Cromotor Dyno, 13 HP

[zombiess]

I had a customer send me a dyno graph of his bicycle and I thought I would share it here. Something about an ebike on a motorcycle dyno makes me smile

Setup was 125V pack, 175A phase current and 175A battery limit with an EB324 controller built by me (one of two EB324's that I was able to get working). Sadly the controller killed a MOSFET on his 3rd pull.

Here is a video he made of the cycle analyst screen during a run
[youtube]zPX8oQU5Ou8[/youtube]

Here is the graph

 

[Lebowski]

Man that's some serious power way more than my little 8hp 110cc. You're easily talking 150cc equivalent

 

[John in CR]

Nice to see someone push one at higher voltage than Methods did. Did he tell mention peak battery amps and minimum voltage to give everyone an idea about the input side? I'm all too accustomed to that kind of sag with 30s, so I can't wait for the stiffness of my new 65/130C pack.

Zombiess, I'm a bit perplexed about his controllers settings. No motor I've tuned has ever sounded right with battery and phase limits set the same. They either get noisy on startup or don't give full speed and sound starved at the top. Maybe proper tuning would help him get some more out of Cromotor. Of course doing like Doc Bass does and hitting the dyno in the winter in a sub-zero shop certainly helps him make Xlytes look better...got any customers in southern Australia where it's winter now?

I sure am glad the 24fetters I got from you aren't EB324's. I've run a pair in daily use for a year and a half without a hitch, each at 123A/185A, and the steep climb to the wind turbines is far more torturous than a few dyno pulls. They do run pretty warm though. With my heavier moto wheel to help keep the front down, I've been considering bumping the combined battery side up to 270A to offset the slightly larger tire I put on. Do you think it's already too close to the actual limits to risk going to 135/200A and 31s, or are the high voltage 24's I bought from you sufficiently better than the EB324's? Remember I do have the one spare just in case. If one blows, I'd just pull over, cut the phase wires on the dead one (no connectors for me) and hobble home on one, that is if you can call 14kw input "hobbling". 8)

 

[zombiess]

Customer made a mistake on the shunt setting and accidentally set the battery current too high. When this is done on EB324 controller the phase amp limit becomes a hard limit for the battery side. The battery limit is always <= phase current limit on that controller type. EB224 controllers multiply both the phase and battery current limits by the shunt multiplier. I've written some threads on this previously if you want to dig them up.

I would not increase the settings from where you are at John. I've been greedy with power many times and pushed things to the point they break. Not as bad as LiveForPhysics does though. I usually sneak up on the failure point unless I'm in a mood.

The reason you should not increase the settings is you run the controllers at a higher duty cycle than people with lighter less powerful bikes do. Your hill climbs are not an easy task for controllers. Your highway runs aren't exactly easy on controllers either. If you were only doing short bursts you could probably get away with upping the current. Having the controllers get warm uses up most of the headroom the MOSFETs have. Just remember that if your controller is hot, the MOSFET junction temperature is about 1.7 hotter during operation. This is why I replace the unknown thermal insulators with quality ones when I built the controllers. A key to making them live under conditions such as yours is thermal management.

Your method of keeping the phase amps / battery amps ratio low works really well to cram more power into a motor. At 185A your MOSFETs have to handle 46A each which is a lot for near continuous use. For comparison a Sevcon Gen 4 size 6 is rated around 25A per MOSFET for continuous use. If you were going to do continuous use such as in a generator, you would probably need to back off your settings a bit, but it's hard to say without real world testing and logging.

The below data is part of what was sent to me in a spreadsheet. I can calculate the controller efficiency from this.

I have the data for this pull that was sent to me.
RPM
130.515
186.45
248.6
310.75
372.9
435.05
497.2
559.35
590.425
621.5
683.65
745.8
795.52

Motor Efficiency (calculated 3 phase RMS input to the motor). A little disclaimer here is that this efficiency data is from much higher input than is used to create charts by the manufacturers. It looks like the motor still has more power to offer since the efficiency > 50% for most of the pull and maximum power transfer occurs at 50%. I wonder what it would feel like to ride a controller that maintained a 50% efficiency through the entire RPM range At minimum it should be a tuned curve to never drop below 50% efficiency as the rest is just wasted at heat and raises the winding temps thus raising the resistance thus lowering overall output capability in the higher RPM ranges... after typing that out I think I better understand the relationship between phase amps and battery amps and I now have some new ideas

36.71093043
42.69263724
45.322109
56.29738681
61.5114747
62.87840499
68.31178102
68.02354687
65.95241261
73.70239128
73.14275958
80.47586666
85.0250371

Output Watts
3412.541633
4794.611627
6185.796073
7267.930946
8321.185512
9492.260143
9960.963147
10545.80818
11256.88604
10653.95507
9787.323875
8033.73503
6878.765873

 

[Doctorbass]

John

We have 4 seasons here and actually it's 25 degree C outdoor now.... lol.. all dyno test was done in a shop at normal 20 degre C ambiant temp... no winter lol...

and yes a simple 5302 with 32mm stator did a real 10,5hp output with 10.5kW inpyt making efficiency to 75%...

