Export thread

Hi all. Over on the following thread:


I talk about the idea of rejuvenating Lithium Polymer (LiPo) batteries that we find in common electronics, although the thread itself is about Palm Pilot PDA's.

Here's some thoughts on rejuvenating Lead Acid batteries. I am specifically talking about Sealed Lead Acid (SLA) batteries of the type that power alarm systems, backup lighting systems, even trailer brakes, and motorcycles. I've been enticed to get into these experiments as I've bought SLA batteries several times which were new but didn't live up to their name plate rating. Batteries in stores sit around for way to long, maybe years. For some NOT fun, try explaining to someone at a battery store why the little cube battery you bought doesn't work even though their volt meter shows it's fine. These techniques can work with car batteries too, but are a little harder to implement. I'm not discussing anything about adding water to batteries but only maintenance free batteries. I did find some good information on YouTube, and lots of not so good information. So, you have to look at lots of info to know what's credible. You could search for rejuvenating lead acid batteries or desulfating lead acid batteries. I'll give anonymous credit to the helpful YouTuber's I found since I can't recall their channel names at the moment.

As I understand it, lead acid batteries hate sitting in a discharged state. Lead sulfate crystals form on the plates and increase the resistance. That means it's harder to charge and, when you discharge it, you get less energy out before the voltage drops too low. Many batteries are rated at the 20 hour discharge rate. Interestingly, you get less energy out of the battery if you draw the energy out faster. So, if it's a 5 AH battery (theoretically but not actually 5 A for 1 hour), then if you draw .25 A for 20 H, that's your 5 AH. The "C" rating is the same as the AH rating but without the hours. So, for the 5 AH battery, 1 C equals 5 A. You can generally charge these batteries at about .3 C. For the example battery, this would be 1.5 A. You can discharge at different rates but higher rates risk overheating the battery and, as stated, give lots less energy. I've found that discharging at the 5 hour rate, or .2 C gives about 85 % of the energy as doing so at the 20 hour rate or .05 C. Check battery label or specs for specifics.

You can charge one of these batteries with a standard constant voltage constant current lab power supply. This gives you more control over the process than with an automatic charger. If you're going to be closely watching the process, you can set the voltage to 14.7 V (in the case of my 5 AH trailer brake battery). If not watching so closely, set the voltage to 13.7 V (this is the float voltage shown on the battery). Then set the maximum current to .3 C, which is my case is 1.5 A. You connect the red positive lead of the power supply to the (usually red) positive terminal of the battery. You connect the black negative lead of the power supply to the (usually black) negative terminal of the battery.

If the battery is working normally, the current will immediately jump to the max of 1.5 A and the voltage on the power supply will reduce to whatever it takes to keep this current flowing. The voltage will gradually increase as the battery charges. Eventually the voltage will hit its limit of 13.7 V or 14.7 V. Then the current starts decreasing. I would consider terminating the charge when the current reaches 1% or .01 C, 50 mA in this case, or possibly .5 % or .005 C, or 25 mA in this case. Don't leave the battery at 14.7 V for a long time. Leaving (this particular) battery at 13.7 V should be safe. Do not charge the battery unattended until you've verified that it goes through a cycle with no problems. Monitor it for overheating. Do not leave it at 14.7 V.

To drain the battery, I use a microprocessor controlled electronic load tester East Tester ET5410 that I got from Amazon. This is a Chinese unit of pretty good quality. I think it was about $ 200. If you want to spend more, you can look at a Rigol product. If you want to spend way more, you could look at Tektronix or HP, etc.

It's not draining a lead acid battery that kills it as some believe, as long as you stay above 1.75 V / cell or 10.5 V for a 12 V battery. It's letting it sit in a drained state and having lead sulfate crystals form inside. To drain the battery, I set the load bank for constant current mode and set the current for the 5 hour drain rate of the battery or .2 C. For a 5 AH battery, this is 1 A. I set it to drain the battery down to 10.5 V. If it's working properly, I expect to see about 85 % of the name plate rating when I'm done. Remember, they're usually rated at a 20 hour rate. After running the drain at the 5 hour current, I'll run it at the 10 hour current (500 mA in this case) and then at the 20 hour drain current (250 mA in this case). Eventually, I'll get around to doing a charge cycle and a drain at the 20 hour rate to check for rated capacity. If the battery has its rated capacity, great. But, in my experience, they almost never do. A battery is generally considered at END OF LIFE it it's at 80 % of rated capacity.

But, you can often bring them back from the dead. Based on watching lots of YouTube videos, the consensus is that the magic desulfators generally don't work. There may be exceptions, but what you need is physics, and TIME. What may work, if the battery isn't internally damaged, is to simply charge it and drain it repeatedly. This may take 5, 10, or more cycles. As I said in the other thread linked above, batteries are weird and they have weird aging processes. Once you drain the battery and the load bank cuts off, the voltage on the plates rises again and you can run the load bank again for a while. So, I'll drain the battery at the 5 hour rate. A little while after the load bank shuts off, I'll start it again, and drain a little more energy out. Then, I'll do it again. Then, I set it to the 10 hour rate and do that a few times with the same battery. At this point, the battery is almost dead so it won't take 10 hours, maybe 1 hour, maybe a few minutes, etc. Then I set it to the 20 hour rate, and do that several times. Note that I'm not charging the battery, just restarting the load bank at lower and lower current levels. Once I consistently get almost no run time at the 20 hour rate, I get ready to charge the battery again. I note how much energy I got out of it on each of the main cycles.

Then I charge the battery again as noted above. I keep a note card with it and record each procedure. If I'm watching it, I charge it at 14.7 V. This pushes higher current into it (still limited by the constant current setting) for a longer time. This may help reverse the desulfation process. I do NOT leave the battery charging unattended at 14.7 V. If I'm going to leave, it's at 13.7 V. Again, note that the power supply will drop the voltage initially to maintain the constant current that was set. I do not leave the battery at all until I know it has successfully charged one cycle. If it starts getting hot or some cells start getting hot, I discontinue charging or reduce current. This is not a set it and forget it process.

I keep draining and charging the battery until I've reached 10 cycles or until it's performing at about 95 % or better of its rating. This is time consuming. Even though much of the process is automated, moving the battery from the power supply to the load bank and setting things takes some time. I've done this to several of these small SLA cube batteries. I've increased their capacity from 50 % to over 95 %. Once I've rejuvenated them, I put them on a "Battery Tender" 800 mA automated charging and maintenance device that maintains them at a float voltage. Note that I use the Battery Tender for maintenance, not charging.

You might say, why go to all this trouble? Just go get the battery replaced. Well, it's not so simple. As I said, the people at the battery store will pop a volt meter on it and say, it's showing 14.5 V or whatever. The battery is fine (even though it's not). They're resistant to do a warranty replacement. By the way, they also won't warrant them below about 10V. Also, it's likely that the replacement battery has been sitting on the shelf or in the warehouse just as long and will have the same problem. For some of these batteries, which I just use for experiments and maybe lighting when the power is out, they're not too critical. But, for my trailer brake battery (the 5 AH one), I darn sure want that thing to work if it's ever needed in an emergency. So, that's why I do this. Also, I just am curious to know how this stuff really works.

I'm also doing an experiment on one 2.9 AH SLA battery that I cannot recommend yet because I don't know the outcome, but you might find it interesting. As I said, once the load bank turns off, the battery voltage will increase. This battery has been very stubborn in completely discharging. I did the 5 hour drain current with a couple of restarts. Then the 10 hour drain current. Then the 20 hour drain current. Still the voltage kept rising at the end. That tells me it still has more potential energy, even if not much. I INTEND to flatten this battery to 10.5 V (not less). So, I've switched the load bank to constant resistance mode. I'm applying a continuous ~ 1000 ohm resistance to the battery down to 10.5 V. It's drawing about 10 mA. Every time the load bank shuts off, I turn it back on. I've been doing this all day intermittently. I'm hopeful that this process is reducing the sulfation internally. Once it really is flat, I'm going to charge it again. I will not leave it discharged.

I'd love to know if y'all have had experience working with batteries in this way.

May your bits be stable and your interfaces be fast. Ron

Just checking voltage is useless, you need to also load test them, to see if they are capable of supplying current for 30 seconds, so I built a quick and dirty tester, using resistance tape from an old toaster, selected to have a 10A draw at 12V, and with a voltmeter to measure battery voltage. Meter is connected to the cables all the time, and a push button allows you to select in the resistor, so you see the resting voltage, and then press the button for 30 seconds, watching the voltage drop under load. Dying ones drop down rapidly to the cut off voltage of 10.6V, and really dead ones drop down immediately to almost nothing. Even works on lithium iron phosphate ones, though there you see good packs are pretty much flat, and drawing over 15A for a few seconds simply makes them shut off due to the overcurrent protection.

Stop Start lead acid batteries for cars, ie EFB & AGM technology, require special SMART chargers, not the older regular chargers. Buyer beware! Also LiPo batteries are not supposed to be drained past 80% of their capacity. I have many helicopter LiPos that have swollen up after charging by letting them drain too much.

