[etischer]

Originally Posted by Voltswagen

So you are suggesting very small cells for example...the A123 cells?

The original question was asking about the Tesla pack. I read though Teslas pdf too quickly...
"The Tesla uses 186540 form-factor cells", 6800 of them.

18650 = 18mm diameter, 650mm length.

[JRP3]

6831 actually

[tomofreno]

"The Tesla uses 186540 form-factor cells", 6800 of them.

Wow, a lot of development and reliability lab effort must have gone into making low contact resistance, reliable connections.

[lnstech]

I have a mail order company that sells electronic kits (www.techkits.com). I am also an EV builder for hobby. I would be willing to sell the Lead Acid version of the Volt Blocher kits if I could get Brian's permission & schematic. I have tried to contact him by email, but with no response yet.

[jrickard]

Originally Posted by Propellator

allow me to throw in some experience from a different field of use.
For the last 5 years or so I use LiPo cells in my radio controlled models.
In the beginning balancing was virtually unknown and my first packs were charged and discharged without balancing. But since they were hilariously expensive I never really charged them to the top and never discharged them completely. Those cells still work today.
After the top notch sponsored pilots started to balance their packs (at the top) this technology quickly spread to the point where battery packs came with balancing cables already attached (to pick up single cell voltage for the charger).
LiPos became more and more affordable and so, when a cell in a pack became pregnant, we considered the cells to be of bad quality and that his would be the statistical failure one encounters once in a while. So we continued to balance at the top and accepted the occasional cell failure.
Never even considering that the problem might be the preprogrammed cutoff voltage (measured across the whole pack) of the controllers we use.

Last year I started to use a controller which employs the single cell voltage pickup (because it was available and fit the model) and from then on I never had a cell failure again.
The problem was solved, I thought, because cell quality became better and better. Sadly my grey matter didn't allow me to make the right conclusion.

So I have to say that, when watching Jacks video on Sunday, jaw dropping and forehead slapping, it was an extreme pleasure to witness one of the rare eureka moments in engineering.

I'm now off to check whether the radio control guys knew about the balancing issue all along.

I'll add an interesting aside to that - and I get a LOT of info from RC guys. I probably have nearly a dozen of their little balancers here. Yes, some do charge at the top. By far the majority of what I have, do a very small 100 ma or 200 ma BLEED to the LOWEST cell voltage. Cute little aluminum cases. Typically do 6 cells. I have one that does 8. If you hook it up without a charger, it bleeds each cell in turn in a kind of cycle until they are all at the level of the lowest cell.

Jack Rickard

[jrickard]

Originally Posted by Dolphyn

Going back to the original topic of this thread, I think the EV Power cell modules from Australia would be a good replacement for Volt Blochers. No need to buy the "master unit", since the modules can be used on their own (or with a wire to detect if any cell has high or low voltage). They can be purchased here. I inquired about shipping to the United States, and the cost is reasonable. Enjoy!

Actually, I suspect the head unit is the most important part. What I THINK they're doing (I don't have mine yet) is summing optoisolators. The forward voltage on these is something, I don't know what. But say 1/2 volt. Over a hundred of them in series that's 50 volts. And you're not really isolated from teh pack.

I think the head unit converts this to a relay switched. Actually, it can do several things. But the most important is it isolates you from the optoisolator loop. I know what isolators are SUPPOSED to do, but you have to power them somewhere, and to do it with minimum wires you have to series them.

Jack Rickard

[jrickard]

Originally Posted by bblocher

There is a lot of good talk going on here. It's hard to be on the cutting edge and trying to figure all this out so I wanted to throw in my experience as well as it seems to be quite different than others are describing here. I have 10k miles on my pack now and the characteristics of the cells have changed over time.

When my cells were new the VBs had quite a bit of work to do to get them all topped off. I bet it took a day at least. After that each charge there wasn't much time needed to balance and you could argue that a little balance every month, or maybe never again would be fine.

Seven months and 10k miles later it's completely flipped. It takes 80% (technically a bit less) of my pack to get to work (35 miles up hill). I noticed I started getting cells that were lower and lower in voltage when I arrived to work as the days went on (it's also cold now). Eventually one cell hit 0v on the way up and was lost, it wouldn't charge. I had a couple of other cells that I could see would be next. I manually charged these cells and they were fine for a few days (even being the first to charge up instead of being the last like before). Within a few days they were the lowest cells during charge again and low when I arrived to work. I ended up shunting even longer during a charge to give these cells more time to charge and now all is good. I can make it to work without them dropping too low in voltage.

So it seems the internal resistance of these cells are all changing at their own pace as they age and creating a lot of problems for series charging. It's even requiring more shunt balancing than I expected. I guess you could just replace these cells, but I'm not sure if others would soon follow or not. There is more to learn here on how cells change over time, but I couldn't disagree more that nothing needs to be done. Maybe if you don't need but half of your capacity this is the case.

I know this goes against some theories that a few of you are tossing around and I'm not trying to discredit anyone. This is just my experience thus far.

It would seem to me that individual cell charging (of course what a pain for an EV) is the best solution for new and aged cells.

Ahemmm.... I hate to be insensitive. And I've kind of made you the poster boy. Understand, my own personal belief is your a HERO that designed a circuit to do something quite well, it was just the wrong thing, although EVERYONE, including me, thought it was the RIGHT thing at the time.....

But you top balanced. And you have a 0 volt cell that can't recover. With 10,000 miles. And a couple more going that way....

That's EXACTLY what got me started thinking. I had a smaller set of cells, right in front of me, in a vehicle where it was NO problem to reach any terminal, with a voltmeter in my hand, and I had to go through this THREE TIMES in as many days. So I'm real sympathetic here. The old thinking doesn't just go away by itself....it takes actually QUITE a bit of whiskey....

Here's a couple of graphs that might help.







Jack Rickard
http://evtv.me

[Dave Koller]

Originally Posted by jrickard

Actually, I suspect the head unit is the most important part. What I THINK they're doing (I don't have mine yet) is summing optoisolators. The forward voltage on these is something, I don't know what. But say 1/2 volt. Over a hundred of them in series that's 50 volts. And you're not really isolated from teh pack.

I think the head unit converts this to a relay switched. Actually, it can do several things. But the most important is it isolates you from the optoisolator loop. I know what isolators are SUPPOSED to do, but you have to power them somewhere, and to do it with minimum wires you have to series them.

Jack Rickard

Jack:
Isn't this the schematic for those at (first page) http://www.aeva.asn.au/forums/forum_posts.asp?TID=900

[JRP3]

Jack, any chance you could resize that for better readability? I'm scrolling up and down and back and forth just to see it.

[jrickard]

Originally Posted by Dave Koller

Jack:
Isn't this the schematic for those at (first page) http://www.aeva.asn.au/forums/forum_posts.asp?TID=900

I don't know for sure. But a device built to that schematic would do precisely what they're describing on their web site.

Jack Rickard