Voltage sag (was: 1500w motor overheating on < 20A?)

[Jabotical]

If you have the default winding 4T leaf in a 26" wheel, then your battery is very much undersized for what you're asking it to do. It runs best at 30-40 amps continuous or more. Most motors will not perform well when they are being lugged. NCRBs are very weak cells, and you'd want about 60ah of them to comfortably output what the motor needs without torturing the battery.

Yeah, I kinda figured that out after I'd already ordered everything, heh. Hadn't realized how important the continous current rating was (though it's actually the 5T winding, but that doesn't change the situation much)
Spinningmags gave me much the same advice.

I'm ordering me a new battery this week (probably one of lunacycle's Panasonic GA-based packs that can do 50A continuous, 70A peak, in a smallish form factor -- I must have such a thing now that I realize they exist!)

Gonna try to sell my bottle battery if I can find a buyer who'd better appreciate a discount decently-long range 20 A pack (some mid-drive goon, probably). And if it can't be sold, well, sometimes you make expensive mistakes. I can't give up now that I've had a taste.

 

[neptronix]

Ahh.. okay. 5T does change things. I was referring to a 4T situation, which is kinda unique due to the freakishly high amp requirements.

If you want a lunacycle GA battery pack, you'll want the biggest one money can buy.. 25ah or more, so that you run them around 1C.
Those newer high energy 3C chemistries are best at 1C-ish. All of em.

Samsung 25Rs are a different matter.. you could get away with 15ah. They should be happy around 2-3C even in the heat.

If your motor has a temp sensor inside, i would get some readings off that instead of the hub case. But any motor out there will perform worse starting from 110f! ( hat tip to you for even riding in those kinds of temps - that's heroic! )

 

[Jabotical]

If you want a lunacycle GA battery pack, you'll want the biggest one money can buy.. 25ah or more, so that you run them around 1C.
Those newer high energy 3C chemistries are best at 1C-ish. All of em.

Samsung 25Rs are a different matter.. you could get away with 15ah. They should be happy around 2-3C even in the heat.

Hmm, good to be aware of. I know you'd said earlier you like to run yours at 1/4 their rated continuous current capability, which does seem like a nice luxury. What's the consequence of drawing their advertised continuous current (at 3C)? It's nice to avoid voltage sag, but a little of that might not be the end of the world... 25ah is getting pretty enormous. I was hoping to avoid anything that dominated my frame too much, and was looking at a 17ah option, with 5p @ 10A each = 50A continuous. I'm generally not going to be going much over 30 mph, which in my setup (at steady state) looks to draw at most half that many amps from what I've seen so far.

Ah, and the 24ah GA pack they had is now out of stock. As are the 25R options. I'd like to stick with 18650 packs out of convenience and beginner-ness, for now. I wouldn't mind making my own pack at some point, but would rather get settled into the ebike thing a little more first. Maybe I'll consider one of the eBay or aliexpress sources...

*EDIT* Looking at the discharge voltage graph for GA cells at different currents, I see what you mean about voltage sag. I'd probably have gone with that 24ah pack at this point, but alas. The 17ah will probably do me pretty well for the next couple years (certainly compared with what I've got now!), until something markedly better comes along.

If your motor has a temp sensor inside, i would get some readings off that instead of the hub case. But any motor out there will perform worse starting from 110f! ( hat tip to you for even riding in those kinds of temps - that's heroic! )

Yep! The problem was it had the wrong type of thermistor for the CA to interpret (100k vs 10k). But I looked up a couple typical tables and translated the readings I took close enough, I think. And it never went over 77 C. Which I gather is considered quite safe territory.

Hehe, yeah, it's crazy hot out here these days. Thanks -- I discovered ebikes and didn't want to wait until cooler weather to explore them and start commuting on one. Figured I'd lose my momentum otherwise. The sweaty backpack back is the biggest problem so far.

 

[neptronix]

Well, when you run the batteries at their maximum continuous current, or close to, you get a lot of heat, shorter life, lots of voltage sag ( thus lower speeds under heavy load, especially when climbing hills ), and less amp hours / watt hours out of the battery, as you're already noticing.

