5 Battery Management - Porsche 356 Electric Conversion

[maddexter]

Great project and nice car but is there a video of the car running?

[maddexter]

Is there a video of the car running?

[lpgas1]

thankyou for the great videos! i am glad you treat them as batteries instead of just cells! a lead acid has 6 cells but because they are in one case we treat them as a unit. your doing the same with lithiums and i love the simplicity of it. i will be doing exactly what your doing in the 356 next time my lead acids die in my car! just one question mate...what sort of charger are you using? sorry i am slowy looking at your videos. you may have one on it already!

regards blair

[marionrickard]

We use a Brusa NLG-511 charger on the Speedster. It is quite programmable as to charge curves and voltages, which is the important part with these cells.

[resal3000]

I generally work with RC lithium batteries.

Each cells are 5Ah, and I am preparing a 100V 100Ah pack(26S20P). They are Turnigy cells from HobbyKing.

These cells are more sensitive to overcharging and discharging.. And its is more of a NECESSITy to monitor each cell voltages.

Right now I am working on designing a affordable active balancing system. It is going to be microcontroller based. Noise filtering Algorithms will also be put into action.

[resal3000]

hello Richard :) Nice video.Very Informative. But I must say, to be on the safe side, a non-microcontroller based LowVolt Cutout and a Chrg balancer is required on every cell.

From my experience, checking on the terminal gives absolutely no evidence of what is happening inside. At capacity difference of 10% with off-throw the batteries during charges and discharges.

I absolutely respect the fact that you mentioned we should buy extra batteries. And I realised that too from practical experience

[marionrickard]

Well, we've driven the car for a year without all that and it is working very nicely thank you. I know you guys think its needed, but I cannot match that with what I see and experience directly.

Jack Rickard

[7926645]

Do you charge to 80% capacity, or to 100%. If to 80%. How do you stop the charging at 80%.

[marionrickard]

Well neither precisely. Normally you charge to a voltage and the manufacturer defines the maximum or 100% state of charge as 4.25 volts. They do acknowledge extended cycle life if you charge to 4.10 volts. This is in a lab on a single cell.

In a string of 36 cells, we found variations in internal impedance and capacity could cause problems at those levels with some cells exceeding 4.25 volts dramatically, while others not. We charge to 3.65% and I would say that corresponds to 96% charge.

[marionrickard]

You charge using a CONSTANT CURRENT source and the charger monitors the pack voltage. As it charges this voltage will rise. When it reaches 130 volts, it switches to a CONSTANT VOLTAGE source and holds that voltage there using whatever current is necessary. That current level will then gradually fall while holding 130v. As it decreases to below 2 amps, the charger shuts off.

[7926645]

Very interesting, thank you.

[erroneus]

Something that I have built, but haven't implemented yet is a low voltage indicator. This is more useful with lithium batteries, (i'm using lead acid). but it simply closes an isolated switch (optocoupler) when the voltage drops below a certain threshold. I am going to enable an LED to turn on at that point to show that there is one battery that is lower than the rest, thus catching a failure before anything gets destroyed. this analogue circuit is also immune to noise.

[marionrickard]

That sounds sufficiently simple to be useful. Again, I don't necessarily need to know WHICH cell is too low, I can find that back in the garage easily enough. But if you had a single LED that lit somewhat brightly if ANY individual cell or battery went below a certain threshold, the driver would know to go home and have it looked at.

Jack

[marionrickard]

Not a bad idea all in all.

[cant7think7clearly]

That's good advice on ordering an extra few cells for spares when buying your battery pack in case of future battery chemistry changes. I wouldn't have even thought about it.

Love your porsche conversion too mate, it's stunning!

[marionrickard]

We first rolled in December 2008 and are currently right at 5000 km. We have had NO cells go bad in the car, nor have any had to be replaced.

I have tortured a number of cells to death in the lab, both by overcharging, and by undercharging.

Jack Rickard

[JRP3]

By "undercharging" I assume you mean over discharging, not charging to a low level?

[marionrickard]

I stand corrected. You're quite right. Over DIScharging cells below the 2.5 volts and even to zero volts using a resistive load.

[marionrickard]

But here is a more advanced version of the same failure, and more realistic to the way it really happens in the car.

Take a four cell "battery" and charge 3 of the cells to maximum capacity. Charge one to 70% say.

Then put the BATTERY on a resistive load and discharge the whole thing. It is essentially "unbalanced". by undercharging one of the cells.

As it discharges, the low cell hits the steep discharge curve first, and as the other cells are still producing (through the low

[marionrickard]

cell. The low cell is destroyed. And the pack is basically disabled. But there are no fires. No great heat. And I can't detect any particular damage to the adjoining cells.

But you are quite correct in that undercharging it didn't cause the problem, it was the over DIScharge.

[JRP3]

How long have you been driving on these cells and how many, if any, have gone bad or been so out of spec you had to replace them?

[randycarter2001]

Where is the details of the DC/DC converter you promised in the other parts? I'm very interested since the EV conversion currently I'm involved with is using the IOTA unit.

[JRP3]

My understanding is that the TS should not be charged to 4.25, you obviously charge them much lower, and that the new SE cells are being rated lower to avoid people over charging them, and they can still go to 3.6. As for capacity, if you replace the 90ah cell with a 100ah cell and charge it with the other 90ah cells, the 100ah cell will never be overcharged. I don't see how it matters if a newer cell of the same or better capacity is physically smaller.

[marionrickard]

The maximum charge voltage on the new BlueSky SE cells is 3.6v while the maximum charge voltage on the Thundersky LFP cells (made by the same manufacturer) is 4.25 volts. They also have differing low voltage cutoff points. But worse, they are not the same capacity. The BlueSky is 100Ah in the same form factor as the 90Ah TS. It just doesn't work. And this is over a 6 month period. Over three or four years, it will be even more of a problem.

[JRP3]

On replacement cells, as long as the cell you use has a bit more capacity than your lowest cell and is never over charged or over discharged, I'm not sure why there would be a problem mixing different types of cells. Yes, it might throw off your monitoring setup if it ran in a different voltage range, and you might need to monitor it separately from the rest of the pack, but as long as it's better than the older cells I think it could work. Buying an extra $1-$2K of cells seems wasteful.

[AndY1ksi]

As always, it's a pleasure to watch & listen to you :-)

Very informative! Thank you!