Hi all. I appreciate your comments thus far. Keep them coming. I want to acknowledge what @SeanBZA and @Mervyn Haynes have said and elaborate a bit. Maybe more than a bit. Batteries can be extremely dangerous. I'm VERY careful and meticulous, and I've made some bad mistakes. Batteries can contain huge amounts of energy. I was once tightening the screws on a 100 AH LiFePO4 battery with a crescent wrench. The wrench was just about the length of the distance between the terminals. The wrench was attached to one terminal and barely grazed the other for just a brief fraction of a second. There was a huge spark which scared the heck out of me and said, better than any actor cursing on any movie, what the heck are you doing dummy? Both the battery and the wrench now have a gouge where they touched. If the wrench had welded itself to the terminal, my house would have probably burned down. I immediately wrapped that wrench completely in packing tape except for the adjustable end. Ugly, but functional. Your watch, or your ring can do the same thing. Bridge a ring across the terminals of a battery and you'll probably get second or third degree burns and maybe lose a finger. I once connected one of these small SLA batteries to the load bank backwards. BAD idea. The load bank wasn't active, but it shorted the internal FET array and effectively shorted the battery. Note that I was expecting NOTHING and was unaware of any problem when I connected the wire. Within 1 second, the small jumper lead I was using, which was good for maybe 1 amp max, had melted it's insulation and was glowing red hot. I was within seconds of a life changing catastrophic event. I slapped the jumper lead away from the battery to get it disconnected as quick as possible. There was no time to "grab" or "disconnect" it. I burned a small stripe across my fingers as a reminder, and was literally only in contact with the (now bare) wire for 1/4 second.

I am not proud of these events. I'm glad that I and my house survived them. I don't just grab a soda, a comic book, and casually start connecting things. I'm very careful. There's an old adage in the construction industry - measure twice (or thrice) and cut once. This concept applies here although not the cutting part. Pre consider and pre plan EVERYTHING. RED positive wire goes to RED positive battery terminal. Other end of RED positive wire goes to RED positive power supply or load bank terminal. Same for the BLACK negative wire. If the terminals on any item are not color coded be doubly, triply careful. Are the settings on the load bank or power supply correct for your SPECIFIC battery and SPECIFIC battery chemistry. I wouldn't test multiple different chemistries at the same time. Is the voltage right, the current (compared to the battery rating), the voltage limit, the current limit? Triple check everything before connecting wires or activating equipment. This is not a spectator sport.

As @Mervyn Haynes points out, other battery chemistries are MORE dangerous than the relatively simple SLA (sealed lead acid) batteries I've mentioned here. Car batteries can explode or non sealed ones can leak acid. A shorted car battery can supply 800 A or more. That's WELDING current. For exotic car batteries, use automated chargers and maintainers so you don't destroy a $ 500 battery, etc. All lithium batteries are very snarky about upper and lower battery voltages. If you overcharge or over drain a LiFePO4 battery, you'll kill it. These are more resistant to fire internally but they can still provide plenty of current to start a fire externally for the bigger ones. Many of these have battery management systems (BMS) to prevent you from killing the battery. Hobby style batteries for model cars, boats, planes, helicopters, and drones are usually LiPo, not LiFePO4. They are usually built for high surge current, light weight, and usually no or minimal safety circuits. These are NOTORIOUS for catching fire or exploding if they're ticked off. YouTube for lithium battery fire. Note that, for the LiPo batteries mentioned in my Palm PDA in the other thread, I was discussing draining the battery in the device till the device shuts off. Nothing more. I was not manually draining that battery on a load bank.

So, I wasn't by any means trying to ignore the safety issues. If you're going to manually work with, test, charge, and discharge batteries: Learn the attributes of the chemistry and its limits. Learn the limits of your particular cell or pack. Plan exactly what you're going to do. Arrange everything carefully. Set everything carefully. Connect everything carefully. Activate everything carefully. Be conscientious and meticulous (that means slow and boring). Especially for LiPo, try to keep them away from flammables. One reason I stopped working with drones, other than FAA meddling, is that there is no place in my house that's not flammable. If a cell or pack ever starts swelling or overheating, discontinue immediately. If you can, get it into a flame proof bag or bucket of sand. If it explodes, flames can reach out several feet. LiPo battery fires are SELF FUELING and cannot be extinguished with water. If you can, figure out how to safely discharge it the rest of the way all the way to zero in a place where you don't care if it catches fire. Take it to a professional company like Batteries Plus and have it recycled. Note that it is illegal and dangerous to ship damaged batteries by USPS or a carrier like UPS or FedEx. A battery fire in a truck or plane could get people killed. It's not worth it just to make a warranty claim with a vendor. The vendor may not know that you can't do this.

That turned out to be a bit longer than I anticipated. But, this is critically important so I might as well get it out in the open. You may only get one chance to make a mistake. Stay safe. Other comments are welcome.

May your bits be stable and your interfaces be fast. Ron

My rule of thumb for SLA batteries is that they typically only produce about 1/2 of their 'rated' AH capacity, before they hit that magic 10.5V limit. Any discharge past that voltage will likely permanently harm the battery. Also, they are HEAVY for their capacity. As a ham radio operator that likes to do "Parks on the Air" (POTA) activations, a SLA is just too heavy and has such a poor capacity rating that I tend to leave them at home these days, and use instead good brand name (BioEnno) LiFePO4 batteries. They also have a slightly higher terminal voltage, which works better with the 'nominal' 12VDC radios, which actually are usually much happier at 13.2VDC, especially when transmitting. And the LiFePO4's have a very flat discharge curve, even under load, which again makes my radios much happier when transmitting. I typically get 90-95% of the rated capacity out of my BIoEnno's, and they come with a dedicated smart charger, so I don't have to worry about having a bench full of test equipment to monitor them. LiFePO4's are also much more stable and safe than LiPO's. I have a variety of BiEnno LiFePO4's - 3AH, 6AH, and 12AH. I rarely use anything larger than the 6AH with a 10W PEP SSB (I run it 5W CW or digital modes) Elecraft KX2 transceiver, and it will run for hours on a 50% receive/transmit duty cycle.

@Dave New You make some good points. I just happened to be working with some SLA batteries so I decided to share this information. I'm not that fond of SLA's per se. For lots of small electronics and not so small backup power banks and boat motors and electronics, etc., the LiFePO4 is a really attractive option. I've got a couple of 100 AH Ampere Time (they recently changed their name to something else) LiFePO4 batteries ganged together in parallel with BIG cables for backup power. I can run a mini freezer, computer, and a lamp for a day or so. Longer power outages than that get really complicated. I've got a Victron Smart Shunt to monitor the parameters and power consumption. For those that don't know, a shunt is a very precise high current resistor element (looks like a metallic bar) placed in series with the main power source. It produces a very specific voltage proportional to current, which allows energy usage to be measured. As you said, I can get 100% of the rated energy out of the battery, which is really cool. LiFePO4 batteries usually have a BMS to keep one from over discharging, over charging, over current, shorting, charging while freezing, etc. You're right about weight of SLA, they are HEAVY for higher capacities. When someone uses the phrase heavy as lead, they mean it. The capacity thing is kind of a scam in the SLA industry. They say don't drain an SLA past 50 % to avoid shortening its life a lot. I don't believe if you drain an SLA flat to 10.5 V and immediately recharge it that it hurts it that much. But, I do believe that letting it sit in a discharged state even for a short time does hurt it. That's probably why they give the 50 % recommendation. So, if you stick to that logic, then yes, you only get half the energy out. Technically, though, I'm able to get 100 % of name plate rating from an SLA that's good and new or one I've successfully refurbished. Lead acid can be more desirable for engine starting with large current levels or in automotive applications where you don't have a custom LiFePO4 charger. But, LiFePO4 is definitely one of the best alternatives out there. LiFePO4 is much safer than LiPo in terms of fire danger.

May your bits be stable and your interfaces be fast. Ron

SLA will lose capacity with each cycle, and also with time, so pretty much will be dead after a few years, even if not actually having anything like a full discharge, and only having perhaps a 5% discharge before recharge. The ones used for old mining lamps are 2 cell packs, with a pouch surrounding each plate, and are wet, and those only have a 3 year life, before they are removed from service and either repacked or replaced. when new capacity is such they can do 48 hours of work illuminating the bulb, but are invariably pulled from service at 3 years, where they often are turned in for replacement, because they will not actually last a shift, even though the miner turns off the light for the 2 hour each way commute to the active face, and is working 8 hours at the face. Heavy duty plates, thick interconnects and well built, and they still do not last. The new lamps, with LED lamps replacing the incandescent ones, lasted 5 years before failing, and the miners welcomed the new lithium packs, half the volume and double the life, that have almost totally replaced the lead acid packs.

The mine packs are charged on a rack for 12 hours between shifts as well, about the best charge possible for SLA, but still drop off in capacity. Same for the electric batteries used for traction, though those do have a 6 month cycle through a battery shop, where they get the acid drained and filtered, the SG verified, and also get a full discharge and charge cycle per cell, used to get an exact capacity, to make the packs all from close matched cells. Charging on those is deliberately to the point of dissociation of water, with a fill system to make up, so they do not sulphate easily, and service will pull any weak ones out and replace them. Plus charged immediately at end of shift, so never left low. They still fail around 5 years and go for recycling, or are put into use for static power supply in larger banks.