Just consider how these cells are rated. The max continuous C rate you see is what they'll just barely produce in a temperature controlled lab, not assembled into a pack where all the cells are heating each other up and there is no venting. That's the absolute edge of what they can do in a totally ideal situation.
You can see this in the spec sheet for the GA or any other cell for that matter by just googling the model name + 'data sheet'
Vendors often go on to advertise them as high power just because of what the lab said... but your pack does not operate in that condition.. unless you add liquid cooling or something ( but then you still get lots of sag )

25ah is enormous but high energy packs are not designed for high power in mind. They are designed with smaller motors in mind, or extreme range on more powerful motors.

So yes, if you want a small battery pack, you have to go with a cell that can handle the output properly by adding overhead beyond the manufacturer's specs.
The 25R's at their 8C rating will do the job pretty nicely at a minimum of 15ah or so, since your constant draw is around 30A.

You'll get around 30 miles with a 15ah 48v or 52v pack of 25R's. That's not too shabby.

 

[dogman dan]

So, unless I'm confused, your CA was ramping down the power because of a false hot reading on the thermistor? Personally, i'd disconnect that feature. You won't be frying that motor unless you ride a rig the size of Amberwolfs stuff.

Yeah, my guess was correct, 77c is not frying your motor, not unusual for a desert ride in the city, or similarly harsh off road riding.

Your little battery was not the whole problem then, but a larger one is a good idea. Even limiting to 20 amps, that big motor will whack it. Why is that? Because the big motor can pull that max amps very hard and very long, compared to a smaller motor. So it gives that poor smaller battery a long hard amp spike, and then lots of them, riding in the city.

 

[Jabotical]

Well, when you run the batteries at their maximum continuous current, or close to, you get a lot of heat, shorter life, lots of voltage sag ( thus lower speeds under heavy load, especially when climbing hills ), and less amp hours / watt hours out of the battery, as you're already noticing.

Oh yeah, my current situation is certainly not ideal. Though part of the problem right now is that the pack can only do ~20A even nominally, and I need more. Makes sense to want more headroom too, and you're right about what the spec sheet reveals.

So yes, if you want a small battery pack, you have to go with a cell that can handle the output properly by adding overhead beyond the manufacturer's specs.
The 25R's at their 8C rating will do the job pretty nicely at a minimum of 15ah or so, since your constant draw is around 30A.

You'll get around 30 miles with a 15ah 48v or 52v pack of 25R's. That's not too shabby.

Yeah, that sounds pretty ideal for my situation. I just need to find such a pack for sale now! (Or make one, but I wasn't planning on that, this time around...)

 

[Jabotical]

So, unless I'm confused, your CA was ramping down the power because of a false hot reading on the thermistor?

Heh, well almost. I had the issue confused myself, but the CA was ramping down the power because I'd set the low voltage cutoff too high in the face of how much loaded voltage sag I was getting.

Personally, i'd disconnect that feature. You won't be frying that motor unless you ride a rig the size of Amberwolfs stuff.

Yeah, I don't want a repeat of any mysterious power reduction. The CA was actually misinterpreting it in the low direction, fortunately. Right, I'm not in Amberwolf territory yet, haha.

Yeah, my guess was correct, 77c is not frying your motor, not unusual for a desert ride in the city, or similarly harsh off road riding.

Sweet! It's extremely valuable to have your experienced input about safe internal heat levels. Things that are obvious to veterans can be uncertain and vexing to newcomers like myself!

Your little battery was not the whole problem then, but a larger one is a good idea. Even limiting to 20 amps, that big motor will whack it. Why is that? Because the big motor can pull that max amps very hard and very long, compared to a smaller motor. So it gives that poor smaller battery a long hard amp spike, and then lots of them, riding in the city.

Makes sense. Hence the voltage sag (as neptronix has emphasized). Plus I want more than 20A anyway -- I can't be satisfied with that when I briefly got to taste 30A before I reduced the cutoff (and if 30A is exhilarating, what about bursts of 40 or 50!). Just gotta find me a suitable replacement battery, now...

