cell_man Li-NCM triangle packs, two sizes now
- Thread starter spinningmagnets
- Start date
SamTexas
1 MW
Who do you want to believe? Those that can only offer talks without any data or hand-on experience or those who can present actual data?pithy said:
Heck, at least one of those guys pretended to be an expert on laptop batteries for about 2 years until I gathered enough information to unmask them publicly.
If you go back just a few posts you will see someone (not me) offering an actual set discharge graphs at various currents. If you don't understand the meaning of those graphs, ask. Then make your own conclusion.
pithy
100 mW
Please do understand that I'm not trying to argue anything here in any direction about these packs--just trying to gather and understand info. 
You are correct in guessing that I don't know how to interpret that discharge graph. More info/explanation would be appreciated.
In another thread, you (SamTexas) suggested that the cells used in this pack (Samsung ICR18650-22P) are "typical laptop LiCo". But the discharge page mentioned earlier says that these are "IMR Li(NiCoMn)". Did you mean that they're similar to LiCo cells? I'm not trying to argue anything here, just trying to understand. LiCo and LiNCoMn are vastly different in terms of safety, aren't they? (please correct me if I'm wrong about that) But maybe you're just saying that those two types of cells are similar in terms of what C-rate can be safely used with them?
At the same time, this page http://www.electricbike.com/lithium-cobalt-magnanese/ suggests that these cells should be "a solid reliable 3C chemistry, which can provide a continuous 30A from an affordable and compact 10-Ah battery".
:?
You are correct in guessing that I don't know how to interpret that discharge graph. More info/explanation would be appreciated.
In another thread, you (SamTexas) suggested that the cells used in this pack (Samsung ICR18650-22P) are "typical laptop LiCo". But the discharge page mentioned earlier says that these are "IMR Li(NiCoMn)". Did you mean that they're similar to LiCo cells? I'm not trying to argue anything here, just trying to understand. LiCo and LiNCoMn are vastly different in terms of safety, aren't they? (please correct me if I'm wrong about that) But maybe you're just saying that those two types of cells are similar in terms of what C-rate can be safely used with them?
At the same time, this page http://www.electricbike.com/lithium-cobalt-magnanese/ suggests that these cells should be "a solid reliable 3C chemistry, which can provide a continuous 30A from an affordable and compact 10-Ah battery".
:?
SamTexas
1 MW
Your post is acknowledged. I'm going to delay my reply for a day. I want to see what the other fools have to say first.
BTW, this forum board is managed by at least one sleazy and stupid moderator. So don't be surprised to see posts disappearing without a trace, or posts that get modified at his childish whim.
BTW, this forum board is managed by at least one sleazy and stupid moderator. So don't be surprised to see posts disappearing without a trace, or posts that get modified at his childish whim.
^-- only if you throw the F bomb into your posts and try to start a flame war..
Yep, i edited the F bomb out and sent a polite PM explaining why..
Sam reposted the flamey post with the f bomb highlighted and bolded..
I deleted the post..
Aren't i an awful person? :lol:
Yep, i edited the F bomb out and sent a polite PM explaining why..
Sam reposted the flamey post with the f bomb highlighted and bolded..
I deleted the post..
Aren't i an awful person? :lol:
Anyway, if this helps.. attached is a spec sheet for the SDI ICR18650-22P that i found with a google search. I believe these may be the same cells used in the eZee packs ebikes.ca is selling but i am not entirely sure.
http://ebikes.ca/store/store_batteries.php
li-co laptop cells are usually designed with safety in mind... generally the big battery companies design for safety as they have liability issues to think about. It's the no-name China companies and RC Lipo sellers that tend to deserve the suspicious eye
( that being said, i run RC Lipo, but i would not expect the average joe to use it w/o a realistic possibility of an impromptu death metal pyrotechnics show occurring during use. )
http://ebikes.ca/store/store_batteries.php
li-co laptop cells are usually designed with safety in mind... generally the big battery companies design for safety as they have liability issues to think about. It's the no-name China companies and RC Lipo sellers that tend to deserve the suspicious eye
( that being said, i run RC Lipo, but i would not expect the average joe to use it w/o a realistic possibility of an impromptu death metal pyrotechnics show occurring during use. )
pithy
100 mW
So would these be considered "li-co" cells or not? Am I missing something fundamental here, like is "li-co" is basically the same as LiNCoMn? If they aren't, why do y'all keep talking about li-co when these aren't li-co cells? :|
And, there's still the basic question (at least in my mind) of whether or not this pack would get really hot when run at 2C continuous.
