Performance of Unite my1020 1000W motor

40 replies [Last post]

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
safe
safe's picture
Offline
Joined: 08/03/2010
Points: 583
Re: Performance of Unite my1020 1000W motor

mc-taichi wrote:

Inductance is like a damper(shock absorber) rather than spring.

Yes, a better description.

mc-taichi wrote:

Is it for peak current not average?

The formula is for full throttle, controller adjusted, voltage for a specific rpm. It's designed to make creating a spreadsheet easier.

To really properly represent the full extents of a motors behavior you would need a three dimensional chart where partial throttle situations were considered.

The controller behavior goes like this:

1) If the throttle directed "duty cycle" exceeds the naturally occuring motor demands for voltage at a specific rpm, then the controller will present a 100% "duty cycle" to the motor and the motor itself will be the limiting factor on current. The range I'm talking about is the upper 20% where the BackEMF is the dominant influence.

2) If the throttle directed "duty cycle" exceeds the naturally occuring motor demands for voltage at a specific rpm and the result is a current flow that is above the current limit, then the controller will "pulldown" the "duty cycle" just as much as is necessary to attain the current limit. The hand on the throttle might register full throttle (100% "duty cycle") but the controller might only be "permitting" 75% "duty cycle".

3) If the throttle directed "duty cycle" is LESS than the naturally occuring motor demands for voltage at a specific rpm then the voltage will be reduced to the motor to a level that matches the reduced throttle setting. Current limiting is in effect "disabled" and control reverts to a linear relationship.

---------------------------------

Load is another confusing factor, but in most cases a higher load applies to option 2) above. When a motor is heavily loaded the controller does a "pulldown" of the "duty cycle" in order to limit heat.

The controller is there to "prevent" you from doing things that are bad for the motor... (it's like a parent telling you to be good)

One of the favorite tricks of the "juvenile mentality" is to alter the controller so that it has NO current limit. This allows nearly unlimited current through the motor and that means excessive heat. You can easily fry the controller as well as burn up the motor this way. The "Rated Load" defines a "Rated Heat" limit... but that's over a long period of time. For short durations you can exceed the "Rated Heat" and get away with it. Some controllers have a setting for this short term spike in current, but they then have a timer that brings the limit back down after a few seconds.

When you take away the controller current limit it can produce huge power... but it can also destroy the motor in a hurry.

safe
safe's picture
Offline
Joined: 08/03/2010
Points: 583
Re: Performance of Unite my1020 1000W motor

Motor Rewinding

For those hardcore and extreme enough to want to actually take one of the 1.4hp Currie motors apart and replace and rewind the copper windings this posting attempts to give some guide about how it will alter behavior.

From the best I can tell the stock motor has:

Kv = 121 rpm / volt
Turns = 14
Gauge = 18 AWG
Resistance = 190 mohm

...so (without all the fancy math that I don't want to go into right now) we get:

Kv = 113 rpm / volt
Turns = 15
Gauge = 18 AWG
Resistance = 204 mohm

Kv = 106 rpm / volt
Turns = 16
Gauge = 18 AWG
Resistance = 217 mohm

Kv = 100 rpm / volt
Turns = 17
Gauge = 18 AWG
Resistance = 231 mohm

Kv = 94 rpm / volt
Turns = 18
Gauge = 18 AWG
Resistance = 244 mohm

...since I have not actually dismantled my own motor I'm not certain that the turn count is 14 turns, so if you do want to try this you might need to correct for that.

Use the spreadsheet with these numbers to see if you get something that you like. Don't forget that as you increase the number of turns you increase the resistance and that means more heat.

Another approach would be to increase the gauge while keeping the same number of turns. Also you can use multiple parallel winds to achieve lower resistance. It's a complete world of relationships where more than one configuration can achieve the same outcome. (sort of like different gear combinations on the front and rear derailler on a bike)

Using the spreadsheet this is how I'd expect the 18 turn rewind to perform:

...overall this would be a nice configuration. In fact, if I were to rewind my own motor (which I have considered doing) this is the setup I'd use.

Note: I HAVE done a rewind with the same motor slot size and was successful in getting it to fit with 18 turns, but it was EXTREMELY tight. First time motor rewinders might not get it to work for them. This is very hard to do.

--------------------------

Also, before people get too far into the idea of rewinding these motors I need to point out that the far better solution is using RC brushless motors. The reason I still have this motor running is that the bike was built back in 2006 and I just kept toying with it. For a new build you ought to consider RC brushless motors.

http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=14427

...this is better performance than the Currie motor at higher power levels. Controllers are more complex for brushless and you often have to modify these RC motors with additonal hall effect sensors to make them work acceptably, but there are people using them and getting big power.

A legal ebike in the US is roughly 750 watts output, so be clear that this is well beyond the law. There are smaller RC motors that would make more sense if you wanted to be more legally compliant. This motor would not work well with a 1000 watt input limitation. (there are better choices)

In just about every way this motor is superior at producing higher power levels than the Currie. Heat is low. Efficiency is very high and "wide". This would produce a wide and powerful powerband ebike.

mc-taichi
mc-taichi's picture
Offline
Joined: 04/14/2010
Points: 18
Re: Performance of Unite my1020 1000W motor

Thank you safe. "controller behavior" is what I would like to know.

So,in case of full throttle, the condition must be always case 2) until rpm goes up to the peak power.

the controller will "pulldown" the "duty cycle" just as much as is necessary to attain the current limit

means the current is stable at the limited(max) level, isn't it?

Now, let me back to my question, why the current curve is like that(down slope)?

