It would be pretty easy to provide opto-coupled comms to a "master" so you could see every cell voltage in a pack of several hundred! But not absolutely necessary.
I have started work on a prototype for a 4 cell block, just to try it out.
Anyone would like a copy of the .pdf file, send me a PM.

User avatar
coulomb
Site Admin
Posts: 3611
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by coulomb » Thu, 22 Jan 2009, 20:43

> The best thing about building something and posting it is that others will know what not to do !
> ... or at least have fun flaming it !

Well, perhaps not a flame. Your opto is normally on, which I think is a good idea, so if the cell dies to zero volts, its "I'm OK" signal (conducting) won't be there. So I assume that you will put the outputs in series, so if any are open, the series string will be open.

Will you be putting all 220 outputs in series? If so, that would be a 66v "low" if the Vcesat of the optos is 0.3v; 44v if it's 0.2v.

Perhaps a group of them will be put in series, and combined with a separate board to generate a logic signal?

BTW, thanks a heap for posting this. It's a nice, minimal design with all the essential features. My only criticism is the PNP transistor in series with the programmable zener across the cell with no resistor or fuse. If both fail shorted, as semiconductors often do (when they fail), there will be a great current. Presumably, the "zener" or PCB tracks will vapourise before too much damage is done. I think it would have been nice to put a small resistor in series with the PNP transistor's base.

Will the PCB layout be available? A friend and I are thinking of a high voltage AC system, and our current dreams for a BMS are unlikely to be ready any time soon.

- Coulomb
Last edited by coulomb on Thu, 22 Jan 2009, 09:48, edited 1 time in total.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Home grown BMS ideas !

Post by Johny » Thu, 22 Jan 2009, 21:01

Hi coulomb. He describes it's intended operation in "hyjacked topic" - which is linked in the first post in this thread.

User avatar
woody
Senior Member
Posts: 1711
Joined: Sat, 21 Jun 2008, 02:03
Real Name: Anthony Wood
Location: Mt Colah

Home grown BMS ideas !

Post by woody » Thu, 22 Jan 2009, 21:06

acmotor wrote: All important.

0 Safe... excludes bipolar /fet junction directly across cell ! and resistor whose power rating is exceeded by full on pwm from micro.
Should the zener / resister / zener balancers include a fuse?

Should the LED-darlington include a fuse too?

How should the fuses + resisters be sized?

cheers,
Woody
Planned EV: '63 Cortina using AC and LiFePO4 Battery Pack

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Home grown BMS ideas !

Post by Johny » Thu, 22 Jan 2009, 21:48

It would be far better to have an intrinsically safe design rather than a fuse. A blown fuse would be a right pain. Zener resister balancers can be rated correctly with only a loose connection causing damage. LED/Darlington system should include resisters (IMHO).

User avatar
coulomb
Site Admin
Posts: 3611
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by coulomb » Thu, 22 Jan 2009, 23:08

Zeva wrote: I'm certainly interested to hear what others may have done or are planning to do for their BMS too, in case I may want to steal their ideas.
Woody wrote: I saw a4x4kiwi contactors which split his pack into 12 x 48V packs and a 24V pack for charging. It made sense for me to join them all in parallel to 1 48V pack for use with a big charger.

The way to do this was DPDT contactors:


    48V Bus + ---O O--- + 48v block - ---O O--- - 48V Bus

                   /                        \

prev 48V block ---O O-----(bypass)-------O O--- - next 48 V block
OK, I have this crazy idea, based on this simple circuit per cell or per group of cells:

Image

It's a simple bridge circuit, capable of transmitting DC power from "input" (the cell(s)) to "output" (part of the input to the motor controller), or vice versa. The voltage ratio is set by the duty cycle of the transistors, and the output can be more or less than the input. So this is a buck/boost converter.

My first conversion is likely to be a high voltage AC system, but I'd like to avoid having to wire 180-220 BMS systems in series, if I can avoid it. (How are those 220 boards coming along, acmotor? Image). So the idea is to use a smaller number of higher capacity LiFePO4 (LFP) cells, and use the DC-DC converter to translate say 3.2v nominal cells to say 6.4v nominal pseudo-cells. This could also operate on groups of cells, say 16 cells at about 50v nominal.

As per a standard buck/boost converter, Q1 is on for say 2/3 of the time, "charging" the inductor with current flowing left to right (conventional current). Q1 is opened, then Q2 closed; in the dead zone while they are both open, the capacitor continues to supply current to the motor controller, preventing 600-700v from messing things up. Q2 conducts for 1/3 of the cycle, doubling the voltage from the cell(s) to the capacitor. 1:1 duty cycle would just copy the cell(s) voltage to the capacitor (neglecting losses).

