Motor / Battery Options for Dual Sport Build

[SteveHare]

Hello Everyone,

I am planning on building a dual sport electric motorcycle and am trying to find different motor / battery options that are available.

GOALS:
Power: 80ft/lb at the rear wheel, 40-60 hp, Minimum of 1 hour hard on/off road riding, curb weight around 250lbs
Price: materials that meet my needed goals at the best price/quality!

Motor:

I would like to do a 6:1 total gear ratio and the goal is a final rear wheel torque of 80ft/lb.

Meaning I need a motor that can produce a continuous 20Nm if my calculations are correct.

Also I would like to find a motor that will produce 30Kw, putting me around 40 hp.


I am looking at the Heinzmann PMS 150 price is $2,500

Link to Specs:
https://www.heinzmann.com/jdownloads/electric-and-hybrid-drives/CAT_Electric_Drives_Product_Catalogue_e.pdf

I believe I can get the desired torque of 20Nm and 30Kw out of this with a 96V 150Ah battery pack.

The price is not terrible for the motor but I do not necessarily need the Disc style and am wondering if I am paying for this technology.
However, I do like the benefits of the axial flux disc configutation.

Another option was Ashwood motors I am waiting on a response from them on any offerings they would have.

I had contacted Yasa, but the price was well beyond my means.


Does any one have any other motor options, or is there something I am missing in my thinking about the motor?

Here are some links to other motors I looked at and the reason I didnt think they would be suitable
Underpowered - http://www.asmokarts.com/index.cfm?pageID=24
Need to contract but probably our of price - https://emrax.com/e-motors/emrax-188/
Need to contact but probably out of price - https://avidtp.com/product/evo-motors/
(My ideal price would be below $1,000 on the motor)


Battery:

What seems to be reasonable for a battery is something that puts out 96V and 150 Amps.
Making that battery, and making it fit into a 250 lbs bike is very difficult.
Looking for any and all advice on battery manufacturers I can use for this build!
What seems to make the most sense is choosing a motor first then building a battery to fit

 

[amberwolf]

Battery:

What seems to be reasonable for a battery is something that puts out 96V and 150 Amps.
Making that battery, and making it fit into a 250 lbs bike is very difficult.

there's a few ways of making a battery taht can do that.

but:

Minimum of 1 hour hard on/off road riding,

at "30kw"... even if you only use half that power most of the time, that's a 15kwh pack. this is very large.

i have about a 2kwh pack, of eig cells like the zero motorcycle used several years ago, and it is close to 40lbs (14s2p), no casing just cells and main battery cables. is the size of good stack of hardback books. seven times that, is....a lot. the pack itself would weigh as much as your whole bike.

even if you only used a quarter of the power most of the time...that's still a 7kwh+ pack, and will still weigh well over 100lbs and be very large. size of an average human torso, perhaps.

i don't know how much power you will *actually* use, how many wh/mile. you may wish to see what other bikes out there have similar power levels to what you propose, and see how long their batteries last, rangewise, or how many wh/mile they get, in the usage scenario you want. then you'll have a better idea of what capacity you will have to have to do what you want.



so, the capacity needed (at the power level needed) is going to make a huge difference to the physical size of the pack, and the type of cells you end up using. (some cells can output a lot of power, but aren't very high capacity...others have a lot of capacity but can't output much power, and some compromise between the two...others don't last very many cycles but can make tremendous power while they live).

also, do you want this to last a long time, or are you willing to replace the pack "often", dpeending on how hard you use it (because there are solutions like rc lipo that can do awesome things...for a very short lifetime).

you will also have to define the actual space you have available for the battery...or else be willing to put at least some of it in "saddlebags", as well as in the frame around the motor and such.

if you were willing to go hubmotor directly in the rear wheel, you could probably get significantly more room for battery inside the frame, depending on the frame design (or what you're willing to do to modify it).

 

[SteveHare]

Amberwolf:
Thanks for the feedback.
My thought process was that if the max Amps provided is 150A, and the pack has a 150AH rating with the regenerative braking in city and off road riding I would get an hour or so of hard riding out of the battery.


Here are the numbers I was able to gather based on your recommendations

Manufacturers:
Alta Motorcycles is a 7kwh battery
It looks like for the DS Zero does a 14kwh and 7 Kwh battery pack
The KTM Electric Freeride has 3.9 Kwh battery
Surron has a 2Kwh battery pack
Cake Kalk OR 2.6 kwh
Electric Motion 6kwh


Forum Builds:
Silent Enduro: https://endless-sphere.com/forums/viewtopic.php?f=10&t=102858
4.8kwh, 3.5kwh modded sur-ron, 4.5Kwh, 4.2Kwh



My Initial though was a 100V 150Ah battery which would be 15Kwh (if my math is correct)
Seeing the numbers from the manufacturers and builders I am wondering if I am being a bit ambitious in size.
As you stated a 7Kwh would be human torso size, I think that is probably of a reasonable size, anything bigger and I wont meet my weight goals.