I welcome input from all. It's important to me to find out solid info about this battery pack.
And, there's still the basic question (at least in my mind) of whether or not this pack would get really hot when run at 2C continuous.
I welcome input from all. It's important to me to find out solid info about this battery pack.
melodious
100 kW
No. These are NOT LiCo. The chemistry of these batteries adds Mn (Manganese) + Ni (Nickel). From what I've read, they are about as safe as LiFePO4. The discharge curve of LiNMC looks similar to lead. You lose some pep (speed) further down the ride. LiFePO4 has a rather predictably flat discharge curve, then at it's end of discharge it drops off a cliff. I have both LiFePO4 and LiNMC in two different EVs. And they behave on the vehicle just like described.
I'd suggest either keep reading the threads and absorb as much info thrown out here, or visit some links (like batteryuniversity.com) that explains the different battery chemistries in usage.
I'd suggest either keep reading the threads and absorb as much info thrown out here, or visit some links (like batteryuniversity.com) that explains the different battery chemistries in usage.
spinningmagnets
100 TW
pithy, it's really hard to get a fair assessment of any battery chemistry. If it is from a large mainstream supplier, the level of quality should be more consistent, with fewer cells that vary significantly from the average (cheap generic cells vary widely). As far as C-rate goes; the faster you pull amps out of a cell, the hotter the cell will get. That is why the graphs are important.
If you can pull a C-rate of 5C from a given cell, is it a 5C chemistry? Does the cell stay fairly cool, like a 25C-50C "labeled" nano-tech LiPo at a 5C draw? does it get warm? does it get hot? how hot? How hot is hot enough that it degrades the life-cycle? If it gets hot, but never catches on fire...is THAT its true C-rate?
Then, there is the factor of continuous vs temporary burst C-rate. Let me give you an example. I have data-logging on my Castle Creations ESC, and I did several back-to back runs where I accelerated from a stop up to cruise at the drives top-speed. I seem to remember that I was drawing about 40A for a few seconds, then I hit top speed and the amps went down to less than 10A to maintain speed (on flat ground). If I use a battery pack that has a certain chemistry and has the Ah-size to allow a low C-rate from the pack for 90% of the ride, how high a C-rate can I draw for the occasional 5-seconds of acceleration? Twice the continuous C-rate? Three times? A direct-drive hub on an uphill will draw more amps than a comparable geared hub on flat land.
The only true C-rate of a battery pack (chemistry + the Ah-size) is...was this one better than that one...as it relates to your personal user profile. This NCM pack is worth the extra price to "some" builders who highly desire 18-Ah in a small triangle pack. If you don't mind a rectangular pack that is 20-Ah size, you could save a few bucks.
There will be no "solid info" until several users have put some serious miles on these packs, and report back with fairly consistent data. Until then, it's just a guess as to whether this is better than "X" or "Y".
If you shop for Lithium Polymer on some website, you will be getting Lithium-Cobalt-Oxide, which sometimes catches on fire (likely due to a variety of charging/discharging/mishandling issues). If you get NCM, you will be getting Lithium-Nickel-Cobalt-Manganese-Oxide...which is very likely likely to NOT catch on fire if mishandled, and is still very power-dense (small for a given Ah-size).
If you can pull a C-rate of 5C from a given cell, is it a 5C chemistry? Does the cell stay fairly cool, like a 25C-50C "labeled" nano-tech LiPo at a 5C draw? does it get warm? does it get hot? how hot? How hot is hot enough that it degrades the life-cycle? If it gets hot, but never catches on fire...is THAT its true C-rate?
Then, there is the factor of continuous vs temporary burst C-rate. Let me give you an example. I have data-logging on my Castle Creations ESC, and I did several back-to back runs where I accelerated from a stop up to cruise at the drives top-speed. I seem to remember that I was drawing about 40A for a few seconds, then I hit top speed and the amps went down to less than 10A to maintain speed (on flat ground). If I use a battery pack that has a certain chemistry and has the Ah-size to allow a low C-rate from the pack for 90% of the ride, how high a C-rate can I draw for the occasional 5-seconds of acceleration? Twice the continuous C-rate? Three times? A direct-drive hub on an uphill will draw more amps than a comparable geared hub on flat land.
The only true C-rate of a battery pack (chemistry + the Ah-size) is...was this one better than that one...as it relates to your personal user profile. This NCM pack is worth the extra price to "some" builders who highly desire 18-Ah in a small triangle pack. If you don't mind a rectangular pack that is 20-Ah size, you could save a few bucks.
