 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Electronics : a new way to balance lithium ion battery
  Page 6 of 11   |
|||||
| Author | Message | ||||
| Warpspeed Guru  Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
That may be so. But I have also heard horror stories about University students buggering a multi thousand dollar battery bought in for a student EV project by being careless only once. I am neither Clint Eastwood or God, so I choose to be cautious with my considerable investment. Cheers, Tony. |
||||
busman Newbie  Joined: 30/10/2016 Location: AustraliaPosts: 21 |
Tony, you are horrified by a few seconds ? Of what, an overcharge situation? We had one member who had a situation where there was a problem when he was away for 6 weeks. I would be willing to bet your uni students dropped a spanner across some terminals ? Yes, you are perfectly within your rights to protect your investment if you feel you must develop all this stuff. I am protecting my investment without it, time has proven it unnecessary as far as we are concerned. Just don't want to see people waste their time and money, simple as that. William |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Hi William, You done it again  , mentioning a product I have never heard about."Supercombies" I mean. So I googled it. It appears to be a 24V/3KW power management system that's on special from Jaycar at present. Is that what you were writing about? If so, do you use this in in place of the inverters we are building? Just curious  Klaus |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
I agree with Busman, there is little need for balances if you dont push your batteries to the max charge level, although as the cells age there is a need to balance cells, as each cell ages at different rates. To offer a couple of suggestion from my knowledge of building equalizes, and no i can not offer how to do the circuitry, but can offer a few points of view to methods of how to equalize cells. Firstly you can use a clunky bulldozer to move a mountain of power in a crawl pace, hence super caps and relays, all good it will work. Then you can use a fast cycling system (bucket and spade to equalize the mountain) that moves small amounts of power at Khz (1000s time per second) between the cells. ( far smaller caps required) The method i prefer is going the other way around and dumping power from the high cells to equal the low cells. Think about it at the point of equalizing you have excess power input and trying to get a mV adjustment to the charge voltage per cell, so yes you can afford to dump power. IMO it far easier to dump power per cell than shuffle 2/5 of bugger all power between cells. Food for thought! Pete. Sometimes it just works |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
The professional EV people are pushing the limits of technology with these lithium cells, and users are more often than not just dumb commuters that could not care less about much of anything, and a vehicle battery sees some pretty hard use. The EV people use a set of interlinked microcontrollers that switch tiny surface mount resistors in and out across individual cells. Only a very few milliamps, but it appears to do the job. Just something very small to shuffle around a tiny amount of power. And there is absolutely no reason why it cannot be made part of the cyclic cell voltage monitoring system that needs to be there anyway. Cheers, Tony. |
||||
| busman Newbie Joined: 30/10/2016 Location: AustraliaPosts: 21 |
I don't know why Jaycar has not deleted that advert, some time ago they sold out of the range when the guru who was looking after that line left the company to join Telstra, the Jaycar/Electus/Soanar group burn through sales execs like a wildfire, if you don't want to do 18 hour days don't apply ! Great range, the SC is a power management centre, charger, inverter and controller of satellites like solar regs, DC/DC switches for DC charging, same switch which is used for load control etc etc. Very versatile and built like an ox. Data linked to all satellites. I think our original post explains our system. As of today I am advancing our home system. We had a power outage for a few hours this morning and we could not open the garage door ! We have a 3 Kw system now which we get 52 cent per Kwh but when the grid is down, does it power our house ? NO it don't we found out. So we will isolate that system and put all it's output to the grid, use my spare SC and solar reg and buy 2Kw of panels, with 400 ah of Lifepo4, that will power the house, with a switch to enable grid feed as well (this can be done via the SC), you can select how much grid or battery you want) Tony, your last sentence, "monitoring system that must be there anyway", well that's the part we found unneccesary. Pete, that's the way our group developed, first burning off excess with a globe, then we rethought and added to the lower ones with Trev's little chargers, eventually fitted a full time balancer and did away with the cell monitoring, just put in place a last line of defence with tight parameters and could not be happier, this is exactly what we will replicate at the house. I got sick of not being able to use the full potential, read capacity, of these banks so we do charge to what some would regard as extreme, 27.8 mostly, just nudging 3.5 per cell. The balancer at about $35 per cell is the cheapest and most effective insurance I could find, and you know what, it works ! Cheers William |
||||
| yahoo2 Guru Joined: 05/04/2011 Location: AustraliaPosts: 1166 |