Also if i think i should look more at discharge characteristics.

I would prefer not to change batteries very frequently, I used worked as a motorcycle mechanic and I learned I care much less about maintaining my own bikes then other peoples! However, I am interested in reading spec sheets on the RC Lipo batteries, do you have any recommendations on specific models?

So I would like to do something that once the bugs are worked out will be relatively painless in maintenance.

Not a fan of hub motors personally.


Guessing at available dimensions:
20cm x 22cm x 60cm

 

[amberwolf]

However, I am interested in reading spec sheets on the RC Lipo batteries, do you have any recommendations on specific models?

no, i havent' kept up with them since i started using the eig cells (which are probalby not power or energy dense enough for your use).

however, there are some similar lipo cells (but with better quality control ) that people like jonescg have used to build racing motorcycle/etc packs. some of his posts about them:
https://endless-sphere.com/forums/search.php?keywords=lipo&terms=all&author=jonescg&sc=1&sf=all&sr=posts&sk=t&sd=d&st=0&ch=300&t=0&submit=Search

 

[SteveHare]

Thanks!!
Very helpful I have been reading up a lot on battery options and trying to figure out how to get the the numbers I need.
One thing I can seem to wrap my head around is voltages.
Let’s say I use a battery to build my pack that operates between 4.2v and 3.2v what voltage do I use to calculate my pack voltage? I’m guessing you base this off the normal operating voltage?
So if I want to build a 72v pack with 50ah out of cells that are 3.7v and 8ah I would do a 20s 6p would be 72v 48ah?
Part of what confuses me is I see some people say they run 72v with 22s pack. I understand different designed batteries have different operating voltages but it seems like 90% of what people use operate at 3.7v so am I doing my math wrong?
Thanks!

 

[amberwolf]

Let’s say I use a battery to build my pack that operates between 4.2v and 3.2v what voltage do I use to calculate my pack voltage? I’m guessing you base this off the normal operating voltage?

yes, it's the "nominal" voltage, which for the eig cells are 3.65v, and for many 18650 /etc cells is 3.7v, and some it's 3.6v. usually the max voltages for these cells would be 4.15v, 4.2v, and 4.1v, respectively.

So if I want to build a 72v pack with 50ah out of cells that are 3.7v and 8ah I would do a 20s 6p would be 72v 48ah?

yes, usually 20s is a "72v".

keep in mind that it's not just capacity you have to satisfy, it is c-rate, or current delivery ability with minimal voltage sag. (more sag means less actual power at teh wheel, *and* more heating inside the battery). often when using less capable cells you have to have a much larger capacity pack than actually needed for range just so you can get the oomph out of it you need to do the work your motor has to do.


if you're limiting to 150a, then 150 / 6p is 25a per cell. that's a lot of current...but it's not as bad as your originally-proposed situation would've been:

if you need the motor to do 30kw, the pack has to put out more (say, 20% more to account for losses), so 30kw * 1.20 = 36kw. to get amps it's 36kw / 72v = 500 amps nominal. if you only have 6p, that means each cell must be able to sustain 84 amps.

that's a hell of a lot of current for a single cell to sustain. even half that, 42 amps, is a lot; the cells are probably going to sag a lot and get hot inside, unless they're really made for performance (like rc lipo or similar), in which case they may not have a long lifespan (you can't have high capacity, high current delivery and long lifespan all in one cell, not yet anyway...usually you only get to choose one, at most two...every requirement you add makes more compromise).

for instance, the eig nmc cells i'm using are rated for 100a sustained, 200a burst for a few seconds, but they are big 20ah cells. so each cell isn't really handling that much, relative to it's size (capacity), which means it's c-rate is lower (5c).

for an 8ah cell, 84a is a little more than 10c. i'm sure a lot of cells claim they'll take that...but if they don't provide verifiable discharge and lifetime curves to show the voltage sag for the one and aging / loss of function for the other, you can't really know what you're getting into.

but you can look at whatever ri / internal resistance they show for a single cell. it'll be in the single digit or less milliohms per cell for a high-c-rate cell. iirc, my whole 14s2p pack is about 35mohm or so, which means 35 / 14s = 2.5 mohm per 2p group, so 2.5 * 2p = 5mohm per cell. im sure they were better than that when they were new, almost a decade ago. :lol: iirc the spec sheet gives around 3mohm, maybe less. but i'd be looking for 1mohm or less per cell if you really want performance from them at high currents.


let's say for my cells, at 2.5mohm per group, and 200a continuous current (the max the cells can take). that's 200a * 0.0025ohm = 0.5v drop per cell. in a 14s pack, i'd see 7v of sag, so from a full charge of almost 58v i'd see a drop down to 51v.

what that means for losses is that if i would've seen 58v * 200a = 11,600w, for 51v * 200a i'd only see 10,200w. meaning, i'd be losing 1400w as heat inside the battery itself, just from voltage sag / internal resistance. that doesnt' count any of the cell-interconnect resistances, either.