There will be no "solid info" until several users have put some serious miles on these packs, and report back with fairly consistent data. Until then, it's just a guess as to whether this is better than "X" or "Y".
If you shop for Lithium Polymer on some website, you will be getting Lithium-Cobalt-Oxide, which sometimes catches on fire (likely due to a variety of charging/discharging/mishandling issues). If you get NCM, you will be getting Lithium-Nickel-Cobalt-Manganese-Oxide...which is very likely likely to NOT catch on fire if mishandled, and is still very power-dense (small for a given Ah-size).
pithy
100 mW
Thanks for the great info everyone! I really appreciate it!
That's pretty much what I had been thinking about these cells before reading some of the posts in this thread and some other threads that confused the issue (at least for me). I'll check out batteryuniversity.com to try to learn more about all of this.
(note--was editing/typing at the same time as spinningmagnets, so, sorry for the extra edits....)
That's pretty much what I had been thinking about these cells before reading some of the posts in this thread and some other threads that confused the issue (at least for me). I'll check out batteryuniversity.com to try to learn more about all of this.
(note--was editing/typing at the same time as spinningmagnets, so, sorry for the extra edits....)
Popstar
100 mW
spinningmagnets said:
Well said qualifier for why this pack would be desirable for a builder. I do not have quantitative data on my triangle pack performance, but it has wowed me so far with its quality and practicality. My reasons for being happy in choosing this pack over Ping, Sunthing, OSN, Cell_man A123, Headway, DIY with supowerbattery111, or full immersion in LiPo-lifestyle came down to:
1) Battery inside the triangle instead of high above rear wheel maintains proper "bike-feel" balance and allows the rack to be used for cargo haulage. This alone makes me giddy to "put some serious miles on" this pack.
2) Using NMC instead of Lipo allows me to let guests to ride, charge and (hopefully not) crash the bike without worry, while NMC instead of LiFePO4 allows ~2X range in same volume.
3) At this higher 52V, it turns out I don't need a high number of amps (>22A) to go a respectable 20mph up a slight incline, so whether it is continuous 1.5C, 2C, 3C doesn't seem as important as before I got it (when I had only used 36V 12ah headway previously).
4) I dealt with Cell_man on my first build and know he takes pride in his products quality and price. At 14s9p, including assembly, BMS, and shipping to west coast, this ICR18650-22P pack comes to $6.62/cell. This ebay posting is selling the individual cells for $6.80 plus shipping. Paul/Cell_man posted earlier saying he is looking to get even better rated cells into packs once he can secure a good supply and price . The INR18650-20R cells migueralliart built his pack with and that spinningmagnets has been talking about on his site sound like an awesome next step, and at $6-7.50/cell from supowerbattery111, hopefully it is not far off.
So I'd say that this triangle pack is a winner for my needs right now. I doubt anyone who gets one will be disappointed with the EV grin it provides, and though time will tell for the life-cycle performance, I think it will be some fun time indeed...
-Popstar
migueralliart
10 kW
For a person that doesn't want to fiddle with soldering or putting together a pack I think cell_mans 18650 packs are the best plug and play solution. The problem is that you DO pay a premium for it. While a lot of people claim they are 3C + rated a lot of us know for sure they won't sustain it too much. That type of cell sags a lot right off the start.
The INR18650-20R sags too but during the beginning of the discharge they hold a higher voltage. Also these cells WERE designed to be used in power tools where constant on/off high current cycles are common. To me this sounds really close to an ebike application. One of the reviews online on the 20R cells said that the cells kinda like doing 15A and 20A discharge and that the capacity was more at that current (but they also sag more). I don't know but to me a cell as small as the 18650 that can do 15A to 20A wins my respect. So much I now made 2 packs out of em and planning the third.
The smallest pack is a 12S4P pack 8AH that I use on my idrive. It is REALLY small and with my lyen sensorless controller set at 30-40 amps and a 9C from cell_man I get EXCELLENT range and efficiency. Also I've tried running the pack out by doing LOTS of starts and stops and checking its temperature and even this small of a pack stays barely warm to the touch at the end of its discharge which means they are excellent for an ebike application. Not to mention I have them glued one to each other. The seller also has the cell separators which will aid with cooling (not that I've needed it so far).