Hi Tony, from my limited experience with commercial EV's they seem to have some pretty tight criteria on when they balance. it will be within specific pack temperatures and in a narrow window of voltage and under a very low charge rate. there will be some cars that meet those conditions most weeks and others that could go for months and not balance. Based on what I have been told from car owners that have been digging around in the CANBUS data, they dont see any indication that packs are balanced continuously despite what some companies say. they also run a 12 volt system that powers the contactors, some sensors and the computer that gets around a lot of the issues with powering things from the high voltage pack. Under normal use with a home battery system I think you could measure the voltage after a month and the cells would all be well within a twentieth of a watt of each other maybe the cells on the ends are different because they are always colder than the rest. we could probably also just switch the whole pack off if the pack voltage is high or low and 95% of lightly used systems would do 1500 cycles without an issue. we are only talking here about what will happen under exceptional or unusual circumstances. So the example I put forward where one cell has 10 watthours more energy than the rest is going to be rare. it will only show up IF we charge very close to or over 100% I guess if you are sampling every minute the voltage is perhaps going to be over 4 volts for less time than that. For a once only event for a few seconds any damage done will not be measurable. Assuming the problem is found and repaired and not repeated every cycle. I have noticed as the LFP banks I have installed age, the owners tend to chat a lot more about who's lead acid battery bank has died since I last visited and they take great delight in rubbing it in when the grid power goes down. It will only get worse as more of the cocky sods get past 2500 cycles, that is the point that most of them are financially well ahead of the alternatives. The problems we have had are things like mice, flooding, vandalism and smatterings of general stupidity. Downwind makes an excellent point about not trying to save every last drop of energy out of a design, it feels a bit ridiculous to be saving 50 watts a day with latching contactors when we are generating 4 kw+ per hour surplus for 6 or 7 hours. it is also tough to spec a contactor for 300,000 switching events when I know it will do less than 400 over a lifetime and maybe 10 of those under heavy loads. cheers yahoo I'm confused, no wait... maybe I'm not... |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Well, we shall see. Just airing a few ideas right now, always anxious to learn from the old hands. EV power no longer list individual cells for some reason and EV works say zero stock of anything, back order only. I asked when they were expecting to have some stock, and never even received a reply. I don't like the idea of handing over several thousand dollars and hearing in six months from now that the company has ceased trading. So who in oZ has Lithium cells sitting on the shelf ? Cheers, Tony. |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Did you check with Trev (drive by Nature)? He advertises on this forum, I got my lithium cells there. Klaus |
||||
| busman Newbie Joined: 30/10/2016 Location: AustraliaPosts: 21 |
I thought he supported this forum ? Would not buy from anyone else. William |
||||
fillm Guru Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Yes , contact me of forum if you want info,, I have 200 Winstons in stock and balancers / monitors loggers ..... ect . Follow the links at the bottom. PhillM ...Oz Wind Engineering..Wind Turbine Kits 500W - 5000W ~ F&P Dual Kits ~ GOE222Blades- Voltage Control Parts ------- Tower kits |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
O/k, thanks guys. Will chase that up. Cheers, Tony. |
||||
Gazz Newbie Joined: 22/03/2017 Location: AustraliaPosts: 20 |
Simple Analog BMS - CURRENTEVENTS - Victor Tikhonov - Metric Mind Corporation - December 2012 View: 2017-03-22_073942_Simple_Analog_BMS_-_CURRENTEVENTS_-_Victor_Tikhonov_-_Metric_Mind_Corporation_-_December_2012.pdf I need Balance circuit for 17x 3.2v Lifepo4 100Ah Cells in Series. The commercially available 17S Balance products are passive voltage based PCB. Specifications: LiFePO4 100Ah Battery Bank - 40 to 60v. 17x 3.2v Lifepo4 Cells in series. Max Charge current .5C (50Ah). Max Load current .2C (20Ah). Charge the cells using Partial State of Charge (PSOC). Charge and discharge voltage is changed to match required load capacity. In my case the charging voltage is adjusted and applied from Solar Controller. The Low Level Disconnect will also be separate, so the balancer needs to standalone. Looking for someone to help with PCB design. I have built many simple circuits using 555 on Proto Strip Board. Building 17S system on Proto boards would be beyond what I could assemble with confidence for $3000 worth of Cells, and it would be a lot tidier on single PCB. It requires the use of separate 555 package for each channel, due to shared V+ supply on Dual or Quad Packages. If you are interested in 17S standalone balancing PCB, I would like to talk about collaboration to get some PCB's made and assembled. Do it here, or I have a few contacts on Alibaba. Cheers, Gary Note: I was debating to post here or "PCB Manufacturing" section. Feel free to move it across or I can do it if better to start a new thread for PCB. Gary |
||||
| Gazz Newbie Joined: 22/03/2017 Location: AustraliaPosts: 20 |