A most interesting post. Thank you.

Would you happen to know how an old sealed car battery and a small sealed battery (used in UPS and golf carts) would react to a permanently connected 13.5v (5A max) supply? I’ve done this for years as I need the battery as a backup for my electronic door lock should their be a power failure and to use the battery to power some 12v bulbs when the powers out.
Time passes so quickly doesn’t it, I‘ve just realised that these batteries are about 8 years on from the time they were taken out of service - so they are getting on a bit.

@squirrel Thanks. I'm not a battery expert, just a person who's watched a number of videos, done lots of reading, and done a good bit of tinkering. So, I cannot give hard fast advice but I can share some general thoughts. Anything I say assumes the batteries in question are not fundamentally shorted or internally damaged. It assumes the electrolyte hasn't evaporated, etc.

My understanding is that SLA (sealed lead acid) batteries don't like sitting around in a discharged state. On the little 12 V 5 AH trailer brake battery I have, it says on the label - charge 14.5 V - 14.9 V and it reduces if it get's hot. And, float 13.6 V - 13.8 V and it reduces if it gets hot. It also says max charge 1.5 A, which is about .3 C. So, assuming this battery is pretty representative of this type of battery, that's where I chose my numbers of 13.7 V or 14.7 V. The lower one, the float voltage, should be safe for the battery over a long time. I have similarly had some of my little batteries attached to a power supply continuously. They're not critical for me, and it's more of a project and a vendetta against the ones that are under performing. In testing, I discovered that some of them had only about half the rated capacity. Hence, this thread. I cannot say if they never had their rated capacity, a distinct possibility, or if they degraded even though I had them on a power supply. I have been able to rejuvenate them to near rated capacity.

I can tell you that the Battery Tender device that I have keeps the batteries somewhere around 13.8 V in lead acid maintenance mode. So, I don't think there would be a problem leaving them on a power supply as you suggest. But, that doesn't necessarily guarantee that they can do what they're rated for. Based on my research, there doesn't seem to be any cheap easy way to test battery capacity. You can buy a battery tester from Amazon or the auto parts store for $ 30 - $ 50. These measure the internal resistance of the battery and the instantaneous cranking amps. But, they don't measure the actual amp hour capacity. As far as I know, the only way to measure amp hour capacity is to drain the battery while watching the output and tabulating what you're getting. My load bank does this automatically. You could do it manually if you were sufficiently motivated and willing to spend the time using a process called numerical integration. I can explain that if needed.

So, I think the only way to know what your batteries can actually do is to test them. 8 years is a bit old, but they might still work for your purpose. You could, if you wish, attach one of your 12 V lights to the battery and just time how long it runs. If you know the amperage requirement for the light, you could calculate how long the light should run. To maximize battery life, it's recommended not to pull more than 50 % of the energy out, although I'm doing more than that for testing. Don't run the battery below 10.5 V. Then, recharge it immediately. If you use the power supply for charging, set the current limit to about .3 C, IE the battery's AH rating x .3. If you cannot limit the current, depending on the size of the battery, the power supply may put in too much current. Something like a Battery Tender, or Battery Tender Jr. is a possible option for charging small batteries. My Battery Tender puts out .8 A. You can charge the battery until the current drops below .01 C (or 1 %) or even .005 C (or .5 %). For my 5 AH (same as 5000 mAH) battery, as an example, 1 % C = 50 mA and .5 % C = 25 mA. At that point, the battery is pretty fully charged.

If you have the motive and the patience, you can repeatedly charge and drain the batteries and see if your lights run longer. Do not leave discharged. (Side note, one YouTube video said if a battery is heavily sulfated, it might have to sit on a power supply at 14.7 V for a month or more before it starts accepting current. Theoretically, your batteries should be charged already.) You can continue this until you run out of patience and give up if the batteries are not working well enough, or you are satisfied with their performance. Remember that if you're pulling current out at faster than the 20 hour rate (.05 C), you may get less than rated capacity even if the battery is working. If you decommission them, consider recycling them at Batteries Plus or similar. If you replace them, you may wish to consider 12 V LiFePO4 replacements. These are much lighter, give 100 % of their rated capacity, and do not degrade as quickly. They must be charged with their own special dedicated chargers though. The charger current could be proportional to the C rating of the battery according to the recommended charge rate. They should have a BMS (battery management system) to protect the cells. They can be drained for testing similarly to SLA, but the cutoff voltages and drain currents are different. LiFePO4 batteries are not generally floated (maintenance voltage), so you have to top them off every couple of months. Having said that, my Battery Tender does have a Lithium maintenance mode that can keep these topped off, but I don't know exactly how it works.

That turned out to be longer than I thought, but I tried to include hopefully useful info. If you have more questions, feel free to ask. Maybe others with battery experience can jump in too.

May your bits be stable and your interfaces be fast. Ron

If you haven't already, take a look at BatteryUniversity^com -lots of info on many type batteries.

I haven't played with small lead acid batteries much, but have kept a couple deep cycles 'ready to go' in case of power failure. Unless there is something I haven't heard lead acid should never be kept on a trickle charge, except maybe at low rates to try to desulfate for a limited time. Float voltage is critical for life, too high will 'boil' off the electrolyte, too low will reduce capacity and increase sulfate. Depending on the exact type battery the float voltage will vary, so following the manufacturer's ratings is a good idea. As mentioned try not discharging below 50% capacity.

I use a CBA (computerized battery analyzer) to check all my batteries, NMH, Pb, Li and I have come to trust it. WestMountainRadio^com sells them if you want to take a look. It is a bit expensive, about $200 last I looked, but especially for batteries in critical uses it is probably worth it. You set battery chemistry, capacity, and a discharge rate and it will plot a graph of the test results which you can later compare against other tests and see how the battery is aging. The software auto sets a safe cutoff voltage, but it can also be changed if needed. It can be set from 1 cell to multiples if testing a battery pack, and has a pass/ fail, usually at 80% rated capacity unless you change it. I like it, and have just found 5 NiMH AA cells that need to go to the recycle bin. Here's one type of graph from my 2 deep cycles which total about 100AH- if I attached it correctly!

Depending on where the battery is kept it's temperature can be a big factor in it's useful capacity. Despite a number of drawbacks, NiCds held up good at low temps and did not mind being kept on a 0.1C or less continuous trickle charge. They also did not mind high current loads, they kind of liked the abuse to minimize 'memory. Back in their day I had a handful of wet cells but don't know how available they are now. At the time I thought of putting 11 or 12 in series as a car battery but the cost was quite high. I still have a copy of the General Electric Nickel Cadmium Engineering Handbook somewhere.

If you can find the battery manufacturer's date code decipher you can check how old a battery is before buying it. It's unlikely a supplier would bother doing periodic charges on small batteries so their age may be a useful. At work the batteries had dates and we would return them after a certain period of time, used or not. Another thought might be to keep a second battery in case the one in use gave out. With some more expense a solar battery charge regulator could keep a second battery charged whenever the engine is running and still isolate it against unintentional discharge. There are options depending on their cost vs need.

Ralph said:

I still have a copy of the General Electric Nickel Cadmium Engineering Handbook somewhere.

Click to expand...

As do I. I got mine back in the early 80's when I was working at IBM Lexington, where we were pumping out a million typewriters a year. The new Models 50, 60, and 75 Selectrics had internal phrase memory that would go away on a power outage, so I was on a team that was designing a NiCd battery backup.

I learned to type in high school way back in 1982. We had the old manual Remmington typewriters and you had to aggressively press each key about .5" to hit a key. We had one IBM Selectric in the room. I got to use it occasionally. That was the best feeling keyboard I've ever touched. The younglings have NO IDEA what they've missed. If you are reading this and you don't know what we're talking about, do a YouTube search on "typewriter". If you're reading this and don't know how to type, even as an adult, consider loading some typing tutor software on your laptop and learning. It takes lots of practice which is boring, but it's a really handy skill to have.

May your bits be stable and your interfaces be fast. Ron

Yes, I took typing in 7th grade in 1967 and like you, learned on manual 'mills'. I had also been taking piano lessons since I was six, and the typing teacher scolded me for 'bouncing my wrists' apparently due to my piano training. I will say that having touch typing under my belt did me all kinds of good when it came to typing up school papers on my dad's portable electric (which is now stored in my closet - haven't thought about it in decades, and it may not work any more from disuse). I started working with computers in high school using a ASR-33Teletype(tm) and a dial-up modem to dial in to the local college GE-255 time-sharing machine, which ran Dartmouth BASIC, ALGOL, and FORTRAN. I still have punched-paper tapes of the programs I wrote back then. I then got interested in ham radio, and got my Novice and Advanced tickets before going to college, where I majored in Electrical Engineering with a minor in Computer Science. That combination of computer skills and RF (plus knowing which end of a hot soldering to pick up) guided me through a long career of embedded systems programming for a variety of start-ups and finally as a Senior Technical Specialist in RF and Cybersecurity for a major automotive OEM.

Dave New said:

using a ASR-33Teletype(tm)

Click to expand...