That was with not tuned Adaptto Max E and 22s Zero MNC.

[youtube]4QVazIxtNwc[/youtube]

Now the bike have the 5403 on 24" and it pull way harder!!

I would say easy 15hp with the 16kW i measured on the adaptto with the traction control tric in teh setting.

Next upgrade: Dual adaptto Max-E and Dual MXUS 3000. The goal is to have the same or better power to weight ratio than the TELSA S. This should give about 30hp total between 90-110km/h The traction control feature of the Adaptto will automaticly reduce the torque on the front Wheel if necessary, to keep all traction.

Doc

 

[Metallover]

I put my Cromotor/Max-E setup on a Dyno last fall. It made about 9.5HP.

https://youtu.be/L7MFAfy9C9I
[youtube]L7MFAfy9C9I[/youtube]

 

[Doctorbass]

Cromotor are capable of more than 10 or 15hp i believe.

all you need is solid battery amp and high voltage or flux weakening. In teh case of the Adaptto even with 18s you can get more than 10hp at the wheel.

if you follow the linear ratio ( that is not really linear in fact lol) of 10.5hp with 32mm stator and lower copper fill of the X5 and the cromotor ( 50mm stator and more copper fill) i would say easy 15-18hp with the proper power, torque arm 8) and driving skill

Doc

 

[zombiess]

Doc,

If it was easy, it would already be done

I really like seeing all these ebikes on dynos. Nyx frame is looking good. I saw one in person when I visited LiveForPhysics last month.

 

[John in CR]

Zombiess,

11,258W output is 15.1 peak horsepower by my calculations.

On the subject of heat, you're making the all too common mistake of so directly linking it to efficiency. The motor makes the same amount of heat in the copper with the same phase current whether the motor is operating at high efficiency or low efficiency. Efficiency is a result, not a cause. Low efficiency doesn't cause heat, current does. There's no getting around low efficiency in the early stages of acceleration, since from a stop it always starts at 0%, and efficiency is low because rpm, and therefore power, is low. Forcing (actually trying to force) a 50% minimum efficiency would make for a horrible ride.

Regarding my controller settings, the current settings were almost the first ones I tried. FWIW, I probably run at a lower duty cycle that almost anyone. I rarely exceed half speed, but when I do get on the highway is when the controllers are the under the least stress, since the motor limits the current well below the programmed limits. Up thru mid 2010 I blew a tremendous number of controllers, but not a single one at high speed. That's where our systems are the happiest, and that's before considering the greater airflow to cool them. Exactly how much my current levels decrease at highway speeds compared to during the few seconds to get there is the kind of data I look forward to seeing with my Grin Tech Analogger plugged in. At this point I have only educated guesses of how much power I use to go 60 or 80 or 100mph.

You're right about the crazy climb to the top of the mountain, because I do need to be careful with the controllers. The one time I did it though I really attacked it, but slowed extremely for the curves, because I was totally unfamiliar with the road. I'd probably lower the settings next time, but even with higher settings I think I could make it easier on the controller than before simply by maintaining a steadier speed.

I've lost 30lbs, which slightly more than offsets the bigger load the motor sees due to the increase in tire size, so as I lose more I should be able to nudge the controllers up a bit. That's because I will ask for the higher current for shorter durations and/or BEMF will climb faster lowering current more quickly to better protect the controllers.


Doc,
Your analysis leaves out the impact of the Kv. We're limited by voltage, so an otherwise identical motor with a higher Kv will always be capable of higher power. That's because different windings have identical torque limits (assuming equal copper fill of course), but the higher Kv motor can do it at proportionately higher rpm and power (assuming it hasn't reached it's rpm limit of course), since power = torque X rpm. A MXUS is less motor than a Cromotor, so unless you got the 3 turn version or you exceed 100V fresh off the charger, each one can't break 10hp. You're a light guy, so it should still be awesome power for you.

 

[Doctorbass]

Doc,
Your analysis leaves out the impact of the Kv. We're limited by voltage, so an otherwise identical motor with a higher Kv will always be capable of higher power. That's because different windings have identical torque limits (assuming equal copper fill of course), but the higher Kv motor can do it at proportionately higher rpm and power (assuming it hasn't reached it's rpm limit of course), since power = torque X rpm. A MXUS is less motor than a Cromotor, so unless you got the 3 turn version or you exceed 100V fresh off the charger, each one can't break 10hp. You're a light guy, so it should still be awesome power for you.

