09-08-2010, 12:20 PM
|
#321 (permalink)
|
|
EcoModding Lurker
Join Date: Feb 2010
Location: Ann Arbor, MI
Posts: 53
Thanks: 0
Thanked 12 Times in 8 Posts
|
How does a three-phase motor work? Does each phase power an specific part of the motor? If so it seems like it would act as 3 single phase motors in one in terms of power use. If it uses all three phases at the same time then I could see how it could be different. So complicated! Sorry for all the questions...
|
|
|
|
 Today
|
|
|
|
Other popular topics in this forum...
|
|
|
|
|
09-08-2010, 12:23 PM
|
#322 (permalink)
|
|
PaulH
Join Date: Feb 2008
Location: Maricopa, AZ (sort of. Actually outside of town)
Posts: 3,832
Thanks: 1,368
Thanked 1,202 Times in 765 Posts
|
There's a rotating magnetic field, by changing the 3 stator currents (which are physically separated by 120 degrees). The changing current in the stator causes current to move in the rotor loops (the squirrel cage loops), which causes a magnetic field in a rotor loop that attracts to the magnetic field in the stator coil that it's close to.
|
|
|
|
|
09-08-2010, 12:35 PM
|
#323 (permalink)
|
|
EcoModding Lurker
Join Date: Feb 2010
Location: Ann Arbor, MI
Posts: 53
Thanks: 0
Thanked 12 Times in 8 Posts
|
So it seems like those 3 stator currents would be the 212amps rms and 102V rms right?
|
|
|
|
|
09-08-2010, 12:39 PM
|
#324 (permalink)
|
|
EcoModding Lurker
Join Date: Jul 2010
Location: Cleveland, OH
Posts: 89
Thanks: 5
Thanked 7 Times in 5 Posts
|
part of the reason 1+1+1 is adding up to less than 3 is that you have some recirculating currents that the individual legs see, but the total input does not. My earlier post was simplifying things somewhat.. I'd need to crack open the textbook to get a better answer.
Also, it IS totally possible to broaden the voltage curve to squeeze in more power.. some drivers call it a trapazoid curve option. I wouldn't mind trying that, maybe have it kick in when you floor it since noise/efficiency wouldn't really be a concern at that time.
|
|
|
|
|
09-08-2010, 12:39 PM
|
#325 (permalink)
|
|
PaulH
Join Date: Feb 2008
Location: Maricopa, AZ (sort of. Actually outside of town)
Posts: 3,832
Thanks: 1,368
Thanked 1,202 Times in 765 Posts
|
Right. Each stator current would be 212 amp rms, and the instantaneous currents would add up to 0.
|
|
|
|
|
09-08-2010, 12:48 PM
|
#326 (permalink)
|
|
needs more cowbell
Join Date: Feb 2008
Location: ÿ
Posts: 5,038
Thanks: 158
Thanked 269 Times in 212 Posts
|
this of course assumes sine wave drive.
but the current through a leg is determined by the sine of its angle, i.e. leg 1 @ 90, leg 2 @ 210, leg 3 @ 330 would be 1, -.5, -.5 which as noted sums to zero. If you examine it at 90 electrical degrees then current in leg 1 should be the max current a leg will see.
__________________
WINDMILLS DO NOT WORK THAT WAY!!!
|
|
|
|
|
09-09-2010, 12:32 PM
|
#327 (permalink)
|
|
EcoModding Lurker
Join Date: Mar 2010
Location: South Africa
Posts: 16
Tazzy - '02 Toyota Tazz Sport 90 day: 43 mpg (US)
Thanks: 5
Thanked 3 Times in 3 Posts
|
If you use field orientated control you get about 15% extra (IIRC) out of the same supply compared to normal sinewaves as is used by V/Hz control. There is very little to be gained in a system, apart from electronic switch efficiencies and careful design, when using FOC compared to other strategies. It also give low-end torque which is lacking in V/Hz schemes.
Unfortunately, using FOC means that the drive is doing funny things to the motor, and the current seen by each phase make sense to the controller only. We therefore have to rely on what the drive tells us, or measure on the DC side, and calculate what it would be if it was a V/Hz drive.
Dawid
|
|
|
|
|
09-09-2010, 12:44 PM
|
#328 (permalink)
|
|
PaulH
Join Date: Feb 2008
Location: Maricopa, AZ (sort of. Actually outside of town)
Posts: 3,832
Thanks: 1,368
Thanked 1,202 Times in 765 Posts
|
Right, you get 2/sqrt(3) more voltage, and you get to use a PI loop so you get all the benefits of the ease of controlling a sepex DC motor. Throttle proportional to current doesn't work very well with V/Hz. It's much easier to keep a good handle on exactly how big the current is getting with a PI loop.
|
|
|
|
|
09-10-2010, 12:13 PM
|
#329 (permalink)
|
|
ReVolt Enthusiast
Join Date: Jun 2009
Location: Michigan, USA
Posts: 239
Thanks: 98
Thanked 47 Times in 40 Posts
|
Preliminary SR Controller !!!
I just added Paul's preliminary SR Controller to the Wiki.
Check it out ......
Open ReVolt/PCB Schematics - EcoModder
Regards, 
-Mark |
|
|
|
|
09-17-2010, 07:33 AM
|
#330 (permalink)
|
|
PaulH
Join Date: Feb 2008
Location: Maricopa, AZ (sort of. Actually outside of town)
Posts: 3,832
Thanks: 1,368
Thanked 1,202 Times in 765 Posts
|
OK, so I've been super bad about the SR controller. Here's the low-down on everything:
I hurried to get the previous SR control board out and it was $350 that I didn't have. I made a mistake so that only one of the 2 low voltage cutout circuits will work properly. So, school was just getting out for the summer, so I was going to have 4 weeks of unpaid vacation (read unemployed) until school started again this coming Monday. So, needless to say, I was pretty depressed. What does one do when there is no money, but time? SOFTWARE! But what software? What's nice and challenging, and isn't mathematically boring at all, like the DC software is becoming? AC! So, I went to work on that, and on the AC control board schematic, and on an AC mosfet based low voltage power section. 144v battery pack 300amp per phase AC. None of which costs anything except time. And man did it take time. When not jumping on the trampoline with my son or talking with my wife, I was working on all that for like 12-16 hours per day. It sounds like a lot, but when you don't sleep very much that frees up all sorts of extra time. So, I got an AC control board made for a $25 special at some board house, and I have been populating it the last few days. Here's a video of the control board and the basic IGBT based power section:
|
|
|
|
|
The Following 5 Users Say Thank You to MPaulHolmes For This Useful Post:
|
|
|
