PLEASE NOTE: If you had an account with the previous forum, it has been ported to the new Genetry website!
You will need to reset the password to access the new forum. Click Log In → Forgot Password → enter your username or forum email address → click Email Reset Link.
Checked an active balancer listing on eBay using caps. If they use generic electrolytic caps, that's going to be a bit of an issue...because a genuine 2200uF 6.3v electrolytic cap is only rated at 1.19A ripple current (2,000 hours).
But the caps look a bit like a new style that I haven't seen before: Aluminum Organic Polymer Capacitors. Seems they might have a half a chance with those...assuming they're true genuine parts, they might be rated to 5.5A ripple current, though still only 2,000 hours.
Maybe not too horrendous in lifetime concept, though the design still has a fundamental flaw.
Great explanation!
Your balancer is active, yes? Guessing you have it moving current between cells, then at some point dumping excess as heat.
In my case with the used byd bank, charging is cut off at 27.2v (54.4v for the 48v), so the high current problem you talked about above shouldn't be much of an issue. Or am I wrong?
58 minutes ago, dochubert said:Your balancer is active, yes?
No. The balancers I have / want to offer for sale on Genetry Solar are total-loss shunt balancer units. Moving current between cells is a great concept...but the "scaled up on a board" designs commonly available on eBay have a serious fundamental flaw. With the quickly multiplied current, the potential for overcharging cells (as a result of balancing) is very high.
58 minutes ago, dochubert said:In my case with the used byd bank, charging is cut off at 27.2v (54.4v for the 48v), so the high current problem you talked about above shouldn't be much of an issue. Or am I wrong?
If the active balancers are working great for you, then enjoy 'em. I've never used them personally (probably because of a combination of home-made shunt balancers and YouTube videos of active balancer fails).
Did a smidge of research online...found a post on the DIYSolarForum where someone explained that these "active balancers" treat the battery cells as pairs, and just bleed power across each pair as necessary. But if both cells in that pair are too high, now what??!?
https://diysolarforum.com/threads/how-to-correctly-install-a-active-cell-balancer-to-lifepo4.9224/
I dunno, I'll stick with my total loss shunt balancers 😉
How hard would it to make a controlled Balancer with a controller and a way to monitor it as well as set the specs you want?
Seem like you could make one with four cell output and one input for a charger and have the CPU handle the rest, of course you want to make sure it don't go over the voltage, but as long as it stay in specs and the solar charger and bms sees the correct voltage it will keep from over charging the whole battery.
Reason why I ask, is that is how I'm balancing my built batss right now, I use the balance wire, and connect gator clips to them and send the 3.60v charge intot he battery cell that needs to be bumped up.
Only problem is I can only do one cell bank at a time, cause you got to move the negative to the next balance wire and the postive to the next in series to charge each cell bank. With a controller, it could do that by it self, and maybe even at a higher charge rate. The flycow charger I got goes down in amprage as it gets closer and closer to the voltage. Which is good, but, is a pain to wait for it to finish.
Using electrolytic caps in a charge pump method is great for high currents...but not for longevity. I seriously doubt the caps have a 5 amp ripple current rating (you have to get physically huge caps to get that level of rating)--and even if so, most electrolytic caps are only rated for 2,000 hours (that's 3 months!) at max voltage/temp/ripple current ratings. With that in mind, I don't see them lasting very long before the caps dry out and fail.
Yep. And when you get into ones that large, you can expect to fork over some large amounts of cash for them.
I took a chance on some used Cornell Dubilier 6300uF 400 Volt ones last year. Looked like brand new and all I have used and checked so far have been right on the money as for the ratings. They can handle a 14.4Amp ripple at 120Hz. But to buy just one new would cost you $150. But the specs on them are kind of way out there. 6300uF -10% to +50% and surges to 450VDC with those tested coming in at 6800uF. But also like you said, only rated for 2000 hours at those ratings.
2 hours ago, The Blind Wolf said:How hard would it to make a controlled Balancer with a controller and a way to monitor it as well as set the specs you want?
Beyond the scope of my interests. The very simple high-amperage shunt balancers have absolutely no limits to the pack size...you could do anywhere from 2S to 600S if you wanted. As soon as I make a "smart block unit", suddenly we're limited to processor precision and design block size, etc.