To give you an idea on how well the cells discharge when I FIRST put together the 12S4P pack I had put a 50 amp fuse in the discharge lead. When I programmed the lyen controller I accidentally used another saved profile that had 55Amps discharge. Going for a test ride the fuse blew in a matter of seconds meaning these 4P bricks were putting out the said current.
The INR18650-20R sags too but during the beginning of the discharge they hold a higher voltage. Also these cells WERE designed to be used in power tools where constant on/off high current cycles are common. To me this sounds really close to an ebike application. One of the reviews online on the 20R cells said that the cells kinda like doing 15A and 20A discharge and that the capacity was more at that current (but they also sag more). I don't know but to me a cell as small as the 18650 that can do 15A to 20A wins my respect. So much I now made 2 packs out of em and planning the third.
The smallest pack is a 12S4P pack 8AH that I use on my idrive. It is REALLY small and with my lyen sensorless controller set at 30-40 amps and a 9C from cell_man I get EXCELLENT range and efficiency. Also I've tried running the pack out by doing LOTS of starts and stops and checking its temperature and even this small of a pack stays barely warm to the touch at the end of its discharge which means they are excellent for an ebike application. Not to mention I have them glued one to each other. The seller also has the cell separators which will aid with cooling (not that I've needed it so far).
To give you an idea on how well the cells discharge when I FIRST put together the 12S4P pack I had put a 50 amp fuse in the discharge lead. When I programmed the lyen controller I accidentally used another saved profile that had 55Amps discharge. Going for a test ride the fuse blew in a matter of seconds meaning these 4P bricks were putting out the said current.
Supertux1
100 W
I'm getting one of cell_man's new triangle packs delivered today for my MAC 10T commuter build.
I'm also getting a CAv3 and I have the Analogger GPS module as well, so I'll be able to collect some real time data about these packs and maybe we can draw some conclusions. I'll have temperature data from inside the motor as well.
Is there any sort of test to determine the efficiency/capacity of a pack/motor/controller combination that doesn't involve actually riding it in an uncontrolled environment? I know that theres is the ebikes.ca simulator, but I was wondering there was an actual experiment I could do in my garage with the real hardware, the data of which could then be scaled by some factor to predict 'real world' performance? Something everyone can do easily. Maybe the raw numbers won't mean anything absolutely but maybe relatively. WOT until LVC with the wheel off the ground? I don't know....
cell_man puts an extra cell in his 14s9p NCM pack to compensate for the voltage sag and markets the pack as 18.5Ah even though the cell data suggests it _could_ be 19.35Ah under ideal discharge conditions. I liked how straight forward he is about the cells and their expected real performance, and adding the extra cell to compensate for the discharge curve to make it better earns him my money.
My needs are simple, a commuter build for 10 miles of range for work on the weekdays and on the weekends towing a cart with a pair of tired dogs.
I decided to go with this battery because it would fit in my frame and balance the bike well, freeing up the front and rear for storage bags for work stuff. I have no illusions about the C rate or life cycle. I plan to be very gentle with the battery and draw at 1C or less, most of my commute is flat and it is not legal to ride with 'the power unit in operation' on our bike paths ($20 fine), so some pedaling will be involved. Might test the law on that one once the motor breaks in and quiets down.
I regard this pack as good for basic, 250-1000W commuting with geared hubs / assist but not much beyond that. It's okay for my needs, but I don't think you'd really want to pair it with a big DD motor if you care about performance or lifecycle.
I'm also getting a CAv3 and I have the Analogger GPS module as well, so I'll be able to collect some real time data about these packs and maybe we can draw some conclusions. I'll have temperature data from inside the motor as well.
Is there any sort of test to determine the efficiency/capacity of a pack/motor/controller combination that doesn't involve actually riding it in an uncontrolled environment? I know that theres is the ebikes.ca simulator, but I was wondering there was an actual experiment I could do in my garage with the real hardware, the data of which could then be scaled by some factor to predict 'real world' performance? Something everyone can do easily. Maybe the raw numbers won't mean anything absolutely but maybe relatively. WOT until LVC with the wheel off the ground? I don't know....
cell_man puts an extra cell in his 14s9p NCM pack to compensate for the voltage sag and markets the pack as 18.5Ah even though the cell data suggests it _could_ be 19.35Ah under ideal discharge conditions. I liked how straight forward he is about the cells and their expected real performance, and adding the extra cell to compensate for the discharge curve to make it better earns him my money.
My needs are simple, a commuter build for 10 miles of range for work on the weekdays and on the weekends towing a cart with a pair of tired dogs.
