Heavy cargo hauler project

The slower winds, when climbing, seem to draw less power
Click to expand...

Dogman, would this be because the slow wind on a hill can be at full-throttle? I believe a faster wind (same voltage) would be at half-throttle? (would a fast winding at full throttle on a hill suffer more heat than at half-throttle?)

For a non-hub, it seems a fast winding at a high voltage that is mechanically geared down works well, because the gearing allows the motor to stay at an efficient higher RPM at the slower speed...
 
Don't quote me on this ;) but I think the reason they heat up less is becuase they draw less current, which is because during acceleration, they reach sooner the levels of BEMF that cancel out most of the input voltage, compared to the "faster wind" motors.

Meaning, they reach those levels at a slower speed, so they can survive longer at lower speeds because the current flowing into them is lower.

That's my theory, anyway, without being able to actually test this (as I don't have but the 2807).
 
There are a number of factors at play here.

Here is one simple one that is interesting. Initially the throttle controls PWM. So over the range of the throttle you have 0 to 100% of PWM.

With a "high voltage" motor wind (aka torque wind) the greater resistance and greater back EMF from the winding allows more throttle range to be effective before "current limit" kicks in and clamps the PWM. You don't get more PWM above the clamp that's preventing overcurrent. The rest of the throttle range is the same. This changes as you speed up, and more of the throttle becomes effective again as the back EMF limits current and the controller doesn't have to step in.

With a "low voltage" motor (aka fast or speed wind) the motor's lower impedance and lower back EMF cause current to soar and kick in the PWM limiting sooner at a lower throttle setting, so the throttle is "touchy" and only a smaller portion of it is useful.

So the "high voltage" motors have better throttle control. Not surprising since they have less speed for the voltage, and the throttle is essentially a voltage control.


This cargo ebike is about double the weight of a normal ebike (system weight) so if you use two motors (one driving each wheel) you should have better results.
 
Trying to express it even simpler, If x watts is only going to get you to 8 mph up a steep hill, then pick a motor that will be in its efficient rpm at 8 mph rpm. One that isn't happy till 15 mph rpm, is the wrong choice.

What actual motor this is, depends on voltage chosen, and rpm in your chosen rim size at that voltage. The motor only wants to be in it's efficeint rpm range.

Others win with another approach, big motor, big wattage, fly up the hill fast enough to keep the motor happy. It definitely can be the way to go. I prefer the smaller slower motor and smaller battery approach because I want good handling on a rocky dirt trail. I want the lightest possible dirt bike that still performs ok for my riding style. I'm gonna crash so 25 mph is plenty, and I still have 2500w of power. On my cargo bike, 20 mph and 1000w is plenty.

For sure, if a single motor dissapoints, then two motors is not a bad idea. A bit harder to get the motors matched though, with different wheel sizes. Looking hard for a used 5305, or special ordering a winding in a 5405 makes sense for the next step after a 9c experiment. Tom will try lots of stuff, I bet, and enjoy the journey like I did.
 
dogman said:
I still suck at interpreting the sim, and have years of just trying it and finding out on the road.

Clarify please why you think the 2810 wont cut it on the 10%. Bear in mind you get to pedal.
Of course, a smart man would just take the 5% route with a load. I just want to learn what you just did.
Click to expand...

making a jerry-rigged simulation for a real world 72v 40Amp setup both the 2810 and 2812 will overheat in two minutes flat up a 10% incline with a load of 300kgs (650 lbs). as the incline is near the end of my journey I'm expecting that some heat buildup will already have happened prior to the climb. Pedaling will help to an unknown degree, but as the simulation may not be accurate when jerry-rigged in this manner I think it is too close for comfort and am unwilling to hinge my bets on it a priori.