 

[Wheazel]

How much power are you pulling through the leafmotor? 1kW?
I think my leafmotor has a more demanding task than many other applications.
The bike it sits on probably weighs 60kg, I am 75kg. The bike has crappy aerodynamics and is most often ridden between 35-50km/h. (Like 20-30mph).
The motor will only warm up fast if I pull 25-30A+ through it. I am running an 18s config, so ~65-72V is what the motor sees.
At around 700W - 1kW where most cruising occurs, the motor will just not heat up.

The big difference here is outside temps tho, rarely above 25C outside in the summertime.
If I run it abit harder, typical delta-Ts with the ambient is 40-60Celcius, but thats with plenty of 1,5-2,5kW running.

 

[teklektik]

You'll get around 30 miles with a 15ah 48v or 52v pack of 25R's. That's not too shabby.

...I just need to find such a pack for sale now! (Or make one, but I wasn't planning on that, this time around...)

Easy. EM3EV has these in a several configurations and can build you a custom pack if the off the shelf versions don't fill your needs.
A well established and respected battery manufacturer on ES run by an English fellow in China.

Apparently the CA has the default max temperature at 130 C (266 F).

I expect I'll set the limit at 200 F / 93 C, and try to stay below 180 F (maybe even 150 F) as dogman suggests, in these beginning days of ebiking. Would rather not push my luck on my commuter vehicle

If you want to look for sound engineering advice on motor heating, you need look no further than Justin. He set the CA defaults of 90 -130degree C to be very safe and to allow you to operate with negligible chance of motor damage. This will stop the motor long before the halls are damaged or, more importantly, you reach the Curie temperature where the magnets cannot regain magnetism lost because of temperature. Running far below the thermal limits will not materially increase the life or reliability of your motor.

Rather than strapping your bike into unnecessary limiting with the added consideration of high ambient temperatures, I would recommend raising the low bound to perhaps 110degC so that if confronted with unexpectedly high loads, the motor has an opportunity to reach a higher (but still safe) equilibrium temperature on its own before limiting begins. In the end, the CA settings are trying to keep you away from the upper temp limit and allowing the motor to do it on it's own with a higher low threshold can be as effective as explicitly limiting power. The primary downside to having the low and high CA limits close together is that the power will fall off more rapidly if the temperature continues to rise in that band.

Clearly, swapping in the proper 10K thermistor would be on the table to allow the CA to operate properly. Here you can just leave the old one epoxied down to avoid removal damage, move the leads, and glue in the new one.

 

[Jabotical]

How much power are you pulling through the leafmotor? 1kW?

Right now I'm limiting it to that, I think I was a tad over that at full cruising speed (esp. up inclines) when I allowed it.

I think my leafmotor has a more demanding task than many other applications.
The bike it sits on probably weighs 60kg, I am 75kg. The bike has crappy aerodynamics and is most often ridden between 35-50km/h. (Like 20-30mph).
The motor will only warm up fast if I pull 25-30A+ through it. I am running an 18s config, so ~65-72V is what the motor sees.
At around 700W - 1kW where most cruising occurs, the motor will just not heat up.

That's certainly more demanding than my 14s config. Though you weight 11 kg less than I do. It's very useful to have your point of reference.

The big difference here is outside temps tho, rarely above 25C outside in the summertime.
If I run it abit harder, typical delta-Ts with the ambient is 40-60Celcius, but thats with plenty of 1,5-2,5kW running.

Yeah, it sounds like you and dogman are right about that. It's often been over 43 C outside when I've been returning from work or whatever. I'll be very interested to see how much cooler the motor runs when the blessed fall (and wonderful winter!) arrive.

 

[Jabotical]

Easy. EM3EV has these in a several configurations and can build you a custom pack if the off the shelf versions don't fill your needs.
A well established and respected battery manufacturer on ES run by an English fellow in China.

