[illuminateddan]

Hi all,

So I've been watching Jack Rickards videos and now have an issue. Jack knows his shit (to put it mildly) and is an advocate of the 'no shunt bms' theory. Now I'm building a bms based on peter perkins slave boards and I've come up with a 'variable point balancing system' which is not a top balancer or bottom balancer but a system that balances all the cells to the lowest cells voltage before a charge. Having seen jacks videos I'm now wondering if I should carry on with this BMS and VPBS, especially since reading voltages from lifepo cells becomes problematic when they are under load, therefore a better system is to check cell voltages when the car is at rest, rather than while driving. As my ideas and theories have now been updated, what is everyone else doing? is shunt bmsing still the way for most or is bottom balancing the main theory? How do you measure voltage on cells the fluctuate wildly when under load and not under load. According to some graphs this can be up to 0.3v between load and non load voltages.

aaarrrgghhh! Thoughts???

D

 

[Dolphyn]

Jack assumes that you won't know a cell is getting weak unless all of your cells get weak at the same time. If you have a system that monitors the performance of the "worst" cell, I think Jack's bottom-balancing theory goes out the window.

I use CellLog 8S to warn me when any one cell is running out of juice, and I can use the rest of the pack to replenish that one cell (using DC-to-DC converter from Ebay), so I don't want all of my cells to run out at the same time.

 

[dimitri]

I've come up with a 'variable point balancing system' which is not a top balancer or bottom balancer but a system that balances all the cells to the lowest cells voltage before a charge.

This describes bottom balancing, so why do you say its not?
If you want your pack balanced, you pick either top or bottom, since those are the only practical options based on voltage reading. Bottom balancing is usually done manually and not frequently, since its time consuming. Doing it before each charge cycle is not practical because you will be wasting time and won't be able to use the car until done. Top balancing is much easier to do at every cycle and can be always skipped or cut short if you suddenly need to drive the car.

Either balancing method typicaly uses shunting, so its funny when people are against shunting, yet keep talking about balancing their packs. Any non-shunting balancing method uually involves lots of power switching components and gets too complex/expensive to be worth the trouble.

Why does it have to be a debate? There isn't one true solution for all, many are perfectly valid. The best one is what best fits YOUR needs for YOUR pack, YOUR car and YOUR wallet.

Jack knows his shit

Now, this one we can certainly debate.

 

[Guest]

Jack knows his shit
Now, this one we can certainly debate.

Can be said about anyone. Knowing ones shit is quite subjective.

 

[dimitri]

Can be said about anyone. Knowing ones shit is quite subjective.

The smiley face was a hint that I was making a joke

 

[Guest]

I can be on the light sided end of things.

 

[illuminateddan]

This describes bottom balancing, so why do you say its not?

Bottom balancing is performing a balance when the pack is at it's lowest point, say 3v. Variable point balancing is a frequent shunt balance before every charge. Here's how my system works:

1. When you plug in the car the master controller -EVMS (electric vehicle management system) sends out a voltage data request to all the slave boards.
2. The slave boards respond with their ID and voltage.
3. The EVMS then looks through the data and finds the lowest cells voltage - lets call it 'nV'.
4. The EVMS then sends out a packet to all the slaves instructing them to set their discharge point to 'nV' and standby.
5. The EVMS then requests the slaves to send back their discharge point for confirmation, which they do.
6. The EVMS then sends a 'GO' command and the slaves discharge all the cells to the lowest discharge point 'nV'.
7. The slaves then deactivate their shunt discharge and the master initiates the charge sequence.

As the cells are being matched to 0.01v the pack should always stay fairly balanced. Bleeding off a tiny amount (a few hundred mAh ish) of juice is far more energy efficient than bleeding lots off the top, and having the system automated in the charge cycle saves lots of time.



As to whether Jack knows his shit or not, well that is another debate. (although the weight of imperical evidence goes in his favour as he has the cars, the equipment, the workshop, the HD TV show, the manufacturer support and the experience!) Anyone else take the trouble to make a 105min video showing a cell charging and explaining whats happening?