Now here is the beauty: we can save the cell(s) during garage charging, motoring, or regenerating (regen). During motoring if the cell voltage gets too low, we just decrease the "turns ratio" of this DC-DC "transformer". So instead of doubling 2.5v to 5v, we multiply by say 1.8, to contribute 4.5v to the pack voltage. This also reduces the current from the cell(s) from 2x the pack current to 1.8x. The pack voltage goes down slightly, which will also reduce the load a tiny amount.

If the cell gets really bad, in the limit, we can just leave Q2 on and Q1 off, bypassing the cell(s) without using contactors.

During charging (garage or regen), if the cell voltage exceeds a preset limit (say 4v), we increase the turns ratio. Instead of doubling 4v to 8v, we multiply it by say 2.2, yielding 8.8v. The current ratio changes from 1/2 to 1/2.2, reducing the current into the cell(s). If this 4v limit is precise, it can be used as the sole balancing method for the cell. If this switching is done at the group level, another balancing method is used for cells within the group.

For garage charging, all that is needed is a source of DC on the DC bus of the motor controller (i.e. across the pack). If the controllers controlling the cells "transformers" are smart enough, they can cope with almost any reasonable voltage level, and use just what current they need by adjusting their "turns ratio". When done, they set the "turns ratio" high enough that the charger doesn't supply any current. The input could conceivably be the unfiltered mains (i.e. a bad boy), and as the waveform rises and falls at 100 cycles per second, the turns ratio could increase and decrease to approximate a sine wave of current, offering good (low) harmonic distortion (often misnamed as good power factor).

When used at the group level, this could be made to do sensible things with a dead cell (bypass the group). It handles over and under voltage, and could handle over temperature as well (at high temperature, just increase the turns ratio if the cell voltage is higher than nominal, and reduce if the cell voltage is lower than nominal).

So, here are the advantages:

1. Relative simplicity, although a lot of the smarts are not shown (driving the gates sensibly).
2. Handles over and under voltage, and overtemperature.
3. A failed cell or group of cells has minimal impact on the pack voltage, so the controller isn't surprised by a huge change in what it assumes is the mains voltage (assuming an industrial controller).
4. Charging is vastly simplified; the charger reduces almost to a rectifier.
5. There is no need for dissipative equalisation, although the FETs will get quite hot.
6. The transistors don't see much voltage in normal operation, so inexpensive 40v devices will suffice.
7. The parts count is quite low, despite needing a microcontroller, gate drivers, isolated supply for Q2's gate driver, etc. The capacitor may take up a bit of room; I haven't calculated its size yet.
8. The microcontroller driving the gates could also act as a monitoring system, sending cell voltages and status on request.

Here are the disadvantages:

1. The transistors handle full cell current, so they need to be capable of handling the current, and on resistance should be very low.
2. The inductor would act as a fuse in case Q1 is turned on hard or fails shorted. That's not ideal.
3. If the two transistors don't conduct for more than a few milliseconds, the capacitor will attempt to charge to full pack voltage, and the transistors would overvoltage.
4. The upper transistor requires high voltage for its gate drive, necessitating transformers or some such to derive a suitable isolated supply.
5. If the circuit isn't very, very efficient, this circuit will dissipate massive power. It may even have to be water cooled. This is the achilles heel of this concept.
6. The losses in all the switchers could add up and affect peak power and range.

No doubt there are other problems I haven't listed. But I think it is a neat concept, and could work out well in some situations, e.g. bicycle or even motor bike sized projects.

So: what does the group think about this idea?
Can it be made to work in a car-sized vehicle?
Are there other problems?
Are there other improvements that could be made?

- Coulomb
Last edited by coulomb on Thu, 22 Jan 2009, 12:10, edited 1 time in total.

User avatar
coulomb
Site Admin
Posts: 3611
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by coulomb » Thu, 22 Jan 2009, 23:43

I should have mentioned this additional advantage:

9. This system is capable of saving the pack in the event that the communications between the BMS and the motor controller (e.g. excess load or regen) is lost. It even saves the pack if the motor controller fails shorted (the groups or cells set their output voltage very low, resulting in low cell currents).

Naturally, I would expect the BMS to ask the controller to wind back the load or regen as needed, as in a standard BMS.

- Coulomb

User avatar
evric
Site Admin
Posts: 500
Joined: Sun, 20 Jul 2008, 01:57
Real Name: Eric
Location: Adelaide SA
Contact:

Home grown BMS ideas !