In a perfect world, we might want to transfer from high to low cells. Perhaps we could make an inverted hybrid with "Simple Analog BMS". On the same theme if the clock was slowed down, and capacitor had time to charge from averaged reference voltage before discharge into low cells. I have not thought the ideas through, just putting something forward for discussion. Cheers, Gary |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Gazz, just finished reading through that Victor Tikhonov article just a few moments ago. Very interesting. I had thought about a vaguely similar idea a long time ago, but using voltage to frequency instead of voltage to period as he does it. His idea is much better, because it gives a proportional control of discharge. At the end of the second article he mentions hard on/off control for the discharge resistors which is how my (inferior) system would have to have worked. Its not simple to do because the individual cell electronics must work down to 3.0 volts which is difficult to do. The 7555 is rated down to 2.0 volts, a normal 555 only down to 5.0 volts minimum. The mosfets have me interested, I cannot find any any data on those NTD4984 he mentions. Finding something with a suitably low gate threshold is also not easy. I still think Klaus's idea of a capacitor hopping system has great merit. But some thought needs to be given to relay contact life and eliminating the inrush current surge problem. I am certain a suitable inductor in series with the capacitor would work. The current would gradually ramp up from zero on contact closure slow enough to prevent any damage to the relay contact. The L and C would create a resonant ringing of a very few Hz that would die out over a few cycles. When the relay contact opens there will be no stored energy remaining in the inductor, capacitor displacement current should have reduced to zero. Its not too difficult to arrange the relay switching in a way that prevents the possibility of any accidental shorts in any conceivable relay or electronic failure mode. The cell sensing wires should all be very lightly fused anyway. I plan to eventually build something like this to both log and monitor individual cell voltages, as well as provide a very weak low power continuous cell balancer. Cheers, Tony. |
||||
| Gazz Newbie Joined: 22/03/2017 Location: AustraliaPosts: 20 |
Hi Tony, I understanding the basics of Victor Tikhonov circuit, and keen to make a PCB. Downloaded the free version of AutoTRAX DEX 9. This is new to me, so hope I can get some help to get it right. Open to paying someone with confidence to design the PCB, see what happens? I also like Klaus's concept, but do not understand the circuit design well enough to build it. My previous thought bubble: But if all cell voltages could be summed and divided into an average reference voltage, this charges the cap between cyclic connection to each cell. The cell is either, above, under or the same. This might not be any better then passive dissipation. Maybe it will spark an idea. Alternatively, there are a few commercial 17S systems I could buy, but its difficult to extract the IC or mythology of balance from sellers. What are you getting? Gary |
||||
| Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138 |
Hello Gazz, I too have a large Lifepo4 bank 16 cells @400AH and have thought of another way to do the cell balancing when the individual cells are nearing 3.5 volts, which is where I terminate the solar charge current. Discharging a resistance load across each cell as it nears full charge on these big cells is dubious at best, even though the solar charger will throttle back the charging current as the pack voltage reaches a set point, this could still be quite high relative to the individual cell monitor discharge currents. My idea is a precision voltage ref and a pulse generator circuit across each cell, the pulse generator supplies a flyback transformer that feeds the rectified isolated output to the whole pack. If the average power from each flyback output was around 50 watts this would mean an individual cell discharge load of 15 amps with the power fed to all cells in the bank rather be wasted as heat. As more cells reached termination voltage relative to those not quite there yet, the lower cells would charge up faster thus bringing them into balance. The circuit across each cell could be quite simple, precision reference diode, pot, 555 supplying 20us pulses which increase in freq with cell voltage, mosfet powering a small ferrite transformer. Something this simple wont monitor individual cell voltages, but really is there a need to in this situation. Cheers Mike |
||||
| Gazz Newbie Joined: 22/03/2017 Location: AustraliaPosts: 20 |
Hi Mike, I was adding above and just read your post. Slow learner, but I know what I know. Got lost at flyback transformer. Take me a while to get up to speed on all this. Can you elaborate on how what flyback? Thanks Add: Because I am using different PSOC during the life of cells, my balancing curcuit must be functioning at all charging capacities to prevent cell runaway. Gary |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
One problem with ANY cell voltage measurement is the margin for error through component differences and component drift over time and temperature. Seventeen cells is not huge number, but it does mean at least 17 potentiometers that need to be individually tweaked initially for each cell. If all the individual cell voltage measuring systems drift out of whack, the whole voltage balancing idea flies right out of the window. All professional quality data loggers use high quality relays, and connect one highly precise voltmeter sequentially across everything being logged. Any slight error in measurement will be the same for every input, so monitoring cell balance can be done much more precisely and requiring no adjustments. That is the power behind Klaus's capacitor hopper. There are no initial adjustments required and no way it can get out of adjustment. Fail yes, but not require a careful regular routine test and re tweak of a whole bunch of potentiometers. All you need is a method to very slowly sequence some relays on and off, it need not even be electronic. Cheers, Tony. |
||||
| Gazz Newbie Joined: 22/03/2017 Location: AustraliaPosts: 20 |
Not to literal, just the concept. Can the bank voltage by divided by 17? Guess that would be to simple to work! Thinking about the slow sequence relays. I designed a latching solenoid valve many years ago. Got fascinated with the latching thing for a while. As in your post, the sequence is key to making this work. I like to keep an open mind to many ideas, even if they seem like the wrong ones. The Simple Analog BMS converts voltage to time and favors low cells, I feel the capacitor could possible be incorporated to create a Mikey Mouse system. Maybe? Gary |
||||
| Page 6 of 11 |
|||||
 Print this page |