 

[mhud]

Anyone else take the trouble to make a 105min video showing a cell charging and explaining whats happening?

This is my problem with Jack. He often has good info, but he repeats himself constantly and takes ages to explain things.

I'd love it if he'd distill his thoughts consicely and put them out there. Instead it seems like he hits `record` and just goes.

But, one can't deny that he has got experience. That's the problem. Do you subject yourself to 1.5 hours of Jack to see if you'll learn something? Maybe I'm just spoiled in the age of Internets.

Anyway, my $0.02 is that it's worth monitoring individual cells and wiring a dc/dc to cycle through the weakest ones while driving, propping them up. This is the only way I know of to get around the weakest cell being your range limiter.

But I'm a newbie with a lead-acid pack, so value my words accordingly.

 

[DIYguy]

Anyway, my $0.02 is that it's worth monitoring individual cells and wiring a dc/dc to cycle through the weakest ones while driving, propping them up. This is the only way I know of to get around the weakest cell being your range limiter.

But I'm a newbie with a lead-acid pack, so value my words accordingly.

Hi mhud,

I have a pack of AGM's which need to be balanced ( without boiling) My charger will do this perfectly as long as there is balance in the pack. So, I use these..... http://www.hdm-sys.com/pdf/hdm_equalizer_specs.pdf
These shunt up to 5 amps during charge, discharge and idle.

So, the charger takes care of overcharge, the controller takes care of running them too low, and the balancers keep em all identical siblings.... The batteries will last longer, and you will get the best performance out of the pack.

 

[Guest]

I watch Jacks show because it is informative and besides being a bit slow it is a good late night watch. Minimized and concise would make for a boring and quick show. What fun is that. Sure Jack's show is not like our evening drama shows but it's fine. I'd like to see more how to's like actually building Lee Hart Shunts and other stuff like that. I hope that stuff will come. Or maybe someone else can do a howto program. Mmmmmm. Also a show a bit more focused on those of us with far less money to spend on building an electric car.

 

[mhud]

Hi mhud,

I have a pack of AGM's which need to be balanced ( without boiling) My charger will do this perfectly as long as there is balance in the pack. So, I use these..... http://www.hdm-sys.com/pdf/hdm_equalizer_specs.pdf
These shunt up to 5 amps during charge, discharge and idle.

I have AGMs too, so overcharging is a serious problem for me as well. I didn't find the HDM balancer until I'd bought my PowerCheqs. They're working well for me. A single unit like yours which could do 10-12 batteries would be awesome.

gottdi, I'd absolutely love to see or hear someone talking about the Lee Hart system. I admired that and was considering building it instead of buying my PowerCheqs. A new system like that which could prop up individual Li cells would be great. If you wanted to limit the cost, you could balance blocks of 4 Li cells with 12V hardware and go off the existing Lee Hart documentation. When you think about it, the Pb cells are 6x2V cells that we don't bother to monitor individually...

 

[illuminateddan]

Yup, I'd agree that he does repeat a lot and the show could do with an edit (and maybe removing the chip on the shoulder, but it does make good late night viewing while doing other things.
The trouble is, there is still no firm answers on what is the best method for looking after lifepo packs, and there needs to be a consus on a definitive besat practice method.

 

[Duncan]

Hi Illuminateddan

Bottom balancing is performing a balance when the pack is at it's lowest point, say 3v. Variable point balancing is a frequent shunt balance before every charge. Here's how my system works:

1. When you plug in the car the master controller -EVMS (electric vehicle management system) sends out a voltage data request to all the slave boards.
2. The slave boards respond with their ID and voltage.
3. The EVMS then looks through the data and finds the lowest cells voltage - lets call it 'nV'.
4. The EVMS then sends out a packet to all the slaves instructing them to set their discharge point to 'nV' and standby.
5. The EVMS then requests the slaves to send back their discharge point for confirmation, which they do.
6. The EVMS then sends a 'GO' command and the slaves discharge all the cells to the lowest discharge point 'nV'.
7. The slaves then deactivate their shunt discharge and the master initiates the charge sequence.