That was a finger strengthening tool.

Although a number of co-workers would rather cut and paste than type, I don't mind the typing. Like many I started on the old mechanical ones. When I started working with computers, the old B3500s used Teletypes as their system console- I don't recall the model. I never took a course on typing, and although I do use 2 hands I am sure I am not doing it 'the right way'.

It's interesting to see some of the really old typewriters in some movies. One I can recall had a sticking key and the person typing (Harold Lloyd) poured some kerosene from the lantern into the typewriter to get it going. I don't know if people actually did that, but we did spray oil into a running Teletype when needed. When I was shown how to do it they thankfully told me to duck so oil from the spinning fan wouldn't get all over me.

My sister used to work in a court, and stayed with a Selectric for many years as typing method, simply because the court computers, old 286 and 8080 XT and AT machines, with MSDOS 4.0, and Wordperfect, could not keep up with a typist capable of doing 30WPM plus consistently. She got annoyed with that 16 character buffer filling up and beeping after dropping characters, or whole words. The Selectric could keep up, and thus she stayed with it till the courts finally got 386 machines that were finally capable of keeping that buffer empty.

As she still types a lot, she uses now Win10, and Office 2008, and at least it can still handle the characters, even if the screen can occasionally take a second or two to show an entire line of input. Anybody looking at her keyboard though will find it hard to use, as all the centre keys, and the numeric pad, are all plain black ,with all the characters that were printed on them missing. Standard cheap Microsoft wired keyboard, she used to use older white MS keyboards, including one of the original split hump ergonomic ones, which she literally used to the point where the keys broke off the mounts, as the centre had worn away. next suffered the same fate, it probably had easily 10 billion keypresses over it's life. She worked a side job as well, as a transcriber for court documents, though she does not want to do that any more, the rate per page is low, and the audio from the courts can at best be described as poor on the best of them, often with whole pages of "inaud" for responses with only the general room mic getting anything, and for a lot of the rural courts also often having extraneous noise like roosters, crickets, frogs and goats as side tracks, along with dogs, cats and cows.

SeanBZA said:

all the centre keys, and the numeric pad, are all plain black

Click to expand...

Stickers can be used to restore the missing labels. I've had to buy these on a few occasions. I'm typing this on my old Windows 7 laptop from 2010. I've had to go through a couple of rounds of stickers.


May your bits be stable and your interfaces be fast. Ron

I've never typed enough on a keyboard to wear out the markings. If I did it is quite possible I would be looking for a new keyboard since I don't think I could type without looking at the keys.

At least for the Selectrics, the key caps were double-shot injected molded (the white markings went all the way through the key cap, so you could never wear them off, but the electronic Models 50, 60, and 75, which had electronic keyboards used a 'Vacuform' like machine to 'inject' a dye 5 mm into the surface of the key caps. This was favored because you didn't have to keep a plant inventory of literally hundreds of different double-shot keycaps for all the 75-plus different regional keyboard layouts. Instead for the 'Vacuform' method, you printed the keyboard layout using dye in reverse on the rubber membrane, and then pulled it down over the blank keycaps on the keyboard assembly, heating it with an infrared lamp. The dye would then transfer into the keycaps. The only drawback was that you couldn't have white legends, since there is no such thing as white dye (duh!).

The same process was used for the IBM PC keyboards. They copied the design invented by IBM Lexington for their electronic typewriters. That's why the IBM PCs had beige keys with black legends, instead of black keys with white legends.

We spent a LOT of time testing various key and legend combinations, in a testing room with high-speed typewriter operators (30+ cps) and measured the number of mistakes per page of copy.

Also, the Selectrics and electronic typewriters were tested to 17.5M characters, which was considered 'life' for those machines. This was done using 'overhead robots' which was a box with solenoid plungers for each key position, driven by 8085 chips to exercise the typewriter. They used paper rolls similar to teletype paper rolls, and each roll was examined by hand for quality of print, etc.

Hi all. Just wanted to give an update on the battery stuff. I'm still experimenting with the little 2.9 AH SLA batteries I have, trying to bring them back up to operational status. I reiterate that these are not critical to me, not super expensive, and not originally working well at all. Batteries that have less than 80% capacity are usually considered to be in need of decommissioning. So, if I trash something that would normally be decommissioned, that's not much of a loss. So, I've been experimenting with flattening the batteries. Not totally flat. But, below the standard limits. The normal cutoff voltage would be 10.5 V (for a 12 V battery). But, I'm experimenting with flattening them down to 9 V. All the experts say doing this reduces the life of the battery. But, if the unit is already unusable, whatever lifespan I get out of it is better than it was. For those that haven't read the whole thread, what largely started this is that I bought a small cargo trailer with a little 12 V 5 AH brake battery that turned out to have only about 1/2 of its rated capacity. Obviously, I'd like the brakes to work if needed. The problem is that, if I buy another battery, even new, it could suffer the same problem depending on when it was manufactured and how long it's been on a shelf. Lead acid batteries hate that. As I mentioned in an earlier post, stores like Batteries Plus don't like the idea of taking a warranty return of a battery when their volt meter says it's OK. They don't have a way of testing capacity. By the way, they also won't warrant something below 10 V when it comes in the door. I think I've got the trailer battery back over 90 % capacity and may work on it further. But, I'm taking greater risks with the 4 small 2.9 AH batteries mainly as a learning experience. Batteries are weird things. Both the low cutoff voltage and the output capacity vary with load. I'm not convinced that it's draining to a low voltage is what kills the battery. It may be that LEAVING it at the low voltage is what kills it. Note that I'm talking only about lead acid batteries. There really are dangerous changes in lithium batteries that happen if you over discharge or overcharge them. For your entertainment, here are a couple of new links I found.


And from within that, a link to a Power Sonic technical manual that has lots of good info. I'll share more as I can.


May your bits be stable and your interfaces be fast. Ron

Hi all. Got a few more thoughts about batteries. Hopefully also a little levity for your amusement. Any reported conversations with batteries are totally fictional. I've been having some success rejuvenating the SLA batteries. I really had few good choices but to try to restore the trailer brake battery (5 AH). It needs to remain lead acid since the car's charging system is not optimized for LiFePO4. My only alternative to restoration would be to seek a quality battery vendor and get PROOF that the replacement SLA was manufactured within the last six months AND test it. I've restored the trailer battery to 95 % capacity, although I haven't tested it further in the last couple of weeks. It's been on float charge on the Battery Tender. I've been having some success restoring the other 2.9 AH batteries as discussed. Indeed, all were under 80 % capacity, sometimes well below that. I've gotten some now over 90 %.

The problem is that this experiment is quite time consuming. Much of it is automated. But, switching a battery from power supply to load bank, checking and setting and changing settings, and recording results is not automated. I can tell you that watching the status meter on a power supply or load bank connected to a battery is ... incredibly insanely boring. But, necessary, to a point. For example, previously I mentioned charging an SLA battery at 13.7 V or 14.7 V, initially current limited to .3 C, to where the current drops below 1 % of C. Upon further reading, I find that this might not be adequate. It turns out final charge may take 24 hours, not the few hours you'd expect initially charging at .3 C. Final current might need to be .2 % of C or even .1 % of C. As a real world example, one of these little batteries has a 2.9 AH rating. Let's pretend it's 3 AH to make the math simple. This is 3000 mAH. 1 % of C is 30 mA. But, the battery I'm currently watching charging on the power supply is down to 6 mA. If I had yanked it at 30 mA, it would not have been fully charged. Also, flattening an SLA battery to 9 V (never do to a good battery, see prior caveats) is hard. I initially drain the battery to 10.5 V at the 20 hour rate. This takes up to 20 hours. Less if the battery is weak. But, the moment the load bank cuts off, the battery voltage starts rising again, sometimes substantially. It's a sneaky little monster.

A small crash course in SLA chemistry (what little I know). In the new state, an SLA battery has plates made of lead as you might expect, and lead dioxide. The electrolyte is sulfuric acid. I read that it can take 200 cycles to truly get the plates and electrolyte conditioned and working well. I think this is called formatting. Wonder if they do THAT in the factory. I doubt it. When you discharge the beast, strange and wonderful chemical magic happens inside. Electrons go through your circuit, and you end up with both plates largely made of lead sulfate. The electrolyte is mostly water at that point. Charge the battery and the process reverses. Pretty amazing really. If you recharge it soon enough, as in immediately, the lead sulfate doesn't crystallize and kill the battery.

So, back to flattening it. Initially, the load bank is set to .05 C and 10.5 V cutoff. As soon as the load bank cuts off, the battery voltage starts rising again and the battery says in a Darth Vader voice "Ah Ha, fooled you. I've got more electrons. Come get them." So, thinking to outsmart it, I switch the load bank to resistance mode, say 100 ohms with a 9 V cutoff, which pulls about .1 A and turn it back on. I sneer at the battery and yell "So There!". Well, I called it a sneaky monster for a reason. The load bank runs another hour, maybe two, and shuts off at 9 V. I don't care how much energy is coming out. I just want the battery fully drained. I turn the load bank back on. It runs maybe another hour. I hear the battery snickering and quietly saying, "You want to drain me, then I'm going to torture you too!" Now, obviously, I'm doing other things while I wait, but any time your brain goes through a "context switch" from task to task, you lose efficiency. Eventually, I get to the point where the load bank shuts off in a few minutes, I raise the resistance to 200 ohms, then 300, then 500, then 1000. All the while, having to babysit the experiment. At 1000 ohms, it may only be drawing 10 mA and may run for another hour or two. After I get the battery REALLY down around 9 V, I'll charge it again, which could take another 24 hours, than drain it again at the 20 hour rate and see what the capacity is. If I try to rush the process, I could make the electrolyte gas out and destroy the battery. This is definitely a test of patience.

