Motor comparison spreadsheet

crossbreak said:
Lebowski said:
Battery current is not equal to phase current at full rpm...
At no load and 100% duty, input should be equal to output? where is the rest? what is the part that's lost?
Its not lost.
The system is still working as a buck converter. Except in the case when some field weakening is used. Then its working as a boost converter.

But non the less its always changing the DC amp Volts around the something different in AC amps and Volts. Its never the same on both sides.
 
At 100% duty cycle, a buck converter does not lower the voltage any more. Input and output are directly connected as the MosFET is on 100% of the time. In a 3 phase converter, the current is distributed to 3 phases. as peak voltage=Battery voltage, battery current must be the same as RMS phase current. So nothing gets lost.
 
crossbreak said:
At 100% duty cycle, a buck converter does not lower the voltage any more. Input and output are directly connected as the MosFET is on 100% of the time. In a 3 phase converter, the current is distributed to 3 phases. as peak voltage=Battery voltage, battery current must be the same as RMS phase current. So nothing gets lost.

Zombiess measured that they never actually get to 100% duty and Lebowski said that they're not equal, so I'll stick with them.
 
crossbreak said:
if there is no controller that does so, then lets build one. Why waste this potential?
Its not potential at all.

The sine wave from the motor does not match the Square waveform you would get from full block communication from the battery.
You have to remember at full rpm the controller will pwm voltage from the battery that's higher then the back EMF at that very spot on the sine wave then the back emf at peak of the sine wave will climb above the voltage from the battery. Having the Controller hold the IGBT or Mosfets on for the whole length of the sine wave is stupid.

Also remember its also a combination of what's stored in the Caps in the controller it self.
 
what i see on the pic i just made with my DSO is not 100% duty at the rise and fall edges of the trapeze. But at least it looks like in the middle :?

the controller does 20 khz PWM (i think) it is a x8m06c from BMS battery. So I sould be able to see about 10 PWM cycles on the "tip" of the trapeze at 5µs capture.
 

Attachments

  • Bafang_BPM_11T_48.8V.png
    Bafang_BPM_11T_48.8V.png
    3.9 KB · Views: 4,781
  • Bafang_BPM_11T_48.8V_5us.png
    Bafang_BPM_11T_48.8V_5us.png
    2.6 KB · Views: 4,780
Did some more measurements with the BPM 11T and x8m06c controller

peak to peak phase voltage is a little less then 100V, i read from the DSO pics (1:10 probe)

RMS phase to phase voltage: 36.9 V
RMS phase current at 1 phase: 1.699A
apparent power: 62.7 VA

Battery voltage: 48.8 V
battery current: 1.083 A
effective power: 52.85W

phase to ground AC: 22.08 V
phase to ground DC: 24.01V

frequency between phases: 218 Hz measured by UT139C DMM

looks like it is not doing PWM at Full Throttle but it's not putting out an RMS phase voltage that is equal to battery voltage, too. This is my bad, sure it cannot output RMS phase voltage close to battery voltage, it can just output peak phase voltage that is close to battery voltage. The difference should be √2 for a sine wave. The trapezoid wave comes close to this figure.
In addition to this, RMS phase current will not be close to battery current at full throttle as the motor itself has inductance and thus a reactive current is flowing that adds up geometrical to real phase current
 
Looks like a waste of energy to me.

Also watch out the little DSO scopes are just toys for kids they don't have the resolution to catch the small PWM events and other things we usually look for with a scope.
I keep mine in my back pack for emergency ICE motorcycle diagnosis. And I use my 100Mhz scope for the real work... (which I'm also looking to upgrade accepting donations :) )
 
at no load conditions there is no energy wasted. FOC controllers seem to use the same amount of energy at this condition. Energy savings seem to rise with current for trapezoid vs FOC

Arlo1 said:
Also watch out the little DSO scopes are just toys for kids they don't have the resolution to catch the small PWM events and other things we usually look for with a scope
sure screen and vertical resolution is low, but you can clearly see PWM event if zooming in

next queston is: Is resistive power calculated right in the spreadsheet? As phase current seems to be at lest √2 larger than assumed, resistive loss should be at least 2 times larger (as reactive current additionally adds up geometrically). Something tells me that it cannot be 2 times larger as there are 3 phases and not just 2. But we measure resistance only across two phases (so called terminal resistance)
 
John in CR said:
Zombiess measured that they never actually get to 100% duty
This was just a test for full throttle at lower then max rpm.
I also did this test and found random 400+ uS on pulses with full throttle at lower then max rpm, which explain some of my failures trying to run low inductance motors.
 
full throttle at lower then max rpm
i dont get it? doesn't full throttle account for max rpm for a given battery voltage? (without using flux weakening).

Maybe some controller do 100% duty at max rpm and some dont? i remember the Kelly KBS-X had a shape much closer to a sine wave with almost no 100% duty period at the tip. I took that measurement with a better osci back then
 
crossbreak said:
full throttle at lower then max rpm
i dont get it? doesn't full throttle account for max rpm for a given battery voltage? (without using flux weakening).

Maybe some controller do 100% duty at max rpm and some dont? i remember the Kelly KBS-X had a shape much closer to a sine wave with almost no 100% duty period at the tip. I took that measurement with a better osci back then

What I am pointing out is when the motor has a load on it and is accelerating from 0-max rpm its not yet at max rpm its at something LOWER THEN MAX RPM we were concerned the PWM was going to full block communication (NO PWM JUST ON OR OFF) but Zombies and I tested this at LESS THEN MAX RPM WITH FULL THROTTLE.

We were not looking for max rpm unloaded we did not care about that.