I decided to go with this battery because it would fit in my frame and balance the bike well, freeing up the front and rear for storage bags for work stuff. I have no illusions about the C rate or life cycle. I plan to be very gentle with the battery and draw at 1C or less, most of my commute is flat and it is not legal to ride with 'the power unit in operation' on our bike paths ($20 fine), so some pedaling will be involved. Might test the law on that one once the motor breaks in and quiets down.
I regard this pack as good for basic, 250-1000W commuting with geared hubs / assist but not much beyond that. It's okay for my needs, but I don't think you'd really want to pair it with a big DD motor if you care about performance or lifecycle.
migueralliart
10 kW
Just want to clarify something here;
Putting an extra cell without making changes to your programmed controllers doesn't mean JOLT in the end here's why;
At a higher voltage your 100% throttle will yield a HIGHER speed and also a higher CURRENT DRAW. The ONLY way to counteract your so called sag is to compensate throttle percentage to limit the max speed.
Take for example the MAC SPEED from ebikes.ca with a 26" rim, a 220lb rider and a 40 amp controller.
Assume you have a 13S pack 18Ah
13*3.6V=46.8V
you get;
Max speed - 28.9mph
Current Draw - 25.8 amps
C rate pulled - ~1.43C
Now lets do the math for a 14S pack 18Ah
14*3.6v = 50.4V
Max speed - 30.6mph
Current Draw - 28.1 amps
C rate pulled - ~1.56C
You see now that you went from a lower C rate (BETTER) to a higher C rate (WORSE)
The only way to compensate is to set the throttle to what the 13S max speed is
92% throttle set at the max speed setting on a 14S pack you get;
Max speed - 28.9mph
Current Draw - 23.7 amps
C rate pulled - ~1.32C
Adding some more...
This calculation is at top speed. If your riding style is more stop and go and you stay below the 25mph zone then having that extra cell will help since your total power will be more due to the added voltage (which isn't really that much and could be overwhelmed by the controller max current setting).
Putting an extra cell without making changes to your programmed controllers doesn't mean JOLT in the end here's why;
At a higher voltage your 100% throttle will yield a HIGHER speed and also a higher CURRENT DRAW. The ONLY way to counteract your so called sag is to compensate throttle percentage to limit the max speed.
Take for example the MAC SPEED from ebikes.ca with a 26" rim, a 220lb rider and a 40 amp controller.
Assume you have a 13S pack 18Ah
13*3.6V=46.8V
you get;
Max speed - 28.9mph
Current Draw - 25.8 amps
C rate pulled - ~1.43C
Now lets do the math for a 14S pack 18Ah
14*3.6v = 50.4V
Max speed - 30.6mph
Current Draw - 28.1 amps
C rate pulled - ~1.56C
You see now that you went from a lower C rate (BETTER) to a higher C rate (WORSE)
The only way to compensate is to set the throttle to what the 13S max speed is
92% throttle set at the max speed setting on a 14S pack you get;
Max speed - 28.9mph
Current Draw - 23.7 amps
C rate pulled - ~1.32C
Adding some more...
This calculation is at top speed. If your riding style is more stop and go and you stay below the 25mph zone then having that extra cell will help since your total power will be more due to the added voltage (which isn't really that much and could be overwhelmed by the controller max current setting).
Supertux1
100 W
migueralliart said:
Of course, more voltage pushes more amps through any given load/resistance resulting in more wattage / work done.
I plan to have my controller driven by the closed loop throttle of the CAv3, regulated by legal speeds and/or wattage with a 1C limit to start; the throttle percentage the controller sees is going to be whatever it's going to be regardless of the pack's voltage.
I did some calculations and the maximum (loaded) speed of my 10T MAC in a 26" wheel at 50.4V is 27.59 mph.
At 42 Volts LVC (3v x 14s) that's around 23 mph.
I will probably set the CAv3 conservatively at 20mph until I get a better feel for the cop interest here.
Since my legal speed is 20mph or 30mph in my state depending on if the bike is commercial electric or home built gasser (not real clear),
I think that the this pack and the protection that the CAv3 should do quite nicely for my needs.
ebikes.ca simulator does not account for the IR of the battery.
IR goes down the more AH you add.
For example, if i have a wimpy 10AH battery pack and i have a 30A load, let's say i am getting 0.5v per cell of sag.
If i have a 20AH battery pack with the same cells and a 30A load, i will get 0.25v per cell of sag.
If i have 40AH, i will get 0.125v per cell of sag.