Now, I'm definitely learning stuff. And I enjoy learning and sharing it with you. But, the value of these batteries is starting to be less than the pain of saving them and learning about them. Next, I will describe the pros and cons of using 4 different kinds of batteries and why I may just recycle these and get a 12V 12 AH LiFePO4. Haven't decided yet. But, $ 70 for a replacement is not trivial to me, so it's not necessarily a slam dunk.

This link is from another serious amateur (prosumer) like me who's been doing similar research.


Hope this helps.

May your bits be stable and your interfaces be fast. Ron

Incidentally a fully flat battery can be charged in reverse, swapping the polarity on the plates, with it seeming to still accept full charge, irrespective of the fact it has been reversed.

It took me some searching to find "Peukert’s formula" without remembering how to spell it. Basically you run a discharge at two different current rates and record the time to cutoff. The values are plugged into the formula found in the link below (replace the * in the link with a period).

https://www.victronenergy*com/media/pg/SmartShunt/en/battery-capacity-and-peukert-exponent.html

There's a link on that page to a calculator you can download to save the math and log calculations.

From that you can use the calculated constant to determine your runtime at any constant discharge rate. I've used this in the past with known good batteries and used the manufacturer's rating, usually 20 hours on larger batteries as one of the discharge values. This may be of use since the closer the constant comes to 1.0 the 'better' the battery. A high value is not what you are looking for, and you will never get a value of 1.0 since even a good battery has losses from internal resistance, operating temperature, etc.

All of the above may not save any time, but if you have test results recorded you may have all the data needed to run the formula already. I've used that constant in a couple of marine/ RV battery meters that give run times at whatever load you are applying at the time at it seemed to work well. If you have a known good battery you can run the 2 discharge tests and come up with a constant for a good battery of your type to use as a rough guide evaluating the others.

When I've run evaluation tests in the past, pre computerized battery analyzer, I would use an inverter and 250 watt bulb as my load which conveniently worked out to an approximate 4 hour run time. I would plug an analog clock into the inverter set at 12:00 and whatever 'time' the clock had when the inverter cut off was my run time. That way I didn't have to keep waiting and checking for an accurate number. In my case I knew to expect about 4 hours to cutoff and just set a reminder for myself to check things so as not to let the battery set discharged for too long.

For some reason this video showed up in my YouTube recommended list. It made me think of this thread.
Here's a backyard (or sidewalk) project.

Tazz said:

YouTube

Click to expand...

Wow dude. That is certainly ONE way to do things. Interesting. A little TOO hands on for my taste. But, it's kind of fascinating that the components shown can make a battery at all.

May your bits be stable and your interfaces be fast. Ron

I don't think I would try what was in the video. If memory serves, in a flooded battery the plates gradually shed lead and there is a space at the bottom of the case under the plates so the pieces can collect without shorting out the cell(s). Disassembling a battery like that was/ is only done on very large cells. Unless there was a very good reason I would go with a new battery before even thinking of taking one apart. Anything that has collected at the bottom of a flooded battery is lost capacity since it is no longer on a plate.

Judging from the background in the video I would say there is no Costco or Sams Club nearby to go to and easily buy a new battery. If that is the case I guess such extreme methods may be justified.

Well, seeing as it is Pakistan, that is where all the old batteries you send in for recycling land up, being stripped down in a country with a lot less regulation about dumping lead in the environment than the USA does. Ship it off and pretend it will be done to EPA standards, just like the US shipping lines do for old ships with asbestos and PCB loaded to the brim.

Yes, ship it away, and "nothing to see here".

Hi all. I just wanted to give an update on my progress here. I am finished trying to rejuvenate the lead acid batteries I was working on. I had 4 ea 2.9 AH SLA batteries and 1 ea 5 AH SLA. All were showing from 50% to 80% capacity. Although I couldn't find a reference, I've heard or read that a battery is considered at end of life it it's below 80%. I really prefer mine to be over 90%, and preferably 95% up to and above the name plate rating. I've been able to rejuvenate the 5 AH battery up to its name plate rating. I've also been able to rejuvenate 3 of the 2.9 AH batteries to above 95%. I'm decommissioning the other 2.9 AH battery as it never got above 88%. Disclaimer, I've been treating these pretty hard with tactics that could kill a new battery. But, these were going to be retired anyway. So, I'm glad I saved 4 of the 5.

I got the basics for my technique from the labeling on the 5 AH battery. Charge and discharge current is about .3 C, IE .3 x 5 AH or .3 x 2.9 AH. So, about 1.5 A or 1.0 A. Charge voltage is 14.7 V but should not be left on them. Float voltage is 13.7 V. Essentially, I've just been repeatedly charging and discharging them since I started the thread. This is NOT time efficient. In fact, it's very time consuming even though it's semi automated. It was more of a can this be done experiment. As I said, it worked for 4 of 5 batteries. Also, I would not have been willing to spend $ 30 each to replace all of them. Other than the 5 AH battery for my trailer brakes, I just use the others for hobby stuff. I now have the remaining 4 on a battery tender for maintenance. It keeps them at about 13.7 V.

During the rejuvenation I charge them individually as follows. Power supply set to 14.7 V with a current limit of 1.0 A or 1.5 A as appropriate. I stop the charge when the current is .005 C or 1/2% of C. So, for the 5 AH (5000 mAH) battery, that's 25 mA or less. For the 2.9 AH (round to 3 AH) (3000 mAH), that's 15 mA or less.

The discharge test is what I'm doing more aggressively than I would to a new battery. The SLA batteries of this type are typically rated at a 20 HR drain with current at .05 C. So, for the 5 AH (5000 mAH) battery, drain current is 250 mA. For the 3 AH (3000 mAH) battery, drain current is 150 mA. Even with a good battery, if drain current is higher, you will get less capacity out of it. A normal test drains the battery at .05 C down to 10.5 V. This is what I start with. If the battery capacity is below around 95% (for my purposes) at the end of the first test, I switch the load bank to CC + CV (constant current + constant voltage) mode and continue the test. In this case, I set the current to be the same 250 mA or 150 mA and set the voltage limit to 7.8 V. Note battery suppliers will often not honor the warranty if the battery comes in the door under 10.0 V. The load bank will consume the stated current until the voltage drops to 7.8 V. It will then allow the current to decrease as long as the test runs. My load bank doesn't tabulate amp hours in this mode. Just as in charging the battery, I like to let the current drop to about 1/2% of C. That would be 25 mA or 15 mA as stated above. The reason I discharge the battery so harshly is in the hope of dissolving all the lead sulfate crystals during the charging process. The battery should never be left dead like this. Once it reaches the low current level, I immediately put it on the power supply to charge again. As I understand it, discharged means weak watery sulfuric acid electrolyte and lead sulfate on the plates (not necessarily crystals). I believe that charged means lead dioxide on the plates and strong sulfuric acid electrolyte.

By doing this repeated charge and harsh discharge 20-40 times, I've rejuvenated 4 out of 5 of my batteries.

This process is unique to lead acid batteries. Others need different procedures. For Nickel Cadmium batteries, I (personally) would look into replacing with Nickel Metal Hydride, which are less finicky about capacity loss. They can be cycled (drained and recharged) to try to rejuvenate them but the voltages are different. Charging should be done with a purpose specific NiMH charger. For Lithium Polymer (LiPo) and Lithium Ion (LiIon) batteries in consumer equipment, it can be helpful to drain the battery until the equipment goes off and then recharge from time to time to reset the battery level gauge. They should be charged with their purpose specific charger. Lithium Iron Phosphate (LiFePO4) batteries should have a battery management system which dictates how much they can be discharged and then cuts off. This may NOT happen though, so don't take a 12V LiFePO4 battery below about 10.5V. They should be charged with a purpose specific charger. Raw LiPo batteries (in RC cars, etc.) can be drained with purpose specific devices and should not be over drained. They should be charged with purpose specific chargers.

Hopefully, this info is helpful to people for maintaining various batteries.

May your bits be stable and your interfaces be fast. Ron

12.7 volts seems about right for a float voltage, I use that on my 2 deep cycles and they are still good about 10 years later- with some but not a lot of capacity loss. Float voltage seems finicky, I've ruined a few batteries floating them under 12.7

It's probably best to follow the manufacturer's spec if known since the specific gravity of the acid will change the battery's voltage some. From playing with car batteries when they were all flooded, a long slow charge at a low current ignoring the voltage would eventually de-sulphate most of them- but you could add water to those and get away with that method. I do recall some sealed batteries used some kind of catalyst that would convert the off-gases back to water, but it's been too long for me to remember details. If a battery did have that feature I would imagine the manufacturer would be sure to point it out.