Hopefully the addition of a cycle analogger with GPS from ebikes.ca will get me some real world numbers to crunch when I put the bike to the test :)

But yea... the power(red) and load(black) curves are awfully close so I expect the motor will be near its stall Torque limit.
Playing with lower inclines and throttle speeds indicates that the 2812 has better efficiency at slow speeds, making it take longer to overheat on hills, and thus be a better climber. The 2810 is much faster on flats, but as I don't need speeds in excess of 25kph that is kinda irrelevant for me :)

Thats as far as my greenhorn understanding will take me, anyhow :)
-Tom
 
Two of them would help a lot here. Half the load each. Or run one through the gears and get motor speed up. This is a steep, slow, heavy load application. Not really what this hubmotor is designed for.

Those overheat times are based on static conditions, no cooling airflow.

You'll need a temperature sensor in addition to a CA to really collect good data.
 
Alan B said:
Two of them would help a lot here... ...You'll need a temperature sensor in addition to a CA to really collect good data.
Click to expand...

Agreed on both points. However I am trying to keep my departures from EU regulations as stealthy as possible, So for me, two motors are right out (A "Speedy" rear hub with regen and a BMC V2T front would be awesome though. One can dream :) ).

As for temperatures, I reckon I will need to figure out which part of the system is the thermal bottleneck, then monitor that (and cool it if I can) and a couple of ambient temps as well. Internal hub temps will be hard to get to, but it should at least be possible to model the heat transfer dynamics, measure the outside temp and then calculate backwards.

I've also been pondering battery options. My needs are 13-15Ah. I can't find any Yttrium chemistry packs that small :( It would be fairly simple to just order 2x36v15Ah from Ping, But I think 20Ah would be wiser to allow for continuous 40Amp discharges. Ordering from emissions-free would also be an option, but ordering 4x39V9.2Ah Packs would be overkill both in terms of C rate and needed range.

Do the BMS systems from either allow for worry-free overnight charging? Id love some kind of auto-shutoff capability for when the pack is full. Maybe the cutoff board from Methods would suffice in that regard?

-Tom
 
I'm not sure there is any such thing as overkill in C rate or Capacity. :D

It is not too difficult to put a temperature sensor inside to get coil temperature. That's the leading heat source.

A quality balancing charger, or a quality bulk charger with some type of secondary shutoff protection like Methods or Goodrum's HVC boards provides pretty good protection. It would be good if a fault didn't endanger the neighborhood fire-wise, maybe a smoke alarm in the space too.
 
The slower wind motors have more torque per amp. That's why they call them torque models. It's maximum torque that is equal between different wind count motors, because the fewer turns can take more current. I call the slow wind motors low power motors because of our realistic voltage limit of 100V, which makes their max power lower than the same motor in a higher Kv wind. The 2810 would need to go to 200V to be the same power as the 2805. That's assuming that quote about half the voltage being correct, which I'm unclear about, since inductance doesn't vary linearly with turn count. If Kv varies linearly with turn count then yes it's valid, though the highest Kv hubmotors are harder on controllers, say 12rpm/v and up.

Another thing I don't like about the slower wind motors is how the power curves and efficiency curves are so separate. The result is that you almost need to ride downhill or with the wind to get to their prime efficiency range, because there's no power there, and at peak power efficiency really sucks. That's due to 2 factors. 1. The motor is being run at way too low a voltage and the motor can't really shine without going to an unrealistically high voltage. 2. Current limiting isn't the dominant factor. Running a motor with heavy current limiting really flattens out the power and efficiency curves that are matched very well, so you have a broad band of both. I've posted those graphs many times.

Regarding the 10% hill, I wouldn't even use my larger hubbies to haul 650lb up a long 10% grade, not running it in-wheel. That's like one of those big 300lb e-scooters tackling a 10% grade with a passenger aboard. With a motor Kv and wheel that probably wants to see a 20-25mph maintained up any long grade, it's just asking for trouble. The long way around at half the grade is the best answer. It might cost you a penny extra in electricity, but on the bright side you get a longer ride, which is always better unless it's raining or you're low on juice.