Don't use the ebikes.ca simulator until you know the IR ( in mOhms ) of the pack.
Unfortunately this is a stat that few, if any, battery sellers publish.
IR goes down the more AH you add.
For example, if i have a wimpy 10AH battery pack and i have a 30A load, let's say i am getting 0.5v per cell of sag.
If i have a 20AH battery pack with the same cells and a 30A load, i will get 0.25v per cell of sag.
If i have 40AH, i will get 0.125v per cell of sag.
Don't use the ebikes.ca simulator until you know the IR ( in mOhms ) of the pack.
Unfortunately this is a stat that few, if any, battery sellers publish.
cell_man
100 kW
migueralliart said:
Adding another series cell does make a difference. Firstly there is the additional energy from that additional 1 series group, so you have an extra 1/13 more energy than you would with 13S, it's approx 7.7% additional energy (14/13). It means the 14S is more in line with a 16S LiFePO4 pack, it also means you are making better use of the typical controllers we supply witth such kits, including 75V fets and 63V caps. Seeing as 13S, based on 3.6V per cell is equal to 46.8V and 14S based on 3.6V per cell is equal to 50.4V, it is pretty safe to say that the voltage is going to be 48V plus for more of the time with a 14S pack than a 13S Pack. I fail to see how providing a pack that gives you a higher voltage and also higher energy is a negative thing. I don't overstate the Capacity, I don't overstate the nominal Voltage and I recommend usage at a little over 2C Max.
I also don't understand the reasoning behind how a lower voltage and therefore slower bike is an improvement. You have the ability to go faster with that higher voltage pack, but that is not how many/most people use their bike. If you have a commute, you might settle on a certain speed and stick with it, adjusting throttle as required. So long as you are using less than full throttle, a slightly higher voltage pack will continue to allow you to maintain a higher speed until the packs discharge voltage drops down and slows your progress.
cell_man
100 kW
I'd also like to add that building battery packs, especially triangle packs takes a lot of time and effort, at least the way we do it, it does. It seems that all we need to do is pull some cells out of the box, string them together with some skinny strips of nickel and drop them in a carton for shipping. If everybody is happy to do all the hard work, testing, fully charging every cell before building the pack so it is balanced from day 1, please let me know, I'll save myself a load of work and make money for doing a whole lot less. Putting the pack together, preparing terminations, fitting BMS, including a PTC thermal fuse on every wire, takes time, doing triangle packs takes a fair bit more time on top, including lots of irregular shaped terminations and copper reinforcements where terminations are long, with fairly narrow cross section. After a discharge test on every pack to confirm it meets the expected capacity, they are finished off and adding the heat-shrink, charging to 50%. The heavy black Heat Shrink we use, can cost up to 7USD per m for the the widest stuff we use, it's not as much as that for the smaller sizes but it's not cheap. We need to buy a 25m roll (up to 175USD per 25m roll on the biggest stuff we stock, I kid you not) and we keep a lot of different sizes. When you look at that triangle pack, yes we do add some additional cost over a regular pack, it's about 50USD premium over a regular rectangular pack on the 14S 9P, then we add the bag on and that adds another 25USD (reduced from our standard price of 35USD) to the Total. We had to custom build those bags and order 500pcs MOQ and it took about 2-3 months to get them arranged and built. I don't see anyone else offering bags as good, at the same or similar price, or anything significantly better for a good deal more.
I originally decided to offer the SDI 22P cells because Justin was so impressed with them and I was tired of the supply issues we were having with A123. I think we made the right choice to move to a higher energy cell type, it certainly makes it much easier to fit a good sized pack on a bike, that still feels like a bike.
Just for info, if I was gonna do a regular finished pack from the various cells we now offer or will be offering in the near future, the retail price per cell price would work out to as below, for price comparison. This would be a finished rectangular rectangular pack (not triangle pack), with everything completed, but without a BMS (for comparison purposes only, I would never supply a pack without a BMS unless specifically requested), a BMS is typically about 50USD extra for a 40A continuous 10S-17S model:
Cost (USD)
ICR18650-22P 4.71USD/cell
INR18650-20R 6.40USD/cell
INR18650-29E 7.65USD/cell
Regarding shipping, I do not ship batteries by regular post, so if some ebay sellers, illegally ship cells that way, it is hardly a fair comparison, we ship them as batteries, costs are competitive, get a quote on my site, select DHL Battery for anything including batteries, see here for shipping info:
http://em3ev.com/store/index.php?route=information/information&information_id=6
If all that is required is some hot glue ( we use RTV, hot glue is not good enough) and a few strips of nickel on untested and unbalanced cells, I'm sure I could do a fair bit better than the above and if quantities are small business quantities, better prices again. I only buy from reliable, proper sources and guarantee that these cells are A grade, not in any way grey market. We have a good quality cell testing rig that can test 64 cells at up to 6A charge/discharge (not the typical 2A that the cheap test unit units can do) and a powerful spot welder that we bought last year to replace the previous unit we had been using, it can weld most things, easily welding 0.2mm copper to 0.2mm nickel.