Ooh! Perfect. I'd glanced at EM3EV earlier but hadn't really seriously considered them as a potential battery source. Thanks for the tip! Their BMS limit is 40A continuous, but I don't know why that troubled me -- I doubt I'll ever want to maintain more than 40A (and they advertise ~80A burst)

Now I just have to decide whether to get Samsung 25Rs or 30Qs in the 6p pack. 14.7ah vs 17.7ah, 120A vs 90A rated battery constant current.
It's hard not to want the extra capacity, but I think neptronix is right that I'll be more satisfied with using a smaller amount of the batteries' rated current, as opposed to having a few more AH.

If you want to look for sound engineering advice on motor heating, you need look no further than Justin. He set the CA defaults of 90 -130degree C to be very safe and to allow you to operate with negligible chance of motor damage. This will stop the motor long before the halls are damaged or, more importantly, you reach the Curie temperature where the magnets cannot regain magnetism lost because of temperature. Running far below the thermal limits will not materially increase the life or reliability of your motor.

Rather than strapping your bike into unnecessary limiting with the added consideration of high ambient temperatures, I would recommend raising the low bound to perhaps 110degC so that if confronted with unexpectedly high loads, the motor has an opportunity to reach a higher (but still safe) equilibrium temperature on its own before limiting begins. In the end, the CA settings are trying to keep you away from the upper temp limit and allowing the motor to do it on it's own with a higher low threshold can be as effective as explicitly limiting power. The primary downside to having the low and high CA limits close together is that the power will fall off more rapidly if the temperature continues to rise in that band.

Awesome. I love having breathing room, and based on my experience so far 110 seems quite generous.

I'm a little unclear about which low CA limit you're talking about, though. The upper bound is the high temperature limit, but there's not a minimum temperature limit is there (doesn't really make sense?). Are you talking about the low voltage cutoff?

Clearly, swapping in the proper 10K thermistor would be on the table to allow the CA to operate properly. Here you can just leave the old one epoxied down to avoid removal damage, move the leads, and glue in the new one.

[/quote]

Yeah, I definitely want to do that, even as used to the current rough mapping as I'm becoming. I should have a couple 10k thermistors arriving in the mail soon. Just need to grab me some thermal epoxy, and hopefully it won't be too tricky. Good idea, hadn't occurred to me to just leave in the old one.

 

[teklektik]

I'm a little unclear about which low CA limit you're talking about, though. The upper bound is the high temperature limit, but there's not a minimum temperature limit is there (doesn't really make sense?). Are you talking about the low voltage cutoff?

There are two temperature settings: Temp->ThrshTemp (the low limit) and Temp->MaxTemp (the high limit). The CA does no temperature limiting at all until the temp reaches ThrshTemp at which point it begins to clamp the ThrOut signal (i.e. small throttle is unaffected, but big throttle gets clamped to a lower value so you don't notice anything at all unless you are being Bad).

As the temperature increases from ThrshTemp to MaxTemp the CA increases the ThrOut voltage limiting from 0% to 100% so at the high limit the throttle is shut down completely. The idea is that as the temperature creeps into this limiting band, the CA will cut into your WooHoo fun so the motor temperature will stop rising but you keep riding. If the temp keeps increasing, the CA cuts into your fun more and more until you may just be left pedaling. The closer together the settings, the more abrupt the cutoff. If you set them far apart, the rollback will be more gentle, but the power loss can be noticeable if you are climbing a hill, etc where the loss of power may actually push the motor into slower operation and less efficient operation (makes more heat). Since this isn't all that cut and dried, the two limits are adjustable so you can tune this up for your terrain and riding style. Of course, if you get stuck in a desperate pickle on a mountainside in the rain, you can just quickly reconfigure the CA to change or disable the limiting and blast home.

I should have a couple 10k thermistors arriving in the mail soon

Ya, pretty effortless installation - although it's always annoying to crack the case. The only thing to look at twice is the thermistor beta which ideally should be 3900. That said, the beta mostly affects the high end of the curve and anything in the neighborhood is fine - you will only get a few degrees of error at 130degC if you're off by +/-100 or so. 3950 is pretty common and just fine, but I've used 4050 with no difficulties at all.