I have a 40amp 'smart' 3 stage charger that has a built in de-sulphate cycle. Just from watching the volt meter it appears to charge up to a certain voltage, stop charge, wait for the voltage to drop to a certain level, and keep repeating the process. The charger was made for SLAs.

I was in ChatGPT and thought it might be interesting to hit it with the following question.


As it applies to small (under 10 amp hour) lead acid SLA type batteries, what methods can be used to recover lost capacity? Include sulphation in your reply.

ChatGPT

Sealed Lead Acid (SLA) batteries, also known as Valve-Regulated Lead-Acid (VRLA) batteries, can lose capacity over time due to a number of factors, including sulfation, corrosion, and general wear and tear. Sulfation is the most common cause of lead-acid battery failure. It occurs when a battery is not fully charged and the lead sulfate that forms on the plates during discharge is not completely reconverted into lead, lead dioxide, and sulfuric acid.

Here are some methods that might help to recover lost capacity:

1. Desulfation: This is the process of removing the sulfate deposits that build up on the lead plates. Various techniques can be used to desulfate a battery, including:
   • Pulse Conditioning: This involves sending pulses of energy into the battery at varying frequencies and voltages. The goal is to break down the lead sulfate crystals that have formed on the battery plates.
   • Chemical Additives: Certain chemicals can be used to dissolve the sulfate crystals and can sometimes restore a portion of the battery's capacity. It's important to remember that chemical additives should be used carefully and in accordance with safety guidelines due to their potential risks.
   • Equalization Charging: This involves charging the battery at a higher than normal voltage to try and reverse sulfation. However, this method must be performed with caution to prevent overcharging, which can damage the battery and may pose safety risks.
2. Regular Maintenance and Proper Charging: Regular charging and maintenance are essential to keep a lead-acid battery in good working condition. Keep the battery fully charged whenever possible, and avoid deep discharges to minimize the risk of sulfation.
3. Battery Reconditioning: This is a more extreme measure that involves completely discharging the battery, then slowly recharging it to its full capacity. It's a risky process and not always successful, but it can sometimes help to restore a portion of the battery's lost capacity.

Please note that not all these methods are suitable for sealed batteries because they can't be opened for servicing like their flooded counterparts. Also, some of these methods might not be successful depending on the state of the battery and the extent of the damage. Always prioritize safety when attempting any kind of battery maintenance or reconditioning.

ChatGPT may produce inaccurate information about people, places, or facts. ChatGPT May 24 Version
ChatGPT


Aside from step 3 which I was unaware of, you probably tried most of these.

Ralph said:

what methods can be used to recover lost capacity? Include sulphation in your reply.

Click to expand...

Here are some opinions based on my research. They're still just opinions though. That's actually not a bad chat reply. A bit general, but not bad. As they say, the devil is in the details. There's lots of controversy on pulse charging. Some of the people that seem more credible on YouTube are skeptical of it. Some people, though, report good results with the little pulse gadgets you attach to batteries. I am also skeptical, but haven't tried it. I would at least consider it plausible. As Chat GPT itself said, you cannot add chemicals to a sealed battery. I also haven't tried that. Equalization charging can evaporate electrolyte. While sealed SLA batteries do have a pressure vent for safety, once electrolyte is evaporated, it's gone for good. My method was basically a variation on the battery reconditioning process, continually discharging and recharging. I've been studying AI on YouTube extensively for possibly making YouTube videos. There are pros and cons, but, as stated on the AI thread I posted here on the forum, I believe more and more that researchers and creatives need to learn AI or get left in the dust. The TRICK is, how do you tell if the AI is hallucinating or giving just bad or dangerous advice?

May your bits be stable and your interfaces be fast. Ron

Certainly not all, but a lot of the questions I pose to AI I know at least a little bit about. If AI's answers seem reasonable I generally assume them to be correct. While not a perfect technique it's not much different than watching a YouTube video. Also with AI I find it useful to read it's replies and if needed ask followup questions, or rephrase the original. It can contradict itself- and has.

As far as batteries, as with everything else there is also some conflicting information out there. Having played with a number of car and deep cycles in the past the long slow charge has had some good results. In some cases after that leaving the battery open circuit the voltage would drop enough that I knew the battery was bad.

Maybe a totally different approach would be worth looking into. I just bought a Lithium UPS off Amazon. Although my previous 2 (different manufacturer) failed too fast this one caught my eye since it had 5 and 12 volt outputs and was intended for an alarm system, hence built to be plugged into a charger 7x24. The 12 volt output is 3 amps which may not be enough for what you need. It's rated at 7800mAH. Time will tell how long it will last. Perhaps a lithium replacement of the lead acid would be worth looking into. At the capacities you are using a lithium bank is pocket size, they hold charge quite well, and are extremely efficient as far as power in vs power out (Q). You wouldn't need a UPS like mine which would make one even cheaper, and a 5 volt USB wall charger will keep it running.

I understand the interest in lead acid, I've done similar things myself and batteries fall into what I call 'toys' in that I still like playing with them. But if there is a real need or safety is involved a 12 volt lithium bank may provide a safe alternative or backup until the lead acid batteries get sorted out. In the old days I would not have hesitated to suggest NiCds, they have the 'memory' issue depending how they are used but they really love to be abused aside from leaving them discharged. They are very good in cold weather, can supply a lot of current for their size, and a constant 0.1C or less charge can be kept on them without worry. I don't think they are very available now days aside from AA and AAAs.

It has crossed my mind to build my own lithium bank, but I find it easier just to buy one. However if you don't mind some tinkering, Li ion 26650s or 21700s could be made into a 12 volt bank. Both are about 5 amp hours but 2 in parallel will give you 10. You can get protected cells for safety and as long as they are the same types there should be no issues- except maybe charging them. There are charing circuit boards out there but I haven't looked into them.

Ralph said:

Lithium UPS off Amazon

Click to expand...

@Ralph All the info you're citing is good stuff and well worth considering. All modern USB battery banks use lithium batteries. It's logical that the UPS makers would go in that direction. Just make sure it has the right safety systems. I have 2 ea 100 AH LiFePO4 (Lithium Iron Phosphate) batteries ganged together with an inverter attached. They lose about 5 % of their energy per month and so they will drain down over time. Also, any monitoring circuitry will create a small parasitic drain. So, you definitely want to top off the batteries periodically. I try to top off all my emergency equipment, the big batteries, battery banks, radios, lanterns, lamps, flashlights, etc. every couple of months. I suggest you completely drain your lithium UPS fully and charge it fully at least 3 times, then do that again once or twice per year, and top it off every couple of months. Lithium batteries will get damaged, and sometimes dangerous, if drained below a certain point, over charged, or charged if below freezing temperatures. The built in BMS (battery management system) "should" take care of these things but sometimes they don't properly take care of the undervoltage or under temperature conditions.

Ralph said:

12 volt lithium bank

Click to expand...

I was really only messing with these 3 AH batteries because they were there and because I could. I use them for hobby use and might be able to run a LED lamp for a few hours in an emergency. But, they're not critical. If I purchased new batteries, I'd definitely consider, and probably prefer, LiFePO4. LiFePO4 batteries don't like to catch fire the way LiPo and LiIon batteries do. Dakota Lithium makes cool, if somewhat expensive, LiFePO4 batteries. I don't know if they officially sell on Amazon though.


The trailer brake battery is mission critical though, and I had a definite interest in refurbishing it. I have that battery ready to go when I attach the trailer to the car again. Some of you that have small cargo trailers might find this interesting. The trailer has electric brakes which are activated by way of a controller in the car and connected through the cable that attaches the trailer to the car. There is an emergency switch attached with a metal cable that's supposed to activate the trailer brakes if the trailer ever detaches from the car. The 12 V 5 AH battery is on the trailer itself. I pray to God a trailer detachment never happens. I've considered some alternatives for the battery but, it turns out, that a lead acid battery (if it's working) is probably the best alternative for that. There are a couple of reasons. As mentioned above, lithium batteries cannot be charged below freezing temperatures or they'll be damaged. But, the trailer battery starts charging any time the car is on with the trailer attached, including in the winter. Lead acid batteries don't LIKE being cold. No batteries do. But, they don't self destruct either.

Also, there's a more subtle problem with lithium batteries for this application. If the emergency trailer brakes are activated, they might try to draw 15 A or so from the 5 AH battery. This would only last 15 minutes or so before the battery is drained and possibly destroyed. But, that's its job, to stop the trailer from rolling. (PS, if you DO have a cargo trailer with electric brakes and a backup battery, have YOU (the reader) tested it lately? ) Lithium batteries are limited in their output by the BMS. AND, the BMS is usually rated equal to the battery rating. So, my 100 AH batteries have 100 A BMS's. A 5 AH lithium battery would likely have a 5 A BMS. So, in this application, it's unlikely a 5 AH lithium battery could provide the 15 A that the brakes might need. As janky as lead acid batteries can be, it's probably best for this application.

Ralph said:

AA and AAA

Click to expand...