It is worth noting, however, that the next revision of Genetry Solar shunt balancers, will have RS-232 string communication down the cells, allowing monitoring of individual cell voltages.
It would be completely possible (and is in the distant plans) to make a "system central controller" which all of the system units connect to (MPPTs, inverter[s], battery monitors, Arduino projects, etc.), and the user could completely customize every single aspect of interaction between all of the devices.
Most chintzy Chinese 0.03A balancers SHOULD have a cell over-voltage / under-voltage limit, which potentially could be utilized as a shutdown trigger if needed.
Beyond the scope of my interests. The very simple high-amperage shunt balancers have absolutely no limits to the pack size...you could do anywhere from 2S to 600S if you wanted. As soon as I make a "smart block unit", suddenly we're limited to processor precision and design block size, etc.
Is the rs32 the 32 pin thing like on my raspberry pie ?
Starting to play around with those as of late. Getting a new on, the one that is a keyboard.
RS-232 is the "formal" name for plain ol' asynchronous serial data transmission. On older computers, it was the smaller DB-9 connector (as opposed to the big DB-25 that often was a "parallel port" for printers, etc.)
RS-232 is the "formal" name for plain ol' asynchronous serial data transmission. On older computers, it was the smaller DB-9 connector (as opposed to the big DB-25 that often was a "parallel port" for printers, etc.)
ah the Com port 1 and 2. Yep. I think the printer port is lp for the 25 pin.
6 hours ago, Sid Genetry Solar said:If the active balancers are working great for you, then enjoy 'em.
I'm learning as I go. Don't know yet if they will work for me. Learned a lot in a short time due to your insight and experience, but I'm still a dumb@@s compared to you.
6 hours ago, Sid Genetry Solar said:But if both cells in that pair are too high, now what??!?
This is a question that needs answering if I stick with active balancing. Your analysis seems valid. The design flaw seems evident, making the current set of active balancers unlikely to last long and possibly hazardous. Shunt balancing seems much safer, but not available for sale and I'm not talented enough to build my own. Yours aren't ready for sale yet. My battery bank needs balancers. A bms can't keep up. Where do I go from here?
Since I'm keeping my charge level well below the 29.2v max at 27.2v, it leaves some room for transients. Hopefully that will do until an alternative becomes available. My only other option is to just continue to let the bms do the best it can. Current plan was/is to add the active balancer to the bms and let them fight it out. When the new batteries get here, I had planned to parallel all the cells on the new 48v set (16s4p) and use one 5a 16s balancer for the whole set. After this discussion would prefer to use yours. Any idea when yours will be available and at what price for 16s? (remember I'm cheap 😁)
Thanks
6 minutes ago, dochubert said:Any idea when yours will be available and at what price for 16s? (remember I'm cheap 😁)
Well, sometime after I order 'em 😉. I'll see if I can get them ordered over the weekend or something like that...then something like another 2 weeks for manufacturing and shipping. The biggest issue for us is the heatsinking, as we don't have a bulk supplier for heatsinks--and let me tell you, these things will put out a LOT of heat. 5A * 4v = 20W--that's more than some small soldering irons. PER INDIVIDUAL CELL BALANCER. If Sean forgets to turn the fan on with the balancers on his Li-Ion battery bank, they get hot enough (before going into thermal trip) that they will melt hot glue.
I'll see what I can do. Lots of things on my plate to do.
Pricewise, I'm hoping to have 2 versions...one with a tiny little 4-character LCD (for registering battery voltage) and RS-232 battery voltage readout...and the other without any of those "smarts." Price for the non-LCD "non-smart" balance shunt hopefully will be somewhere close to $8/ea. The one with the LCD (and RS-232 string comm) will probably be somewhere around $12/ea.
Those prices do not include heatsinks, wiring, etc. You would need 16 units for your system.
I appreciate it, and I'm sure a bunch of others do too.
I'm willing to do my own heatsink/fan setup if that helps.
I'm sure you've seen Sean's videos with the balancer test setup?
Yes, a lot of heatsink. Seems like more fans would help.