 

[Jabotical]

Thanks for taking the time to explain that so very well. I'd read over the CA unofficial manual, but clearly not everything has stuck yet.

And I'm sold -- 130 C limit with 110 limit threshold it is!

I do love how easy it is to change things in the CA on the fly.

Ya, pretty effortless installation - although it's always annoying to crack the case.

Yeah cracking the case is the part that sounds potentially daunting. But I wouldn't mind seeing inside there, anyway.

The only thing to look at twice is the thermistor beta which ideally should be 3900. That said, the beta mostly affects the high end of the curve and anything in the neighborhood is fine - you will only get a few degrees of error at 130degC if you're off by +/-100 or so. 3950 is pretty common and just fine, but I've used 4050 with no difficulties at all.

Good to know. I also just found the relevant section in the manual which I should have read before ordering. Fortunately the ones I've got coming advertise the B25/85 Value as 3960K, which sounds like it'll be swell.
Thanks for all the help!

 

[Jabotical]

Neptronix, do you (or anyone else) happen to have a relevant opinion on the Samsung 30Q vs the 25R?

I ordered a ~15ah 25R pack, which will probably do the job. But for $100 more I can make it a ~18ah 30Q pack.

30Q: 5c, 90A continuous, 18ah
25R: 8c, 120A continous, 15ah

The extra capacity margin would be nice. If I'm generally sustaining at < 30A, would I notice much of a performance difference in real world use?

[EDIT: Fixed which AH went with which battery]

 

[spinningmagnets]

30Q @ 18-Ah = 6P.....6P X 15A per cell = 80A

25R @ 15-Ah = 6P.....6P X 20A per cell = 120A

Both would be the same size and weight, so...I guess it would depend on if you want more range, or if you think you will ever want to raise the amps someday?..

 

[Jabotical]

Yeah, I suppose that's the obvious tradeoff.

Mostly I was trying to figure out whether I could get 30 sustained amps (with bursts of more) from the 30Qs without the voltage dropping unduly.

They sound like they provide almost comparable performance to the 25Rs, just with a little more range and a tad less (but still solid) current. Possibly exactly what I want...

 

[neptronix]

Neptronix, do you (or anyone else) happen to have a relevant opinion on the Samsung 30Q vs the 25R?

I ordered a ~15ah 25R pack, which will probably do the job. But for $100 more I can make it a ~18ah 30Q pack.

30Q: 5c, 90A continuous, 18ah
25R: 8c, 120A continous, 15ah

The extra capacity margin would be nice. If I'm generally sustaining at < 30A, would I notice much of a performance difference in real world use?

[EDIT: Fixed which AH went with which battery]

Let's get out calc.exe and look at some data.

Samsung 30Q data sheet
http://www.nkon.nl/sk/k/30q.pdf

Look at the discharge current vs what you get out.

At 1C, you are not even getting the nominal voltage of 3.6v, you're getting 3.53v nominal. Pretty close, and not too saggy.
At 3C, you are getting 3.39v nominal, so 10% of your battery pack's capacity is being turned into heat. If you built a 13S 48v pack, it is going to be 44v average and cut off early if you did 3C all day.

The best use case for any battery, if you want what's promised on the cell spec sheet, the cycles advertised, the voltage advertised, the amp hours advertised, etc...
is to use 1/4th of the maximum C rate specified by the manufacturer. So in this case, 1.25C is what you'd want from a cell like that.

The maximum C rate is basically the point where the battery just barely avoids blowing up or catching on fire in a temp controlled lab, and not enclosed. The manufacturer publishes this and the vendors parrot the figure, but like to leave out the fact that the cell will perform really bad that way. hobbyking and panasonic, samsung, lg etc all do this.
And the capacity is always rated at a really cherry picked 0.2C or 1A figure. Hobbyking is the only manufacturer that underspecifies this... IE, at 0.2C, you will find 16.5-17ah in their 16AH packs.