I'm not a fan of NiCd any more because of their finicky problems. However, I have now converted all my 9 V, AA, and AAA batteries to NiMH except for my smoke detector. (NiMH packs are also available for RC cars and such.) The reason I converted my batteries is that I was tired of buying batteries and I was tired of them corroding in my equipment. They work pretty good. They may last more or less time than the Alkaline's depending on the application. Alkaline's start out over 1.5 V but drain down to below 1 V near end of life. NiMH's maintain about 1.2 V for most of the drain cycle. All my stuff works pretty well with these and they can be recharged hundreds of times. Equipment sometimes complains of low battery prematurely but otherwise works OK. I end up charging some of these about once per week. Maha Energy is a great source for these. I've also heard Panasonic Eneloop's are good.


I get the "precharged" batteries which are supposed to be ready to go, but I still charge them and drain them and charge them 3 X before using them. Sometimes they've been sitting in a warehouse and should be charged before use. This company also has great NiMH chargers.

Ralph said:

build my own

Click to expand...

There are definitely some cool things that can be done DIY. Lithium batteries should always have safety circuits and / or a BMS. I bought a 10 X AA battery holder. I can put the NiMH batteries in there and get a nominal 12 V 2.7 AH battery pack. (10 alkaline's of the same size would make 15 V, not 12 V.) I also bought a 12 V to 5 V buck converter with a USB output so I could charge phones and such with the battery pack. One problem with so many batteries in series is that, if they're not closely matched in capacity, charge, and age, one or two cells can end up draining too fast and end up going too low and being destroyed. You can also do a battery pack with 8 alkaline batteries for a one time use emergency battery pack. A lithium battery bank, as you say, would also be cool.

May your bits be stable and your interfaces be fast. Ron

Electric trailer brakes seem to be a USA mostly item. By me most small trailers are unbraked, and the rest of them use mechanical braking, the most common being a caravan, where the coupler is the actuator, and the disconnect being a very strong spring that engages the brakes on overrun. Then a cable that will operate them if the coupler is disconnected, and it is on the safety chains. But then you do not tow massive loads with a regular license, you need a truck driving one to tow over 3500kg.

As to the lithium battery use on a trailer, charging will be an issue, it will never fully charge either, as the voltage off the vehicle is just a little too low, unless you have put in a split charge system off the alternator, so as to have a slightly higher voltage to charge it. Same for SLA, though the existing running voltage of the vehicle will charge it, and the (by me the only one allowed) standard 7 pin trailer electrical connector does have a pin for an ignition switched 12V supply, normally used to charge batteries in caravans while driving, often the fog light pin repurposed for this over installing a 13 pin connector.

There are vehicles that come with the 13 pin version in addition to the 7, but those are all vehicles intended for the EU market as well. None come with the 4 pin one, it has to be replaced with a 7 pin one, as combined stop indicator lamps are not allowed. The USA really should move away from Henry Ford wanting to save 50c per vehicle, and combining stop and turn, as after all you can get vehicles in more colours than black now. You need 5 cores to have separate indicators, plus stop and side marker lights, which is common on most trailers, though very few small car towable ones come with a reverse light, and normally you only see a reverse light on a truck towed trailer.

Yes lead acid can deliver a massive current in relation to capacity, for a short time, unlike most lithium chemistries. I would say trailers are often the most neglected item vehicle wise, never checked, never serviced, and often abused. A lot will be driving around for years without having the bearings checked and repacked, and the lights checked, unless you get pulled over. Same for the tyres, never checked unless they are actually flat, or missing off the rim on arrival. How many of you with a trailer have checked the spare tyre, and have a jack that fits the trailer, and that it works, and the wheel bolts can actually be removed. I know with me it was in March, because I had it inspected to register it, and the tools and jack are in it, along with a spare. But then, it is a trailer that is rated for 400kg only, made from fibreglass sheet, though after repairs it is almost half steel plate and urethane adhesive now, to fix the cracks.

Sounds like a lot of requirements I was unaware of. I never had a trailer so I never had any reason to look into them. I was toying with the idea of replacing my 2 deep cycle lead acids (when they start to fail) with lithiums. At least as far as my looking turned up there were not too many ways to keep 100AH or so lithiums charged. I am sure there are, I just didn't turn up much when I looked. Prior to that I was thinking of using 2 GC2 6v 200AH golf cart batteries which are flooded. They are mighty heavy and lithium may be a better choice now that big ones are available.

My still new lithium UPS was mainly designed for an alarm system with a USBA port as a bonus. They say to charge it from a 12.0 - 12.7 volt DC adapter. I gave it an initial charge as per manufacturer and started using it to charge a tablet I use with an inline USB power meter to monitor. It keeps around 5.2 volt output under load which makes me think it may be perfect to power a Raspberry Pi. According to a test someone did on Reddit complete with graphs and tables of test methods it performs up to spec. It should prove interesting as I get to use it more. It may just replace the much larger and heavier lead acid UPS once I get around to getting the Pi running again.

Same here as far as NiCds. The last ones I bought were for an old cordless phone I no longer have. Although I have a lot of NMHs my actual battery use has shifted to 18650s followed by 26650s and most recently 21700s. Eneloop NMHs seem to be about if not the best of the brands I've tried. EBL and Amazon's brand of NM have a somewhat higher capacity than Eneloops but at the expense of a lower cycle life. I strongly suspect Amazon and EBLs are the same. From a number of discharge tests on both the discharge curves are virtually identical. I've also had a small number of EBL and Amazons fail and neither were too old or used a lot. Although I've also had a few Eneloops fail they are pretty old and have seen their share of use so that's kind of expected. I've had the failed Eneloops almost as long as they've been available, and most of that original batch are still going.

Now and then I run my batteries through a computerized tester to see if any should be pulled from service. I also rotate through them so that none stay sitting idle for long periods of time. The failure mode I've seen on NMs is they cannot supply more than 100 or so milliamps without a severe voltage drop. I've tried multiple charge/ discharges on them but they still fail. I'll let all my NMs gradually fail without replacing since I've pretty much moved to all lithium.

Ralph said:

replacing my 2 deep cycle lead acids

Click to expand...

Definitely consider LiFePO4 for any deep cycle application. You can drain them flat without harming them, meaning flat as in drawing out the rated capacity and stopping at 10.5 V for a 12 V battery. Much lighter than lead acid, and should last longer.

My 2 ea 100 AH batteries are Chinese brand and cheaper than domestic, but these seem to have a good reputation. Amperetime is now LiTime. I got them from Amazon. Caveat, I haven't had them too long and haven't fully drained them but once. But, they seem good for now.


Based on research, not experience:

Two good domestic suppliers but not as cheap.



A good solar products dealer but their batteries may not be cheap.


Look for Will Prowse YouTube reviews but I don't know what he's been up to lately.

Ralph said:

keep 100AH or so lithiums charged

Click to expand...

Not too hard. A lithium solar controller can do it. I think some inverters have battery chargers but don't own any like that. I have two AmpereTime 20 A 240 W chargers. I manually connect one to the battery bank every couple of months and top the batteries off. Full charge would take 10 HR for my 200 AH battery bank, but the last top off took less than an hour. I had to replace about 5 % of the energy. You can also search for Lithium Iron Phosphate or LiFePO4 chargers on Amazon. You want a battery that has a good BMS that protects against over voltage, over current, over temperature, short circuit, under voltage, and under temperature at a minimum. They may do other things. I found my BMS didn't protect against under voltage apparently but the inverter shut off at 10.5 V and I didn't try to drain the batteries any more. I also don't think my BMS protects against under temperature. You MUST NOT charge the batteries below freezing or they can be damaged. Some chargers may have an under temperature sensor.

Ralph said:

new lithium UPS

Click to expand...

Cool, do you have a link?

Ralph said:

computerized tester

Click to expand...

Also cool, do you have a link?

Sounds like we have some common interests and experience.

May your bits be stable and your interfaces be fast. Ron

Only drawback of lithium is the price, around 8 times the price of SLA for the same capacity, and so far I get 2 years and a bit out of SLA, even though living in South Africa we get regular rotational load shedding, minimum 2 hours a day, because of reasons not discussable on an open forum. So far the SLA is sort of winning, in that 50% DOD seems to be working, at least till the capacity drops to the point that that is a full discharge, meaning time to replace 2 12V units, using the cheapest ones on the market, which are $36 each at current exchange rate, and which are made locally. Same capacity lithium is $800, and only has a 5 year life expectancy as well. Made from recycled EV packs, so they might have a high failure rate, which I have seen with solar lights that also use salvaged EV cells. Replaced with a lot of old laptop cells, which came to more than double the capacity, and a new BMS board, as the failing cell dumped electrolyte all over the existing DW01 based board. Yes I am aware of the poor choice of a DW01 for LiFePO4, with lower voltage, but I suspect the OEM did not care. At least now the cells are matched to the BMS.

@SeanBZA I'm not really getting the math here. Perhaps you can elaborate. But, for example:

Here's a 12 AH LiFePO4 from a good Chinese brand.

LiTime 12V 12Ah LiFePO4 Lithium Battery, Built-in 12A BMS 153.6W Output Power, 4000+ Deep Cycles& 10-Year Lifetime (I think the warranty is 3 years.) ~ $ 60

Here's a 12 AH SLA which has good reviews.