John
 
You might even look at a wheelbarrow hubmotor. Those I've seen online look like high turn count scooter type hubbies mounted in a small fat wheel. It may need high voltage to get to the speed you want, but if they're 12kg motors then they may actually be up to the task, since slow and heavy loads is what they're designed for.

John
 
Yeah, I can definitely agree, at 72v 40 amps, the slow wind 9c motors are getting flogged. They do ok at 1500w, maybe 2000, above that, they are making heat for sure. Whether it's too hot by the top depends on how much you bog down, below 10 mph, a 2810 will start really making heat. But I think you will make it up the 5% grades. But you'll still have to pedal your guts out for at least that part.
 
> wheelbarrow hubmotor.

Where????

here is my guide for installing a temp sensor in a hub motor
https://sites.google.com/site/shelbyelectro/Mods-and-technical/motors/installing-a-temp-sensor-in-a-hub-motor

I'd say just scrap the 9c idea and go with an x5 or larger hub motor.
 
Barrow motor would reaally be a slow wind I bet. I still think a 9c experiment is worth a try. But not if he can source a slow wind of a heavier motor. I think he can do the 5%, don't forget, it's not a 5% mountain it's just a bump. The motor will be toasty at the top, but I think it can do it with lots of pedaling.

Something through the chain with really low gearing for the eventual ideal solution of course.
 
Looking at the sim, using Toms half and double trick. The 5% grades still look tough. But the sim is for no pedaling, and even if you can't pedal up more speed, you still see dramaticly lower numbers on the CA when helping the motor in a low enough gear. Anyway, without pedaling it shows overheat in 8 min for both 10 and 12 turn windings. Should be able to get up the grade in less than 8 min, and pedaling should extend that time just a bit. Play with it more, and lower the weight to 500 pounds and everything gets much better. overheat in 14 min with 500 pounds.

So just carry a little bit less and you should be ok. If the bike and you is 400 pounds, you have 100 pounds for cargo.

I was simulating with 30v and 70 amps. So similar to 15s lipo and 35 amp controller.
 
Hey cargo_tom -

Do you have any updates? You have my curiousity up - I have both a 2810 and a 2812 motor and I love 'em both. I'm wondering if you've got a build going yet? What you have done so far, all the research and your objectives, has been very similar to the considerations I've had for my builds, even though I'm shooting for long distance, high efficiency.

We're all pullin for ya man!

JKB
 
Howdy :)

No substantial upgrades yet, im afraid, as I have been bogged down with engineering classes lately (introductory classes on mechanics, thermodynamics and circuit theory... useful ebike stuff)

I do have a project timetable and have done most of the parts shopping in the last few days. So things should kick off in 4-5 weeks tops, as I get exams out of the way.

I have ordered a 9c front hub wheel, throttle and CA at Method's estore. I have also ordered a 40A infinion controller, GPS analogger and various odds and ends at ebikes.ca, so I only need to get battery and charger sorted before the tech end of the project can get underway.

I have also managed to borrow a woodshop for a few days come summer, so hopefully I'll get the new marine ply cargo enclosure ready as the parts trickle in.

In other news I've gotten my hands on a sweet dual channel oscilloscope/multimeter complete with logging capabilities and data analysis software. So I will endeavour to really get under the hood of this thing and figure out exactly how everything works.

Stay tuned
-Tom
 
Cargo_Tom said:
Howdy :)

No substantial upgrades yet, im afraid, as I have been bogged down with engineering classes lately (introductory classes on mechanics, thermodynamics and circuit theory... useful ebike stuff)

I do have a project timetable and have done most of the parts shopping in the last few days. So things should kick off in 4-5 weeks tops, as I get exams out of the way.

I have ordered a 9c front hub wheel, throttle and CA at Method's estore. I have also ordered a 40A infinion controller, GPS analogger and various odds and ends at ebikes.ca, so I only need to get battery and charger sorted before the tech end of the project can get underway.

I have also managed to borrow a woodshop for a few days come summer, so hopefully I'll get the new marine ply cargo enclosure ready as the parts trickle in.