Yes i have the 3T version(Motor is 21 x 3T.) on 25.5" wheel and 22s and flux weakening on the Adaptto that add about 20 30% more speed.
Actually with a 4T equivalent(my 5403) without ovs the torque fall at about 70km/h (normal mode)but with OVS( in BOOST mode) , it feel it never decrease ! in fact i get 116kmh max from now.

Doc

 

[Doctorbass]

Doc,

If it was easy, it would already be done

I really like seeing all these ebikes on dynos. Nyx frame is looking good. I saw one in person when I visited LiveForPhysics last month.

Glad you say one of these frame Zombie! :wink:

Soon i'm sure we will see one with a cromotor!

and on my side... with two MXUS 3000

 

[John in CR]

Doc,
Your analysis leaves out the impact of the Kv. We're limited by voltage, so an otherwise identical motor with a higher Kv will always be capable of higher power. That's because different windings have identical torque limits (assuming equal copper fill of course), but the higher Kv motor can do it at proportionately higher rpm and power (assuming it hasn't reached it's rpm limit of course), since power = torque X rpm. A MXUS is less motor than a Cromotor, so unless you got the 3 turn version or you exceed 100V fresh off the charger, each one can't break 10hp. You're a light guy, so it should still be awesome power for you.

Yes i have the 3T version(Motor is 21 x 3T.) on 25.5" wheel and 22s and flux weakening on the Adaptto that add about 20 30% more speed.
Actually with a 4T equivalent(my 5403) without ovs the torque fall at about 70km/h (normal mode)but with OVS( in BOOST mode) , it feel it never decrease ! in fact i get 116kmh max from now.

Doc

Sounds great. I'll be especially interested in what the traction control allows you to push through the front motor, because I have a couple of 2WD's in the project queue.

 

[zombiess]

Sounds great. I'll be especially interested in what the traction control allows you to push through the front motor, because I have a couple of 2WD's in the project queue.

By this time next year I hope to have a controller that will be able to overheat any ebike/scooter motor in one pass. Something that makes 2WD utterly useless for performance, much like fast AWD cars that convert to RWD only for increased performance.

My dream of manufacturing controllers is getting much closer I'll be dyno testing my 18 MOSFET prototype for current handling in a few weeks.

 

[John in CR]

Sounds great. I'll be especially interested in what the traction control allows you to push through the front motor, because I have a couple of 2WD's in the project queue.

By this time next year I hope to have a controller that will be able to overheat any ebike/scooter motor in one pass. Something that makes 2WD utterly useless for performance, much like fast AWD cars that convert to RWD only for increased performance.

My dream of manufacturing controllers is getting much closer I'll be dyno testing my 18 MOSFET prototype for current handling in a few weeks.

Sounds great if 2 controllers can be run from one throttle. You did leave out one very critical word, "economical".

There are real advantages to 2wd though. It doubles heat dissipation, and with traction control there's potential for improved handling in the same way that a front motor is different.

 

[Doctorbass]

Sounds great. I'll be especially interested in what the traction control allows you to push through the front motor, because I have a couple of 2WD's in the project queue.

By this time next year I hope to have a controller that will be able to overheat any ebike/scooter motor in one pass. Something that makes 2WD utterly useless for performance, much like fast AWD cars that convert to RWD only for increased performance.

My dream of manufacturing controllers is getting much closer I'll be dyno testing my 18 MOSFET prototype for current handling in a few weeks.

Sounds great if 2 controllers can be run from one throttle. You did leave out one very critical word, "economical".

There are real advantages to 2wd though. It doubles heat dissipation, and with traction control there's potential for improved handling in the same way that a front motor is different.

John i guess you could be interested to read about the Adaptto Max E and mini E. They have traction control and also 2WD function that allow you to operate two controllers with the same throttle and individual settings for each motors.
The idea of double heat dissipation of both motor and controler and also maybe minimize saturation of the stator are one of tbe adva tage i want to test. One more controller is 1kg and one motor is 10kg so adding 11kg out of 135kg total but having twice the power dissipation will certainly help.

I know what i get with only a 40mm stator and 16kw and this is ery impressive. I am very impatient to see how 90mm sator total and 32kw will do!

According to Luke the 25Ah Farasis cells i use can do 400A burst from 90 to 100%soc. That future setup will test that for sure!

Doc

 

[Scott]

Thanks for posting this Zombiess! I totally forgot and never got around to it.

I had always wanted to get my bike on a dyno but I hadn't since it's so expensive. These tests were for a Presentation in physics 3 in college. The professor loved it and I ended up getting a 4.0 with honors credit in that class!