So am i really so expensive, or do I just offer more? I'll leave it up to the viewers to decide.
I originally decided to offer the SDI 22P cells because Justin was so impressed with them and I was tired of the supply issues we were having with A123. I think we made the right choice to move to a higher energy cell type, it certainly makes it much easier to fit a good sized pack on a bike, that still feels like a bike.
Just for info, if I was gonna do a regular finished pack from the various cells we now offer or will be offering in the near future, the retail price per cell price would work out to as below, for price comparison. This would be a finished rectangular rectangular pack (not triangle pack), with everything completed, but without a BMS (for comparison purposes only, I would never supply a pack without a BMS unless specifically requested), a BMS is typically about 50USD extra for a 40A continuous 10S-17S model:
Cost (USD)
ICR18650-22P 4.71USD/cell
INR18650-20R 6.40USD/cell
INR18650-29E 7.65USD/cell
Regarding shipping, I do not ship batteries by regular post, so if some ebay sellers, illegally ship cells that way, it is hardly a fair comparison, we ship them as batteries, costs are competitive, get a quote on my site, select DHL Battery for anything including batteries, see here for shipping info:
http://em3ev.com/store/index.php?route=information/information&information_id=6
If all that is required is some hot glue ( we use RTV, hot glue is not good enough) and a few strips of nickel on untested and unbalanced cells, I'm sure I could do a fair bit better than the above and if quantities are small business quantities, better prices again. I only buy from reliable, proper sources and guarantee that these cells are A grade, not in any way grey market. We have a good quality cell testing rig that can test 64 cells at up to 6A charge/discharge (not the typical 2A that the cheap test unit units can do) and a powerful spot welder that we bought last year to replace the previous unit we had been using, it can weld most things, easily welding 0.2mm copper to 0.2mm nickel.
So am i really so expensive, or do I just offer more? I'll leave it up to the viewers to decide.
jonescg
100 MW
cell_man said:
Paul, you could offer these packs for 20% less than cost and give tickets to the movies, and some folks will STILL complain that it's too expensive!
I call it "spoilt customer syndrome" and the double edge sword that is globalisation is entirely to blame. Nowadays, you can walk in to a big box hardware store and buy a $20 cordless drill. You know it's not as good as the $200 drill on the top shelf, but it's still a drill right? So when it breaks, and your second one breaks, you go shopping for a good drill - and scoff at the $200 price tag. Yes people, quality costs money!
spinningmagnets
100 TW
Paul, if you offer these packs at cost, just to be a good guy...how long before you stop offering these packs and spend your time on products that help you stay in business in this competitive industry?
If anyone thinks a reliable product that is offered by someone who has taken years to establish a good customer service reputation is just too expensive, then...perhaps this is an opportunity for you to make a spot-welder for $100 and start buying cells in bulk and become a competitor in pack building?
I say: put up or shut up, nobody is forcing anyone to buy packs from cell_man/em3ev. Competition is good, and if you think you can do a better job for less, then you can make a lot of money making packs for sale. The new Samsung 20R cell looks really good (Migueralliart), so...what's stopping anyone?
I want Paul to make a profit, and I want him to stay in business for a very long time.
If anyone thinks a reliable product that is offered by someone who has taken years to establish a good customer service reputation is just too expensive, then...perhaps this is an opportunity for you to make a spot-welder for $100 and start buying cells in bulk and become a competitor in pack building?
I say: put up or shut up, nobody is forcing anyone to buy packs from cell_man/em3ev. Competition is good, and if you think you can do a better job for less, then you can make a lot of money making packs for sale. The new Samsung 20R cell looks really good (Migueralliart), so...what's stopping anyone?
I want Paul to make a profit, and I want him to stay in business for a very long time.
migueralliart
10 kW
cell_man said:
With a lower capacity pack having more voltage and going at a higher speed means worst sag. Specially 18650 cells. I did say I assumed a high speed cruising riding style. This is definitely not the case with LIPO or for a big capacity pack that is close to the current draw while cruising at say 25mph+.