Yes some controllers are different for sure.
MOST SINE if not all sine controllers will not go to full block communication ever. You can not get a sine wave if you do.
 
thx, got it. that wasnt no load. Cannot do any loaded testing though. Sadly i have got no dyno... would be nice to make some loaded tests and measure torque

did not find the old shot from the kelly measurement, i'll repeat that when i find the time. ATM i'm reading a bit about reactive current in BLDC machines... may have some influence, need to dig out some formula to predict it's influence on rpm and resistive loss. I think for motors that have a lot of inductance like the MAC Hub motor, it may have noticeable influence

MOST SINE if not all sine controllers will not go to full block communication ever. You can not get a sine wave if you do.
sure, that is also a reason why you get a different max rpm figure at the same battery voltage. And thus you calc a different (higher) Kv. A 13% higher Kv leads to a 28% lower Km² and thus a 28% higher resistive loss

At the end of the day, this would not matter if we all would use the same equip to measure motor constants...which isn't the case unfortunately
 
Some measurements of phase to phase resistance on a couple of Astroflight 3210 motors that I have to hand.
Measured at 20 deg. C.

3210 12t

A-B: 0.1648 ohms
B-C: 0.1651 ohms
A-C: 0.1705 ohms

[specification: 0.190 ohms]

3210 7t

A-B: 0.0625 ohms
B-C: 0.0607 ohms
A-C: 0.0607 ohms

[specification: 0.060 ohms]

3210 6t

A-B: 0.0440 ohms
B-C: 0.0423 ohms
A-C: 0.0417 ohms

[specification: 0.045 ohms]
 
crossbreak said:
Now that you have your meter. What does it show you if it connects to one of shunts I sent you?
0.005014 ohms.

It's not quite the same as the one you have. I got this one: http://www.ebay.co.uk/itm/White-OLED-Four-wire-Digital-Micro-ohm-Meter-Milliohm-Micro-Ohm-Resistance-Meter-/281907885385?hash=item41a304f549:g:SxIAAOSwYIhWlb1j
 
Here's the complete measured data for the Astro 3210 12t [manufacturers data in brackets].

Controller: Castle Creations Phoenix HV85

Rm: 0.167 ohms [0.19 ohms]
Kv: 113.8 rpm/V [113 rpm/V]

N.L. amps @ 2930 rpm: 0.43A (battery side) 0.33A RMS (phase)
N.L. amps @ 5458 rpm: 0.50A (battery side) 0.39A RMS (phase)
[0.5A at 7500 rpm??]

Weight 0.957 kg [0.964 kg]
 
Astro 3210 7t

Controller: Castle Creations Phoenix HV85

Rm: 0.061 ohms [0.06 ohms]
Kv: 206.5 rpm/V [194 rpm/V]

N.L. amps @ 5334 rpm: 0.85A (battery) 0.67A RMS (phase)
N.L. amps @ 9894 rpm: 1.12A (battery) 0.89A RMS (phase)
[0.9A @ 7500 rpm?]

Weight: 1.022 kg [0.964 kg]
 
Astro 3210 6t

Controller: Castle Creations Phoenix HV85

Rm: 0.043 ohms [0.045 ohms]
Kv: 234.9 rpm/V [225 rpm/V]

N.L. amps @ 6067 rpm: 0.85A (battery) 0.77A RMS (phase)
N.L. amps @ 11227 rpm: 1.14A (battery) 0.93A RMS (phase)
[N.L. amps @ 7500 rpm: 1.0A]

Weight: 1.028 kg [0.964 kg]
 
MXUS 28mm DD Hub, 12X5T Winding, V2 Stator

Phase Resistance: 0.122 Ohms
No Load Current: 1.64 Amps, at 50V
No Load Power: 82 Watts, at 50V
Bare Motor Weight: 6.45Kg
RPM Per Volt: 596 RPM/50V = 11.92 Kv
 
What would be ideal high toruq, highpower under 12lbs motor to 16" dirt bike? I am looking for Revolt 120 Pro but is there any better with more torque? :)
 
https://endless-sphere.com/forums/viewtopic.php?p=1177299#p1166771
VasiliSk said:
Got this data from RV160 short
[..]
80mOhm
[..]
33.36kV (RPM/V)
[..]
this gives me 39.6Nm on shaft of motor with 180 phase amps

Losses_zps6evqqecl.png~original


so this motors Km² is only ~1 :( weaker than the RV120 that gets 1.6 if we can trust the specs from Revolt
 
Just an idea here but since we've now added the ability to input a custom motor on the ebike simulator page ( see https://endless-sphere.com/forums/viewtopic.php?p=1324922#p1324922 ), you could in principle add a column to this spreadsheet that uses the string and math functions to produce a direct URL for viewing this as a custom motor in the online simulator. The only thing missing in your columns right now is the one for the inductance, but everything else is there.

For instance, if we look at the Joby JM1S motor, we have

11 pole pairs
94 rpm/V
.015 ohms
0.107 Nm hyst trq
0.00007 Nm/rpm eddy trq, which is .0000073 Nm/Rad/sec

If I assume that the winding inductance is .05 mH, then this can be expressed in the custom URL string as motor=cust_94_0.015_0.05_11_0.107_0.000007_0
The order is just KV_Ohms_mH_PolePairs_Hyst_Eddie_TransmissionLoss Append it to the URL like this:
http://www.ebikes.ca/tools/simulator.html?motor=cust_94_0.015_0.05_11_0.107_0.000007_0

Of course the default shows it directly coupled to a 26" wheel diameter, while a motor like this would be geared down, so it might make sense to have the string default an initial effective wheel size based on the KV. Here's a more realistic set of parameters for this motor on an ebike:
http://www.ebikes.ca/tools/simulator.html?bopen=false&motor=cust_94_0.015_0.05_11_0.107_0.000007_0&wheel=2i&batt=B5216_GA&cont=C35
 
Back
Top