Home Creators Posts Import Register Favorites Logout

Content

I’m finding that many people aren’t using my Recommended Batteries tables because they are too complicated, with too much information to try to understand.

Most people have just one application they want the “best” battery for and they don’t need a table with all the best choices for different uses and power/current levels.

So I’m going to try a test…setting up separate tables for each of the different main uses. Unfortunately, the safety info needs to still be there in each table but I think I can make these tables a lot simpler. But I need your help with the list of different tables to create!

And I have no idea how I’m going to handle 18650 vs 21700 either. A single table could have four columns, 18650 high and low power and 21700 high and low power…getting complicated again!

What should I add/remove/change to the categories here (in alphabetical order)?

  • E-flight (not racing) - a combination of highest voltage under load for power and longest run time down to around 3V. Total Wh might be the best number to use? Extra emphasis on available power at only 50% charge too, for landing. Cell size/weight is critical.

  • Flashlights - a combination of highest voltage under load for max brightness/power and longest run time down to around 3V. Total Wh might be best here too? Or are some crazy bright lights only run for short periods and highest voltage is critical?

  • General Use - the existing Recommended Batteries tables.

  • PEV/LEV/micro-mobility - a combination of highest voltage under load for power and longest run time down to around 3V.

  • Racing (air/ground) - Highest voltage under load down to around 3.6V with size/weight being a priority too.

  • Vaping (regulated) - a combination of highest voltage under load and longest run time down to around 3.2V. Wh might be best here?

  • Vaping (mech) - highest voltage under load down to about 3.6V. Lower power mech users can use the regulated vaping battery recommendations.

There will be a LOT of duplicate entries in these tables with a couple perhaps being identical. Some tables will probably be split into low/high power columns.

Eventually this will be on a web site but a huge part of the world can’t access US web sites reliably (or at all) so not sure what to do their either. Having JPG’s on Facebook makes the info accessible to a lot more people.

So what should I add, remove, or change in the categories listed above? Thanks!

Comments

BongleHeimer

"Or are some crazy bright lights only run for short periods and highest voltage is critical?" this is my main interest, personally, but runtime is perhaps interesting to *most* other people.

Phaser3D

What about a MCP so u can ask ai what is the best battery for my application etc

ibey

I just want to know which cells can hold 20, 40, 60, 80 and 100A discharge down to 3V without overheating.

EUCrusade

Flashlights are a bit more complicated as they use various drive circuits. Linear drivers actually want a saggy cell as the excess voltage(Vin above LED Vf, so 4.2V at the battery, on low mode the Vf of the LED might be under 3V meaning about 1.1V difference) is burned off as heat. Linears are mainly found in older enthusiast and super low budget lights. Buck drivers want the highest voltage so they can keep the LED going for as long as possible as bright as possible. Boost drivers crave delivered energy as they jump the voltage up to the higher LED drive voltage. Fet drivers are usually found in combination with buck or linear and act as a turbo mode. V is king as many combo drivers just let the fet go unregulated. These are not sustainable levels in any sense so short bursts of high current 15-50A depending on exact light and setup. With regards to PEV, there would be a need for tiers. As someone looking to re-cell a dead ebike battery likely has different priorities to someone building a custom pack. Maybe like a casual/low power and a high power category split?

Andrea

E-flight: highest voltage and run time till they hit temperature limit or voltage limit. Nothing else is really needed.

Volvoracing Germany

Power tools with recommendation for pack configuration as the most common packs are 5s1p, 5s2p, 5s3p (i own a 5s4p Makita DIY pack but this is very uncommon). I.e. my actual most overkill pack is a 5s3p Tenpower 50XG. I dunno the actual load profile for an Makita impact driver (DTD 173) but find a 5s1p 18650 pack very handy. Could this be a category to consider?

osen

As for flashlights: yes to all the questions more or less. I'm not sure if it's going to be less complicated than the current table honestly. Buck drivers: mostly total Wh down to Vf of the emitter at reasonably high output, lets say 3.2V or so. For highest output often higher. Boost drivers: just total Wh down to LVP, usually ~2.8V (i.e. current table). For FET basically just as little voltage sag as possible. Linear I wouldn't even bother with including. TL;DR: The main thing missing for flashlight use currently is probably Wh delivered down to a higher voltage relevant for buck drivers, like 3.2V. Maybe highest voltage for FET at different levels but you can already get that info from the shootouts. I.e. a table would probably need to be segmented into separate categories for buck/boost/FET drivers.