Post by evric » Thu, 22 Jan 2009, 23:53

Hi Coulomb,
Voltage doubler circuits are inherently lossy. They will only supply very small currents. ...Ric
Prius Plug-in Conversion: http://www.evplus.com.au ...Holden Barina EV: http://www.evric.kestar.com.au

User avatar
coulomb
Site Admin
Posts: 3611
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by coulomb » Fri, 23 Jan 2009, 02:45

evric wrote: Voltage doubler circuits are inherently lossy. They will only supply very small currents. ...Ric
Yes, that seems to be the received wisdom.

Where are the main losses? Ferrite losses in the inductor? FETs are pretty efficient, and the copper losses are pretty easy to keep low.

Perhaps it's junction capacitance (switching) losses?

- Coulomb

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Fri, 23 Jan 2009, 04:43

Coulomb,

Great to see you are on the ball re my simple BMS circuit.

Yes, I had fitted a 200 ohm resistor to the transistor base at the last minute before the PCB was made (sorry, wasn't on posted cct)( I must have known you would spot that shortfall !).... and a second LED on the PCB to parallel the opto so you can see what is going on.

No fuses, just failsafe design would be my preference. What is the point of a fuse ? extra cost and something else to go wrong.

I see the biggest problem in a series battery pack (particularly higher voltage one) is how to survive an open circuit cell. Danfoss controller will shut down on undervolage but will the offending cell still be there ? A fail to s/c is easier to handle.

Re heat loss / shunting. This is actually needed when you run regen. There is nowhere else to 'shuffle' the power to when cells are full.

Correct on the Vce(on) of the opto and of course its Vce(oc) max.
My battery pack is divided into modules of 22 cells by contactors(70.4V) at present (although I may eventually use half that) so the opto Vce total is less than 10V. This also means that the module can be self contained and pass its status on via another opto to join the other modules, thus maintaining isolated wiring between modules.


Re your BMS thoughts. Interesting.
I guess that the lower the cell count, the fancier the BMS can be. However, after starting with a home made micro BMS I found that the cost per cell and complexity per cell was an issue at 220 off.
Shuffler circuits seem to be around 65% efficient so some energy is saved.

It was also possible to crash the micro on my prototype with the massive electromagnetic fields in the battery wiring when hundreds of amps were flowing. Believe me, I tried it. Sure I could have made it layout less sensitive but it reminded me that simple analogue was fine for bottom line cell protection... use the micro for other nice to know but non critical monitoring.

I am still assembling and have a few layout points I would change on a next run of boards. I will be gettting more boards made once my pack is complete. I will post up results and maybe do another PCB run if you are interested.

Edit:spelling, before Richo sees it Image
Last edited by acmotor on Thu, 22 Jan 2009, 18:58, edited 1 time in total.
iMiEV MY12     105,131km in pure Electric and loving it !

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Home grown BMS ideas !

Post by Johny » Fri, 23 Jan 2009, 05:50

On acmotors BMS. How about cascading the optos so that each opto output goes in series with R4 on the next module. No VCC issues and one opto for the output which can be the closest one to signal ground.

You can also lower the opto drive current because the output is running low current. It might stuff up the 2.5V set point though.

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Fri, 23 Jan 2009, 06:08

Just lazy.
One wire daisy chain x 220 is less than 2 wires for cascade. But yes, it would throw out the set point.

The idea is to have the cells as autonomous as possible, even when they are stand alone. edit: did I say that ?Image You know what I mean though.

I did start on OV and UV separate via 2 optos but cost goes up as does wiring so came back to (Probably Rod Dilkes's) idea of just battery good on one line. After all, you should know if you are charging or discharging !
Last edited by acmotor on Thu, 22 Jan 2009, 19:52, edited 1 time in total.
iMiEV MY12     105,131km in pure Electric and loving it !

User avatar
Richo
Senior Member
Posts: 3556
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

Home grown BMS ideas !

Post by Richo » Fri, 23 Jan 2009, 08:44

As a side point after seeing thousands of different product boards in for service the tracks on PCB's make quite a good fuse Image
esp when you are talking about a battery that can supply enough current to vaporise most tracks 300mA v 300A. Image
Help prevent road rage - get outta my way! Blasphemy is a swear word. Magnetic North is a south Pole.

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Fri, 23 Jan 2009, 20:17

Agreed on track fuses !

They only get into trouble when the voltages are higher as they tend to take the whole PCB, instrument and room with them !

At only a few volts you can get away with them though.
iMiEV MY12     105,131km in pure Electric and loving it !

EVLearner
Groupie
Posts: 51
Joined: Tue, 06 Jan 2009, 02:57
Real Name: Malcolm Moore
Location: Turramurra

Home grown BMS ideas !