But I agree in part with you that there is extra energy in the pack. It just sounded like having that extra cell fixes the problem when the reality is that people go for 100% throttle and by having that extra cell it will take em to a higher speed.
migueralliart
10 kW
flabby said:
If you search under my name you'll see a thread named INR18650 yes these cells can punch.
I go into details on how to build your pack and where to get the cells. The seller gives a good price considering they do the spot welding (pretty well) and arrange your pack into bricks of any size you want. The only drawback is that you might need to order 5S or 6S groups since they can't ship a bigger pack.
I don't run with a BMS. I could never trust a component like that when all I threads about BMS failures every week. But to each its own.
migueralliart
10 kW
spinningmagnets said:
+1 .. everyone is entitled to buy whatever they want from whomever they want to buy it from.
cell_man
100 kW
I'm sorry I think it is a bit of an academic point to say that it is better to have lower voltage so that you can then go slower and have less sag. By that point why don't we all just use 36V and go at everywhere at 15mph, there's be no sag and the range would be huge. IMO there is nothing wrong with a slightly higher voltage to keep the battery at or above the magic 48V (or whatever voltage) that users would be basing the motor speeds around. On faster ebikes, 25, 30 or more mph, I don't believe most users go everywhere at full throttle, all of the time. My reasoning is that they can do the max speed they expect (or more) for maybe 70 or 80% of the discharge instead of maybe 20 or 30% with 1 less series cell. I believe this is the best compromise and that is why I call 60V a 17S, 72V a 21S. Ok, I will call them 50V, or 62V or 75V, there is nothing wrong with that, but the intention is that the user has a descent voltage for most of the discharge. You can't get over the drop in voltage as it discharge, it comes with the territory (unless you use A123), but at least if it starts that bit higher, it stays above the magic number for longer, there is a throttle to modulate speed when on a fresh pack. The pack is recommended for no more than a recommended current level, so as long as the controller is that or better a little lower, there is no problem. If the requirement was a very small high power pack, i wouldn't suggest the 22P cells. When I get the 20R cells I'll suggest those in that application.
I will do some cell tests and post up some comparisons. I don't want to call out anybody, but I have tested large batches of 22P cells, they do not perform as badly as the below listing suggests and I don't know what the test data at 3 and 5A are supposed to be showing:
http://www.dampfakkus.de/akkutest.php?id=73
Just because something is supposedly independent and I am sure it is independent, but it doesn't mean it is accurate. Are those graphs suggesting that a 22P cell delivers only about 60% of it's capacity at 3A? I hate to disappoint, but if that is what they are stating, they need to get some better test equipment, that is absolutely not true. I looked at test data for other cells on that site and most of them looked surprisingly bad at moderate current levels.
The data on Justin's packs shows some curves for finished packs using the SDi 22P cells, in known configurations, so from that the discharge curve for a single cell could be inferred:
http://www.ebikes.ca/store/diagrams/B3614LiM-EZ%20Battery%20Manual,Rev%201.1.pdf
Off the top of my head Justin's data approximately agrees with what I have seen with individual cell tests we've done. I have the equipment and obviously everyone seems to be very interested, so I will make an effort to get some discharge curves made at various current levels when I get a chance.
I will do some cell tests and post up some comparisons. I don't want to call out anybody, but I have tested large batches of 22P cells, they do not perform as badly as the below listing suggests and I don't know what the test data at 3 and 5A are supposed to be showing:
http://www.dampfakkus.de/akkutest.php?id=73
Just because something is supposedly independent and I am sure it is independent, but it doesn't mean it is accurate. Are those graphs suggesting that a 22P cell delivers only about 60% of it's capacity at 3A? I hate to disappoint, but if that is what they are stating, they need to get some better test equipment, that is absolutely not true. I looked at test data for other cells on that site and most of them looked surprisingly bad at moderate current levels.
The data on Justin's packs shows some curves for finished packs using the SDi 22P cells, in known configurations, so from that the discharge curve for a single cell could be inferred:
http://www.ebikes.ca/store/diagrams/B3614LiM-EZ%20Battery%20Manual,Rev%201.1.pdf
Off the top of my head Justin's data approximately agrees with what I have seen with individual cell tests we've done. I have the equipment and obviously everyone seems to be very interested, so I will make an effort to get some discharge curves made at various current levels when I get a chance.