but i the cells were only 1mohm each, then id' only have 0.5mohm per group, so there'd only be 200a * 0.0005ohm = 0.1v of drop in each group, so only 1.4v drop across the whole pack. so that's only 58v - 1.4v = 56.6v. 56.6v * 200a = 11,320w, so i'm only losing 280w as heat inside the battery pack. that's *way* better, because it means i get 1400w - 280w = 1120w more at the controller than i did with the higher resistance cells. and i don't have to worry about the cells getting damaged from heat, so i don't have to invent a cooling system for the pack....etc.

if you were to use something like a common 18650 (which don't come in 8ah, but just to illustrate...), the internal resistance may be 100mohms...which is a hundred times higher than what you want to be using. if you had a 6p pack of those, it means each parallel group would be 16.7mohms. times 14s (for comparison) is 234mohm for the whole pack. at 200a, that'd be 200a * 0.234ohm, = 46.8v. meaning, your 58v pack would sag down to 11.2v!!!! and you would be wasting 200a * 46.8v = 9380w of heat inside the battery pack!!! so it would probably catch fire pretty quickly, if it werent' that the controller would just shut off as soon as the voltage sag hit it's lvc point.


so....get really good cells (which probably won't be cheap).

Part of what confuses me is I see some people say they run 72v with 22s pack.

i have no idea what their reference points are...but that sounds like they are talking about it's lvc, where it shuts off to protect from overdischarge.

a 22s pack would normally be nominally 80v to 82v, depending on the cells used.

 

[SteveHare]

That was a ton of info!
I had to read twice to get everything, thank you for sharing your experience/knowledge!

I’m gonna got my feet wet with this build knowing it won’t be perfect and perfected yet but I want to use it as an opportunity to learn and progress with this stuff.

Qs motor 138 is on its way along with a em-150 controller. Price on these products is great.
Longevity wise the controller probably isn’t the best option but I have limited my expectations on this build and am going for experience and knowledge rather then performance and permanence.

I went ahead and ordered some SPIM08HP cells.
I am sure you know about them.
3.7v 8ah 200 amp continuous, most the controller will pull is 150 amps so that would be 3c on a 50ah pack
Only 10kw but again I need realistic goals! Also these cells were super affordable at only $2.77 per cell.

They are older cells most likely and they sag a bit. From some people on this form it seems they are getting some pretty mixed IR results with these.
I’m gonna go for a 100v 50ah pack so 20s 6p

Any recommendations on easy cell testing for a novice?
I figured once I get them to do standing voltage and Ohming them out (does that provide an accurate reading for IR?)

Then do a couple charge and discharge cycles and try to monitor them to understand their performance then put at storage voltage for pack building.

As far as pack building I will use some foil connectors and nickel strips and copper bus bars.
Nice thing about the foil connectors is they have a crimp portion for a BMS wire. Should make putting these cells together a lot easier and they are not to pricey at .60

Also have the donor bike too. Will not be a true dual sport because not street legal but again experience.

Here’s the bike it’s a kx125 that has a blown motor. Theses were crazy fast two stroke bikes that have a great track record. Hopefully I can do this bike some justice.

 

[amberwolf]

Any recommendations on easy cell testing for a novice?

i'd look at the threads that discuss those specific cells and see what tests they did on them. if they dont' say, ask for specifics of their cell testing methodology.

I figured once I get them to do standing voltage and Ohming them out (does that provide an accurate reading for IR?)

if by "ohming them out" you mean using the ohms or resistance testing setting on a multimeter..you'll damage or destroy your meter putting it across a voltage source when set to those.

there are several ways of testing ri (oddly, it isn't really "ir", as it's notated as "resistance, internal", capital r, subscript of i, so "ri" would be closer to correct...but many will still know what you mean either way). but there are also two versions of internal resistance, usually listed as acir and dcir.

acir is done at a particular frequency, and can *only* be compared to someone else's test if the frequency used is the same. so it's not that useful since hardly anyone bothers to say what frequency they tested at. 1khz is common but not universal.

dcir is more useful, because the measurement is the same regardless (though methods of measuring and equipment variance can make a great deal of difference in results). you setup a measured known constant resistance across the cell, a load, that will cause a significant current, preferably somewhere within the middle of it's current capability (unless you need to know what it would be at the max current). you measure the current that is flowing, and the voltage, and use all those with ohms law to determine the cell resistance.

to really know what it is you'd need a bunch of resistances, or a variable one, and test at many points along the current-capability-curve, as it is different at higher or lower currents, and different at higher or lower states of charge...so you end up with a resistance curve (or set of them).

but most people doing diy testing will probably just test with a single resistance at the middle of the state of charge and the middle of the current capabiliyt, for an average dcir.



once you have the dcir for each cell, you can match them up for parallel groups so that you have approximately equal total dcirs in each group, so no one group will sag much more than others under load.

you can also test for capacity and match them that way...but it is likely that this will follow the dcir fairly well. (high dcir but low capacity, low dcir but high capacity, etc).