Post by EVLearner » Sat, 24 Jan 2009, 03:07

From my limited and dated experience, Zener Diodes tend to have a near zero temperature coefficient at about 6.8 V - hence the desirability to aim for there, but Zeners at about 12 V have a far more stable voltage - almost irrespective of current flow (and their temperure coefficient is not all that bad - but if you check the reference voltage in IC chips, then these can be lower voltage along with temperature stability built in - hence the programmable voltage zeners give excellent results as you have pointed out.

After a few weeks thought on Battery Voltage monitoring, I have a mental picture of using firmware approach using a 1 of 16 multiplexer being fed from say 12 10 M ohm resistors (one from from each incremental voltage), and 12 protective diodes to prevent the Multiplexer from being back-biassed, a couple of op-amps (in the Analogue to Digital Converter (ADC)?) to invert and normalise the current referred voltage and then a 16 bit ADC (with its internal voltage reference) to get the digital voltage equivalent, before feeding this into a microprocessor and making a '12 wide, normalised Bar chart display. Comments appreciated...

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Sat, 24 Jan 2009, 04:15

Just musing here...

I have made a 4 cell ladder voltage measurement to one ADC and subtracted in software to get individual cells voltage to experiment with BMS.

This is fine however voltage numbers degrade as the number of cells increases. Maybe 8 might be a limit. The reason being that there is not much to see on the TS cells I tested once they are equalised. Variations in voltage only occured in the mV range.

At some point I decided I just wanted to know if there was a faulty cell as 220 sets of voltage readings were not very interesting.
I have gone for just reading the overall voltage on modules of 22 or so cells.

Such info as ESR for each cell could be interesting, however the pack is still just limited by the weakest cell. Since you are already measuring energy or Ah in/out of the pack, the individual cell info is fairly dull unless there is something going wrong.

In a pack of 45 TS cells it may be worth reading every cell.
iMiEV MY12     105,131km in pure Electric and loving it !

User avatar
Mesuge
Groupie
Posts: 168
Joined: Wed, 05 Mar 2008, 07:37
Contact:

Home grown BMS ideas !

Post by Mesuge » Sat, 24 Jan 2009, 05:41

Tuarn, I think you are musing in the right direction.
Just look at the AC bigboys, for instance ACP/eBOX frontend seems to indicate they data log only subpacks performance and possibly the highest/lowest VDC cell as an outlier, which should be enough to detect/service any problem with the cells/entire batt. pack. On the other hand detailed BMS and its importance for low voltage system is another story.
ImageCzech EV Club | blog, gallery, video | http://old.elektromobily.org

EVLearner
Groupie
Posts: 51
Joined: Tue, 06 Jan 2009, 02:57
Real Name: Malcolm Moore
Location: Turramurra

Home grown BMS ideas !

Post by EVLearner » Sat, 24 Jan 2009, 06:12

Yes ACMotor, I was thinking about monitoring every cell and then it came to me that if a cell in a battery was faulty, then this would show up in the battery in any case, hence the relax back to 12 voltage test points and not 72 test points - I think we would get the same answer (or information anyway).

The output could be a set of vertical LED arrays forming a horizontal line that shows charge as the height of the line and any leds that are out of line show that the relative battery has a cell that has a problem.

Not that I probably missed it before - how hot do these batteries get, and should this be monitored too??

It almost makes me think that the battery manufacturers could easily install a simple monitoring module in the battery wall that can be plugged into - or have they already done that - or do we have to invent their wheels for them too??

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Sat, 24 Jan 2009, 08:21

I am glad there is support for battery module rather than cell voltage monitoring, I don't feel so bad now ! (still keeping the cell OV/UV and shunt for eq.)

Built in battery monitor... This has been done by several manufacturers but not on any batteries I can afford ! I think Altair-nano were first some years ago.

On the Rodeo my plan is to discharge at average 0.5C (peak 3C) and charge at 0.1C so I am not expecting temperature to be an issue.
220 x 40Ah TS cells = 704V at 0.5C =14kW average power

Heat at cell (worst case ?) = ESR x (Amps)^2 so on average this would be .004 x (20)^2 = 1.6W        and .004 x (120)^2 = 57.6W for peak.

I can't see this heating things up much !

Now if you pull 5C the game starts to change...
.004 x (200)^2 = 160W but even then it would be in short bursts (and not in my EV).

The TS cell's operating temperature is quite wide (-25 to +75°C) and Matt always tells me they should be 'toasty' for best performance.
iMiEV MY12     105,131km in pure Electric and loving it !