All you gotta do is google 'cell name data sheet' to get this info.. most data sheets include a discharge graph that shows multiple amperages to show you how bad the voltage drop is. Multiply that voltage drop by the number of cells you have and you can get an idea of how much sag you're gonna hit when you hit the throttle.

Let's look at the samsung 25R, then.

https://www.powerstream.com/p/INR18650-25R-datasheet.pdf

At 5A, which corresponds to 2C, we're getting 3.54v nominal.. so this cell has the same voltage drop as the 30Q at twice the power.
At 10A, or 4C ( 10 / 4 = 2.5ah ), we get a really sad 3.34v nominal. This is similar to the 30Q's voltage drop on 3C.

I don't know why they even listed it, but they did a 10C test. This gives you 3.18v nominal which is lower than lifepo4's nominal voltage. This condition blows 17% of of the pack's energy into heat. So if you had a 1000whr pack running at 10C, the battery is creating 170 watts of heat in an enclosed pack - maybe you'll get 50 cycles out of the pack, lol. Your 48v pack is now 39.85v nominal under load.

So, the 25R can handle around twice the power, until you start really pushing it.

But do you see what i mean about using 1/4th of the rated power for every cell? it rings true with the 25R and the 30Q.. it rings true with RC Lipo, lifepo4, etc etc.

 

[Jabotical]

But do you see what i mean about using 1/4th of the rated power for every cell? it rings true with the 25R and the 30Q.. it rings true with RC Lipo, lifepo4, etc etc.

I do indeed. And thank you for once again taking the time to explain so elaborately. May someone else find your comments (and spinningmag's, dogman's, ...) when they have similar questions.

All right, I'd wondered if maybe the 1/4 capacity rule of thumb applied most especially to Lipos or something, but you make a compelling case. I always appreciate data-driven conclusions.
I'd looked up those spec sheets before, but wasn't sure what was the relevant information. They provide really useful insights, now that I'm looking at certain portions through your lens.

Those "venerable" 25Rs really hold up, don't they. Shockingly, it seems there is no free lunch -- everything is a compromise one way or another. Fortunately there are sufficient possible configurations out there to get the capacity and realistic current for any given application, if you're willing to accept the necessary weight and size. And also fortunately, the cells really are quite impressive these days. I can't imagine trying to do this back in the Nickel-Cadmium days, or with Lead Acid batts, etc.

 

[neptronix]

You got it. ( by the way, you said 1/4 capacity rule.. i hope you mean 1/4 C rule, which is related to discharge .. )

The problem is that most people don't understand the specifications they're looking at. So they end up buying stuff that doesn't work.
Or they don't look at them at all and trust the vendor, who is touting cherry picked specs off a sheet from a cell manufacturer that also likes to cherry pick off those specs too.

Hence, the 18650GA becomes a high power cell with a cycle life of 1000 cycles, for example.
What they don't tell you is that the GA hits 80% capacity after something like 125 cycles. around 66% at 300 cycles or so.. leaving you with 2.6ah per cell capacity.

If 3C on a 3.5ah cell is 10.5 amps, then 3C after 300 cycles when the cell is only 2.6ah is 7.8 amps. Continue pulling 10.5A from the cell and you're actually pulling 4C and then the pack is sagging like mad and will end up in the dumpster sooner. that's a shame.
You can get that magical 1000 cycle figure if you just run 1C the whole time and buy more battery than you need, so that when it's old and crusty, it still performs well.
And hey, you just saved yourself 2 battery replacements.

The best way to understand battery performance is through graphs. I am a visual learner and my iCharger 1010b+ and cycle analyst taught me a lot of what i know.. the rest is calc.exe and http://www.percentagecalculator.net/ :lol: . break out tools like these if you wanna learn more.

 

[Jabotical]

Yes, that was a mental slip, I definitely meant C (Which is related to both discharge and capacity, no?)

Yeah the nominal ratings of voltage, capacity, current, and cycles alone are definitely insufficient to make a decision. I didn't really appreciate that at all until your explanations with charts and examples. I agree that graphs are king for really getting a grasp of what's going on with this sort of precise technical phenomenon.