Bill Richards

I mainly use my 18650s for vaping and often go back to the graphs and look at what is the best at 10A or 40A down to 3.2 or 3.25 these are not always the best overall down to 2.8V so having lists based on vaping minimum voltages would be good. Fortunately I can read the graphs easily and understand that batteries that stay cooler should have a longer charge cycle life that those that hit temp limits earlier. So I think this new way to rate for different purposes will be good

Frank

I'm on the lookout for 2170 cells that fit in the Racing category, i.e. minimum voltage drop at 80-100A loads for ~30 seconds. Actual runs are longer but max battery load doesn't occur until motor voltage nears pack voltage. This for 1.0-1.5 mile land speed racing. I also drag race which has higher loads but for very short periods of time and it's hard to beat upper end pouch cells for that application but I'm always looking.

Satori80

Sorry to sound pessimistic, but I don't see this working out very well. Even with vaping, I've got devices that range from 60 to 400W. How do you account for that in a simple way? There is just no getting around the fact that li-ion cells aren't a one-size-fits-all solution even if you split their use into categories. I feel like users are going to have to meet this information part way if they are ever going to get anything out of it. This isn't exactly rocket science.

Battery Mooch

I had considered that but a lot of up-front work for a relatively small dataset…not sure if it’s worth the time but it’s still on the table. 🙂

Battery Mooch

You might be amazed at how incredibly complex even the concept of Wh or Ah or voltage-under-load is for the vast majority of people who don’t want to learn anything. They just want to buy the “best” battery. 🙂 But you might be right!

Satori80

Actually, I wouldn't be amazed. I had to write an Ohm's & Watt's laws tool with plenty of notes about parallel vs series behavior because I can't always keep this stuff straight myself. I use it more often than I care to admit. I keep mixing up anode & cathode, too. It doesn't change the fact that there is no overall 'best' battery. As much as I admire your drive to be helpful to as many people as possible, if one isn't willing to put in a little effort, they will never find what they are looking for, and I don't believe anybody can simplify a way around that fact. **edit** After further thought, it's almost possible, but you'd need to break the categories into so many pieces that it would become even more confusing than your current solution. It would still be ballpark, though.

me

I would suggest even a 3.2v cutoff. Does anybody actually run cell down to 2.8/2.5v?

ibey

I dont but at those high currents voltage drops much lower than what the resting voltage of the cell is.

Battery Mooch

I used 3.2V for years but a large number of people wanted it to go down to 2.8V since the voltage sag results in “bounce back” up to 3.2V or even 3.5V or so at higher current levels. Knowing this more people are running their cells down to lower voltages. But many device still do have a higher fixed cutoff. I don’t have a good solution for this yet.

OverStack

For non racing e-flight, careful, the SINGLE most important metric is wh/kg !! Not runtime. The 2nd most important metric would be 60~120s pulse power density (W/kg) at both high and low SoC for vtol, and cycle life for high value fixed wing batteries

OverStack

Hello, for e-flight there's slightly more: first, what you said (voltage and runtime) is energy (Wh). next, if you have a 100g battery that runs 10 minutes, and a 200g battery that runs 12 minutes, by your statement the 200g is better. But in reality you can use 2 100g batteries in parallel to get 20 minutes of runtime instead of 12. So the metric to quote is Wh/kg, because it encompass voltage, runtime, and mass. Mass is suuuuper important in an aircraft !

Andrea

Ok. About weight, we know by default that a 21700 weights more than a 18650. We don't care if a 21700 weights 70gr or 75gr right?

OverStack

That's the thing, it actually makes a very large difference ! There are 67g cells and 74g cells, that's a 10% difference. That's why for example the rs50 beats the 50xg by a fair margin for drone endurance, because it's significantly lighter

izzeww

I think for "PEV/LEV/micro-mobility" then CCC rating should be a requirement because you want that stability in a large pack you invest a lot of $ into.

OverStack

Exigent applications like drones run the cells to 2.7-2.8 for "regular" use, and 2.5v for extreme long range as they need every bit of energy available in each cells