12 Volt 12 AH, F2 Terminal, Rechargeable SLA AGM Battery ~ $ 30

So, it looks like the SLA is half as expensive (but not 1/8). But, that's not the whole story. You can only drain the SLA to 50 %, whereas you can drain the LiFePO4 to 100 %. So, you'd actually have to have 2 SLA batteries being drained to 50 % to equal one LiFePO4 being drained to 100 %. Even LiFePO4's PREFER lesser draining though. The LiFePO4 ad says they have 4 times the cycle life of SLA and 1/3 the weight and a 10 year lifetime (but 3 year warranty). I note that it would be 1/6 the weight if comparing to 2 SLA's.

It just hit me possibly why your numbers look the way they do. You said recycled EV packs. These are probably LiPo or LiIon and are the most expensive batteries you can get, even on the secondary market. That also applies to used laptop cells. I'm not talking about those, but LiFePO4 which is much cheaper even brand new.

It still seems to me that LiFePO4's are pretty ideal for most deep cycle applications that don't freeze or require super high current levels. And, LiFePO4's are much safer than LiPo or LiIon and don't like to spontaneously catch on fire that cannot be put out. NOTE, if your short out a big LiFePO4 with a wrench or screw driver, you STILL may start a fire, while melting the wrench. All large energy sources can be dangerous.

May your bits be stable and your interfaces be fast. Ron

Issue is import duties, and shipping, so things tend to be a lot more expensive, but locally made batteries are cheaper, although there are also environmental levies on them ( that actually never get used to protect the environment, but instead go into general fiscus, but that is another whole can of worms) as well. The lithium batteries are all imported, but the lead acid are a mix of imported and local production, and the imported ones compete with the local ones by being poor quality.

My Computerized Battery Analyzer (CBA) is made by West Mountain Radio. There's a newer version of mine out now but mine fills my needs. Check https://www.westmountainradio com/cba.php replacing the space with a period (I think working links are not allowed here?). Mine without an optional power amplifier can handle 12 volts 5 amps discharge which is perfect for a 100AH lead acid bank at the standard 20 hour rate. You can download the CBA manual for specific info. I use it for AAA cells to the 100AH bank, just about every battery chemistry including multiple cells. I imagine the latest CBA does more. I did buy the optional software some time back which adds different tests to run.

The small lithium UPS I just picked up can be found at
https://www.amazon com/Konnected-Backup-Battery-Mini-UPS-7800mAh/dp/B08JTVY87T/ref=sr_1_8?crid=M2CN83F15A3O&keywords=konnection+ups&qid=1688839760&sprefix=konnection+ups%2Caps%2C261&sr=8-8

..putting a period after amazon- or just search Konnected UPS on the site.

I only had the UPS a few days so I can't speak to it's longevity but it has powered my Pi with external spinning 4TB drive for a while. I haven't tried to see it's maximum run time yet.

The large LifePO4 cells I saw were on the 18650batterystore website. I was looking for 21700s and ran across them. They do carry 18650, 26650, and 21700s at good prices from some well known manufacturers. I've been interested in batteries for some time, dating back to the consumer GE NiCd days with 1.2AH D cells. The batteries are long gone but there's still the old GE NiCd charger hanging on the side of my workbench. Useless now days it's out of the way so it's still there- along with the GE Nickel Cadmium Engineering handbook, and the metal case from a NiCd GE aircraft battery that used to be full of wet cells. Back then I bought quite a few surplus NiCds to play with. When NM started coming out they had very high self discharge rates so I didn't play with them very much until that improved. At around twice the capacity of NiCds at the time they were a nice upgrade.

@Ralph It sounds like you have some really cool information there. Thanks for posting it and the links. I've heard about the West Mountain Radio battery analyzer. It sounds cool, but I've never had a chance to play with one.

Just FYI, I post working links here all the time. I haven't had a problem except sometimes it's snarky about Amazon links. Seems better lately. Also, FYI, on Amazon links, you can chop off everything starting with /ref and afterwards and they work fine. It makes them lots shorter.

May your bits be stable and your interfaces be fast. Ron

I thought I read something here about links, guess not

As with ChatGPT I like playing with my toys to see what they can, or how far I can push them. It is pretty well known that for high drain devices NM AAs outperform alkaline cells, opposite for low drain. I did a few tests with the CBA a while back with fresh Duracells and NM & found that right about 500mA they are equal. So above 500mA you are better off with NM, below 500 use Duracells. The 500mA may not be exact, I didn't want to burn through a bunch of alkalines to get a more precise value but 500 is pretty close. I have to say it seemed wasteful using a battery just once, guess I got used to using NMs!

I don't know why using a solar charger to safely recharge Li batteries never crossed my mind. I was aware of them for lead acid but never actually used one. I checked out the battery links and there are some interesting batteries available. I am pretty sure when my lead acids give out they will be replaced by Li. I believe (not 100% sure) Fedex will deliver up to 50 pounds, and if that is the case I would rather have the batteries arrive on my doorstep than have to lug lead acids in and out of the car- or off a pallet.

Although I don't need one yet, I did some searching for Li batteries in the 100/ 200AH range. I tried Trojan since I looked at their larger lead acids some time back. Although they do make Li they don't have 12v versions. Exide had a few, and as expected Amazon had quite a few. I did notice many seemed to be around $300, some $1000 or more. That's a pretty big spread! Quite a few on Amazon were made by companies I never heard of. Not all of them had built in safety features, but I doubt that would account for the $700 or more difference. I did see that the rated number of charge cycles did vary, some having nearly twice that of others. I'll have to check to see if there are any sites that did actual scientific testing between brands next. As a guess, off grid solar storage batteries would be a good starting point.

@Ralph Don't buy cheap no name batteries. You can end up burning your house down. Ampere Time is a pretty well regarded Chinese brand according to Amazon reviews. They are now Li Time I think. I have two of the 100 AH units. I've drained them completely to name plate capacity (stopping at 10.5 V) for testing and have used them to a much lesser extent a couple of times. But, I haven't had them more than a few months nor tested them exhaustively. I think I paid $ 400 each for 2 batteries plus extra for big fat cables, inverters, chargers, energy monitor, and such. Dakota Lithium and Battle Born are supposed to be good brands. Check out YouTube reviews by Will Prowse although I don't know what he's been up to lately. There are some links earlier in the thread. You definitely don't sub standard Lithium batteries. Make sure you go with Lithium Iron Phosphate (LiFePO4) chemistry.

Some things that add significantly to cost for more prominent companies are R & D, marketing, tech support, customer service, warranty service, and of course profit. It's not just the cost of the battery equipment itself.

May your bits be stable and your interfaces be fast. Ron

I've done some looking around since my last post and it seems there are a number of new manufacturers. Dakota and especially Battle Born seem to be rated well. I agree with avoiding 'no name' brands, I don't even use no names in AAs. On the plus side I am not in a rush to buy any so I have time to do some research. One site did what appeared to be a thorough test except for cycle life. They took the battery apart and actually tested the battery manager functionality. I saved the page as a PDF for reference but it's too large to attach. You can look up PowerUrsus 100Ah lithium iron phosphate battery (LiFiPo4) if interested. Generic case aside they were pretty impressed with it and I like their testing methods with actual numbers. I am not running out to buy one but will keep it in mind.

Ralph said:

I've done some looking around since my last post and it seems there are a number of new manufacturers. Dakota and especially Battle Born seem to be rated well. I agree with avoiding 'no name' brands, I don't even use no names in AAs. On the plus side I am not in a rush to buy any so I have time to do some research. One site did what appeared to be a thorough test except for cycle life. They took the battery apart and actually tested the battery manager functionality. I saved the page as a PDF for reference but it's too large to attach. You can look up PowerUrsus 100Ah lithium iron phosphate battery (LiFiPo4) if interested. Generic case aside they were pretty impressed with it and I like their testing methods with actual numbers. I am not running out to buy one but will keep it in mind.

Click to expand...

It's essential to do your research, especially when it comes to power sources. I agree that well-known brands like Dakota and Battle Born tend to offer more reliable options.
The PowerUrsus 100Ah lithium iron phosphate battery seems intriguing, especially with a thorough test backing it up. It's always a good idea to keep options like that in mind, even if you're not in a rush to buy.
If you're looking into energy solutions, you might want to check out this Residential Energy Storage System. It's another option worth considering for your power needs.

Prior to the last week or so I haven't been on the forum for a while. I got side tracked from a lot of things including the battery research. While away my trusty (?) laptop stopped booting and I couldn't get a restore to run. Long story short I bought a new laptop and while waiting for it to arrive I got my old one back up on it's original Windows 10 from an Acronis image.. With most of the programs gone I configured it as dual boot with Umbuntu on the other side. Unfortunately Spinrite wouldn't run on the internal NVMe drive, but it seems to be OK now.

I reinstalled the West Mountain Radio CBA software, but not the optional extended tests yet. I need to run a test on my lead acid deep cycles since it's been a while. The Lithium research I was doing is intended to replace the lead cells when they are ready for replacement. Those lead acids have a number of years on them but at least as of last check were still in pretty good shape.