I was certainly a bit hasty in buying my first battery, though of course even its rated current proved to be insufficient for my needs, haha. But I chalk that up to the cost of learning -- I'm quite satisfied with the other components (bike, motor, controller, CA), so I can survive one poorly-chosen battery. Thanks for helping me not do that a second time (to a lesser extent), heh. One does want one's battery to last.

I covet getting a fancier charger one of these days, that will give me mucho information. I love data (and control).

 

[neptronix]

Hey man, you might be new here, but you're off to a very good start

It sounds like you enjoy what i'm typing, so here is a rant with you in mind.

Ever since i landed on ES 6 years ago, i wanted a high power, high distance ebike that could i could get really lost on, as a means of relieving stress and enjoying nature to a greater extent than i could with pedal power. Batteries were the weakest link, of course.

Nowadays i live in the suburbs with very few bike lanes and high road speeds. A battery that can give awesome range, good power, and low voltage sag ( thus high discharge efficiency ) is mission critical. I'm currently falling short of that goal, and only have 20 miles of range when i have no bike lane due to the high speeds involved ( 35-45 with a LOT of pedaling while in a full tuck with a huge chainring )

So all in all, i've spent crazy amounts of time analyzing battery specs, and that's how i know what i know. Because damnit, i am not giving up on the dream of having an ebike with tesla model S range.

If you could invest a little money in some nice RC Lipo balancing equipment that does graphing ( as a means to determine pack health and suitability for unbalanced bulk charges ), you might wanna consider going the RC Lipo route if you're obsessed with optimization as much as i am. RC Lipo is the least safe and most temperamental, but an anally retentive user of it should never have a problem. The advantage is that it's cheap and gives you the best combo of power density and energy density.

Here is my current dream battery. These are 250-260whrs/kg cells, which absolutely beats every 18650 on the market today. 45mph requires about 3000w constant on a really efficient hub or RC drive motor, so given that 45mph goes a hair over 1C with the 2.88kwhrs pack, and 35mph lands us in that comfy 1/4th C rate @ 1400w with intense pedaling, i'm occasionally breaking my own rule, but damn if it won't do the distance thing like a champ.

You could expect a 60 miles at 30mph range with the smallest configuration here.

 

[dogman dan]

I swore I was done with lipo. But then I bought 48v 20 ah more. It just simply has the c rate to handle a 40 amps controller. Less sag means you go farther before you have to slow to keep the voltage above lvc.

I keep my lipo outside, or in the fireplace.

 

[Jabotical]

Hey man, you might be new here, but you're off to a very good start

Hey thanks! Yeah I've been very impressed with how knowledgeable and willing to help you and the rest of the community have been.

I'm certainly addicted to this ebike business. I can't believe I only just discovered such a thing even existed/was reasonably possible. My commute has been so much more enjoyable lately, even with my underpowered battery situation.

So all in all, i've spent crazy amounts of time analyzing battery specs, and that's how i know what i know. Because damnit, i am not giving up on the dream of having an ebike with tesla model S range.

I think you must continue to pursue that dream.

If you could invest a little money in some nice RC Lipo balancing equipment that does graphing ( as a means to determine pack health and suitability for unbalanced bulk charges ), you might wanna consider going the RC Lipo route if you're obsessed with optimization as much as i am. RC Lipo is the least safe and most temperamental, but an anally retentive user of it should never have a problem. The advantage is that it's cheap and gives you the best combo of power density and energy density.

Man, I've gotta admit that the longer I've lurked here the more the Lipo approach has tentatively appealed to me. Their sensitivity and danger remains a turnoff, but perhaps if I charge them outside or in a fireplace like dogman (or... a pressure cooker?) I could give 'em a shot. They sure sound like they have every advantage, otherwise. And the graphing balancing equipment appeals mightily.

And that dream battery of yours looks mighty legit. Capable of burning your house to the ground, perhaps, but also of powering your bike into the cosmos. How close are you to making it a reality?