EVLearner
Groupie
Posts: 51
Joined: Tue, 06 Jan 2009, 02:57
Real Name: Malcolm Moore
Location: Turramurra

Home grown BMS ideas !

Post by EVLearner » Sat, 24 Jan 2009, 16:04

I am cool with the internal resistance power dissiapation, but what concerns me is the heat generated by the chemical reaction to charge / discharge the cells - and I am thinking that this reaction may generate far more temperature than the internal resistance alone.

On the conceptual side of it; if there is a BMS sitting over the batteries all the time than this will be a constant drain on the batteries - unless the BMS circuitry is made indivual for each battery and in can be remotely activated (else is open circuit) - apart from say a 10 M ohm voltage monitoring probe.

On yet another angle, has anybody done a Finite Element Fourier analysis of the heat distribution of the plates (based on current flow and current density & chemical reaction), so that the temperature can be read by embedding something in the lead plates (or the acid/gel)??

acmotor: If your've got 220 (40 AH) batteries that is a lot of volume (even if they are Lithium) - like about 2 Metres^3 - where do you sit??

User avatar
woody
Senior Member
Posts: 1711
Joined: Sat, 21 Jun 2008, 02:03
Real Name: Anthony Wood
Location: Mt Colah

Home grown BMS ideas !

Post by woody » Sat, 24 Jan 2009, 16:55

EVLearner wrote: acmotor: If your've got 220 (40 AH) batteries that is a lot of volume (even if they are Lithium) - like about 2 Metres^3 - where do you sit??
Er, LFP40's are just over 1 litre each (+ 1.5 kg), so 220 should be 223 Litres (330kg) - quite manageable :-)

cheers,
Woody
Planned EV: '63 Cortina using AC and LiFePO4 Battery Pack

EVLearner
Groupie
Posts: 51
Joined: Tue, 06 Jan 2009, 02:57
Real Name: Malcolm Moore
Location: Turramurra

Home grown BMS ideas !

Post by EVLearner » Sat, 24 Jan 2009, 17:42

My bad grammar - I was assuming Lead Acid, not LFP40's! Hey, you don't have to cast out the back seat after all!

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Sat, 24 Jan 2009, 19:21

Yep, TS on the tray back of the Rodeo. Image

FEA on cell plates ? Someone somewhere is probably doing that. Maybe CSIRO with the ultra battery (lead sinker unless they make some real breakthrough though, but it is a good idea. Lead working as well as Lithium would be interesting)

Are you offering to head up battery research in Oz ? That would be great ! It is way beyond me. I have enough trouble just charging and disharging batteries ! Image

Your point is good re heat from chemical reaction.
My thinking is that any heat is energy used in the charge/discharge equation. So if the ESR accounts for the cell losses then the heat(energy) from chemical reaction is already accounted for in the ESR number ? Someone will flame me here.

Maybe we need a careful test of a thermally insulated TS cell in charge / discharge unless someone has the data already ?    


EL,
Correct, minimal battery drain by BMS is part of the design.
My BMS battery drain including LED and opto is <4mA.

40Ah / .004 = 10,000hours to flatten, that is well over a year and probably only just more than self discharge.    
iMiEV MY12     105,131km in pure Electric and loving it !

EVLearner
Groupie
Posts: 51
Joined: Tue, 06 Jan 2009, 02:57
Real Name: Malcolm Moore
Location: Turramurra

Home grown BMS ideas !

Post by EVLearner » Sat, 24 Jan 2009, 21:20

AC,
Regards to the UltraBattery have a look at Firefly.com This is a USA based battery company using a virtually parallel technology to the UltraBattery; ETA about 18 months to the mere mortals.

It looks to me that the Firefly "Carbon Sponge Lead-Acid" technology has a Power/Volume figure that also parallels the Lithium cells (at a large fraction of the cost), and is very low internal Resistance, and is very robust.

My gut feeling about batteery heat is that the chemical reaction is one thing and the I2R losses are another kettle of fish, but these heat sources need to be added.

A BMS runing at <4 mA is commendable (is this for a 12 V battery/pack). Is the LED being run from a little power converter? Does this mean thatyou are simply measuring the voltage for each battery pack or does your BMS do much more - sorry I came into this (EV) life very recently!!



EVLearner
Groupie
Posts: 51
Joined: Tue, 06 Jan 2009, 02:57
Real Name: Malcolm Moore
Location: Turramurra

Home grown BMS ideas !

Post by EVLearner » Sat, 24 Jan 2009, 21:34

AC,

Yes I finally caught on from going to the start of this thread and saw your little circuit. So the Opto is off for under and over voltage and on if the voltage range is OK - all makes sense there! (Based on nominally 4 V)

Post Reply