In other news I've gotten my hands on a sweet dual channel oscilloscope/multimeter complete with logging capabilities and data analysis software. So I will endeavour to really get under the hood of this thing and figure out exactly how everything works.

Stay tuned
-Tom
Click to expand...

Damned thermo. :) :) :)

The scope - is it one of the little dual-channel LCD solid-state scopes? Been thinking bout one of those myself.
 
Cargo_Tom said:
jkbrigman said:
The scope - is it one of the little dual-channel LCD solid-state scopes? Been thinking bout one of those myself.
Click to expand...

nope. 'bit bigger than that ;) a Fluke 123 scopemeter with most of the fixings. I need a current clamp and a thermistor module and I'll have full use of the thing.

-Tom
Click to expand...

Holy crap. That's the mac daddy. I worship in that Fluke's general direction....
 
BTW, you did not add a 10% grade in your ebikes.ca simulator test according to the first image.Thinking about it a bit more, i realize that it might take much greater power.

Only the 'noname5004' ( cromotor hubzilla ) could even start to get you up that grade, even then in a 20" wheel, it will only get 70% efficiency according to the calculator, 3300w constant just to do 25km/h.

On a flat, you only need 540w and will hit 31km/h.

If you ever see the scenario mentioned above ( 661lbs up a hill in a 20" wheel ), you're gonna need more than a hub motor. Start thinking about chain driven motorcycle motors O_O
 
Agreed :) 10% incline will be too much. I have the option of a detour with half that gradient. I am not adverse to pedalling to help things along. Also bear in mind that the hill I need to climb is only 300 feet high, so the climb wont take all that long to do. The rest of the commute is pancake-flat.

I am constantly dreaming about more projects to take on, and the Cro-motor is on my long-term shopping list.

In other news Im about ready to pull the trigger on two 36V 20Ah packs from Ping with upgraded chargers.
I'd prefer large capacity single prismatic cells to charge on a per-cell basis and avoid bms altogether (I like the simplicity of having fewer, larger cells), but Ping's reputation seems solid enough.

I will be looking into hydrogen fuel cells for my long term power solution anyway (fuel cells are on my curriculum next year, and will help provide insight and practical experience-), and when that time comes the 36v packs can find use in low-power projects for my offspring

-Tom
 
Oh, then that changes things.
Don't expect your pedal power to do much though, the electric motor will be doing the overwhelming majority of the work even on half the grade.

I looked up hydrogen stuff at one point. Awful power density, high price, and unless you are generating the hydrogen yourself, you are likely getting it from refined natural gas, ultimately costing you as much as gas :|.

If you find any different, you let me know. Until then i laugh at $800 fuel cells that can only output a few hundred watts constantly :roll:
 
The technology focus at my department is HTPEM fuel cells with methanol used as a hydrogen carrier. Serenergy is a spin-off from my department and has already found use in transportation applications.

The advantage for me will be access to inside knowledge and hopefully the know-how needed to go the diy route, bypassing the expense of an emerging market.

In any case it is several years down the road. I fully intend to get my moneys worth out of lithium first :)
-Tom

Edit May 21: Just ordered 2x 36v 20Ah packs from Pingbattery.com. Will need to look into buying the diodes for hooking them up in series.

Edit May 31: Got the 9c wheel, controller, CA, analogger ebrake lever and throttle today.
2" thick moped tire is going on tomorrow.
 
You must log in or register to reply here.

Similar threads

E
Kona UTE Cargo Ebike
Replies
13
Views
849
dogman dan
dogman dan
Peterfr12
Dual suspensions Recumbent with steel frame. New build and futur updates
Replies
5
Views
975
wturber
wturber
O
Strongest/best rear wheel for heavy cargo e-bike!
Replies
19
Views
5,599
Chalo
Chalo
panurge
Luna Cargo Frame with TDCM igh
Replies
18
Views
9,701
wyvernwaddell
wyvernwaddell
Back
Top