I have a couple of questions,
In the other systems I have seen the "SHUNT" does not seem to work by discharging the cell - which would require some way of dumping the energy - but works by permitting the charger to -not charge- that cell by bypassing it

If you use that system and your logic you could use the shuts to "bypass" all but the weakest cell, charge that and then remove the bypasses to cascade the other cells in sequence to charge them as well.

Does that make sense??

 

[Guest]

Yup, I'd agree that he does repeat a lot and the show could do with an edit (and maybe removing the chip on the shoulder, but it does make good late night viewing while doing other things.
The trouble is, there is still no firm answers on what is the best method for looking after lifepo packs, and there needs to be a consus on a definitive besat practice method.

Well I'd suggest a better opening and ditch that little house on the prairie style garbage. He needs something more in tune to the application. But not what he is using. The opening sucks. With all the crap and flack he gets I am surprised he does not just shut down his video for us all to view and go about his work in silence. He has the right and has earned that chip on his shoulder. He has given lots of proof and has shown his metal. He has put his money where his mouth is compared to few around here. Like he said, it's easy to type yourself smart, it's another to prove it. He has proof. Freely given proof.

Pete

 

[illuminateddan]

Well I'd suggest a better opening and ditch that little house on the prairie style garbage. He needs something more in tune to the application. But not what he is using. The opening sucks.

Amen to that. I go make coffee while the intro comes in.

 

[illuminateddan]

Hi Illuminateddan

I have a couple of questions,
In the other systems I have seen the "SHUNT" does not seem to work by discharging the cell - which would require some way of dumping the energy - but works by permitting the charger to -not charge- that cell by bypassing it

If you use that system and your logic you could use the shuts to "bypass" all but the weakest cell, charge that and then remove the bypasses to cascade the other cells in sequence to charge them as well.

Does that make sense??

Hi Duncan.
When I refer to shunt balancing I'm actually just using the common term for 'a fat resistor to discharge'. so my bad. My system doesn't shunt current from cell to cell, it just discharges cells. I'll be charging at 30A so shunting 30A across a cell is hard and costly in heat terms.

Even with other shunt systems, all they are doing is discharging the cell while it charges, usually at about 1-5w. This seemed a little non sensical to me so I had a rethink.

One of my main issues is gather cell V data and making it relevant - i.e. v under load is miles away from V under no load.

d

 

[Guest]

Amen to that. I go make coffee while the intro comes in.

I just download first then just jump past that section.

Pete

 

[JRP3]

Yup, download and Fast Forward are your friends. I couldn't make it through the third drive in the Speedster.

 

[mhud]

Yup, download and Fast Forward are your friends. I couldn't make it through the third drive in the Speedster.

If you play videos in VLC (on any platform) you can play back at various speeds with audio... It's weird at first but pretty handy! And it corrects the pitch so it doesn't sound like listening to chipmunks.

 

[JRP3]

This describes bottom balancing, so why do you say its not?

Sort of. It sounds as if he would pull all cells together even if the pack is only partially depleted so it's not a true "bottom" balance. As you know there isn't much difference in the middle of the discharge curve, so balancing a pack in that range won't give you a bottom balanced pack.

Either balancing method typicaly uses shunting, so its funny when people are against shunting, yet keep talking about balancing their packs. Any non-shunting balancing method uually involves lots of power switching components and gets too complex/expensive to be worth the trouble.

Well you can bottom balance and top balance manually by adding charge as well as removing charge. I use both to bottom balance more quickly if appropriate.

Why does it have to be a debate? There isn't one true solution for all, many are perfectly valid. The best one is what best fits YOUR needs for YOUR pack, YOUR car and YOUR wallet.

Indeed. There is no "best practice" since the ultimate goal is to preserve cells by staying away from the ends of the curve and there is more than one way to do so.