By the way, following is my wiring in which the current mater is installed as shown in red.

controller_wiring2.gif

And, I set 80%(means 40A) for "Max Armature Current[3]" on "step1" shown in Configuration Program for my Kelly controller.

The result of test ride is that the current is 30-35A at low speed then goes up to 45A or more at 35km/h or higher.
I have no idea what is going on.

safe
safe's picture
Offline
Joined: 08/03/2010
Points: 583
Re: Performance of Unite my1020 1000W motor

mc-taichi wrote:

And, I set 80%(means 40A) for "Max Armature Current[3]"

There are two types of current limiting. Armature current limiting means it measures how much current goes through the motor. It should produce a fairly flat result at most rpms, except at the really high rpms where the amps will taper off.

mc-taichi wrote:

The result of test ride is that the current is 30-35A at low speed then goes up to 45A or more at 35km/h or higher.

I'm assuming full throttle all the time...

At low rpm the current is a little low and at higher rpm it's a little high. This might just be some internal behavior specific to the controller. In theory it should be 40A all the time as that's how you set it.

If you change the wiring so that the current is "sensed" on the battery side then you will get "Current Multiplication". When you "sense" the motor (Armature) current you should get no "Current Multiplication".

I think you have Armature current limiting but there is something else going on that's throwing the results off.

Are you really holding the throttle wide open at low rpms or just part way?

e-doggies
e-doggies's picture
Offline
Joined: 12/30/2007
Points: 280
Re: Performance of Unite my1020 1000W motor

One of the posts above shows the top left corner of the worksheet. It shows Current of 40 in cell B5, and Current of 75 in cell D8. The current in column D goes down as the rpms increase. If this motor was controlled by a controller that could limit armature current, would the numbers still look like this? In other words, at 100 RPM, would it show 40 in D8?

I've been trying different volt/amp combinations for the 1.4HP Currie, based on the numbers you have. At 36V, it seems pretty easy to get 1000W output. 48V @ 25A, and 60V @ 20A, both result in 1000W output, but the top RPM's are a little on the fast side.

You run yours at 48V @ 40A. How well do you think this motor would hold-up at 60 Volts and 20 Amps?

The majority of my riding is done at part-throttle (less than 50% throttle). Is there a simple and quick way to get approximate motor data at partial throttle?

mc-taichi
mc-taichi's picture
Offline
Joined: 04/14/2010
Points: 18
Re: Performance of Unite my1020 1000W motor

YES, I am really assuming full throttle all the time including at low speed of course.

At any voltage which varies according to electricity remains in the battery, rpm at low speed is 10-15% lower than rpm at high speed. But it is not my point in this time, it may be an error or character of my controller. My point is current must be constant ideally because it is limited.

So, my question is why the current should be higher at low rpm then goes down as rpm goes up like your chart.

As for the position of current meter, no matter how my controller works, the currents should be same in any position in the circuit.

controller_wiring2_0.gif

I think i1=i2=i3 in above figure because of single way of wiring at this part. Are there any misunderstanding?

safe
safe's picture
Offline
Joined: 08/03/2010
Points: 583
Re: Performance of Unite my1020 1000W motor

e-doggies wrote:

You run yours at 48V @ 40A. How well do you think this motor would hold-up at 60 Volts and 20 Amps?

Actually I run mine as 48V @ 20A which translates to roughly (48*20=960) 1000W input.

60 volts @ 121 rpm / volt would produce a 7260 maximum rpm which is impossible with a brushed motor. At about 5000 rpms the brushes start to lose effectiveness and "float" just like valves float in a car that is running too fast.

Battery Current Limiting should produce this formula for Column B in the spreadsheet:

=MIN(((C8)+(SQRT(C8^2+4*($Voltage)*($Current*$Resistance))))/(2*$Voltage),1)*$Voltage-C8

(produces a curved output line)

Armature Current Limiting should produce this formula for Column B in the spreadsheet:

=MIN($Voltage-C8,$Current*$Resistance)

(produces a straight output line)

----------------

The "bottom line" is that Armature Current Limiting is supposed to maintain a limit of current that goes through the motor that should be flat.

safe
safe's picture
Offline
Joined: 08/03/2010
Points: 583
Re: Performance of Unite my1020 1000W motor

mc-taichi wrote:

I think i1=i2=i3 in above figure because of single way of wiring at this part. Are there any misunderstanding?

The theory is simple.

Battery Current Limiting measures the battery side current and produces a declining current going from low to high rpm.

Armature Current Limiting measures the motor side current and produces a flat current going from low to high rpm.

------------------

In practice... right now I'm suffering from the worst cold I've had in several years and can barely think, so I'm not going to try to figure out your circuit, but my guess is that you are presently experiencing Armature Current Limiting.

Change Column B back to:

=MIN($Voltage-C8,$Current*$Resistance)

...there are many factors in the entire system (battery, controller, sensors) that complicate getting accurate data.

Cheetah
Offline
Joined: 01/28/2011
Points: 3
Re: Performance of Unite my1020 1000W motor

SOUNDS GOOD TO ME

ed hardy
ed hardy
ed hardy

__________________

Cheetah

mc-taichi
mc-taichi's picture
Offline
Joined: 04/14/2010
Points: 18
Re: Performance of Unite my1020 1000W motor

Long time no see.

I found a new 1000W motor in TNC scooters.
http://tncscooters.com/index.php?route=product/product&path=52_56&product_id=512

The picture is same as ones listed in e-bay mentioned in the comment of this topic before. But it costs double.
I wonder what is the difference between them and if I should choose TNC's one if I want more powerful and durable one.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Customize This