PDA

View Full Version : Series Parallel Switching - DC and AC



Nuts & Volts
20 August 2010, 0856
Ok I figured I would bulk this into one main post. Edit me if I am a little bit off on my claims here.

The idea behind this set-up is to use two electric motors, a compatible controller, and some extra relays/contactors to perform the switching. This idea is basically a 2 speed electric transmission. First volt equates to speed/rpm and amps equate to torque. So on acceleration (low speed) you have the two motors connected in series this way both motors see the same amount of amps, but half voltage to each. Then you switch the motors to parallel so that the amps are halved between the motors and the voltage is equal. Series provides more torque to take off and parallel provides a higher operating speed.

So lets make an example to fully understand the whole situation. You have a 72V battery system, one controller, and two motors. You wire up contactors so that you can change the wiring of the motors (ill post a diagram later). Ok so you start the motors in series and pull the throttle so that 100 amps is drawn. That means at max speed the each motor will see 36V (based on equal motor resistances). But each motor will be getting 100 amps each so that is 200amps worth of torque. The power draw from the battery is at 7200 watts. When you reach the maximum speed of the motors at 36V you switch the motors to parallel setup. This allows both motors to see 72V, which will double the operating speed of the motors. But the amps will be halved and each motor will get 50 amps. The battery will be supplying 72V and 100amp (or 7200W) in both motor configurations.

series: motor has 36V x 100A = 3600W ------ torque=2, speed=1, power=2
parallel: motor has 72V x 50A = 3600W ------ torque=1, speed=2, power=2

Advantages: More torque without pulling more amps, saves your battery, faster acceleration, low gear for hill climbing
Disadvantage: Complexity (some), cost of two motors, need compatible motors, anymore?????

Only question I have, is it possible to series parallel switch two AC motors or brushless motors?

Nuts & Volts
20 August 2010, 0920
Ok thanks Ed. Ill change the coloring (thought I would try to be creative with my post!!). I think electricracebikes MAY have had this switching because they had a Zilla which can perform the switching the automatically.

Also I asked about the AC motors, not because of the speed, but of the amp draw on the batteries. Instead of having 1 big AC motor draw 400A from the battery and get 100Nm of torque. I would have two smaller motors that each get 200A but only draw 200A from the battery. This way I am getting the same amount of torque (if done right) but only running the batteries at half the draw. Unless I am wrong about the amp draw???

magicsmoke
20 August 2010, 1023
An alternative is to use one 'meaty' motor and series / parallel switch the battery.
i.e. parallel cells high current capacity, series cells high voltage.

Rob

jpanichella
20 August 2010, 1023
I was reading a thread over at endless sphere where somebody did something similar for a race bike. It's probably the best way to get the most performance with the least money, but if I remember correctly he complained about the complexity involved in the drivetrain and wiring.

frodus
20 August 2010, 1049
An alternative is to use one 'meaty' motor and series / parallel switch the battery.
i.e. parallel cells high current capacity, series cells high voltage.

Rob

except your controller probably won't work with dual voltage. Not sure if any controllers will allow their input voltage to be switched like that, not without a fault.

frodus
20 August 2010, 1109
series parallel switching on the motor side is a cool idea. You need a controller that can support it (FETs/IGBT's need to be turned off briefly during the switch). I know drag racers use series parallel switch to act as a transmission. It's like you described, in series to start, each gets 1/2 voltage but full current. Then they switch and get full voltage, but 1/2 the current. You just need to make sure that your motors can operate at pack voltage (i.e. don't run a 48V motor at 144V) and that your controller supports it.

Doing series parallel switching on an AC isn't possible AFAIK. I know some guys on ES have done delta/wye switching on BLDC motors, but I'm not sure about the details, and its only with one motor. I don't think it can be done with AC the way you're thinking.



Now, about power draw from batteries. Power is power. If you're drawing 7200W (whether in series or parallel) you're still drawing 7200W. if its a 72V pack, you're still drawing 100A either way. 36V each motor and 200A is still 72V 100A on the battery side. Try to imagine the controller as a power converter. 200A motor side doesn't neccessarily mean that you're pulling 200A off the batteries. On my motorcycle, with a series wound motor and synkromotive controller, I would see 400+ amps on the motor side, but I'd only draw 150A or so on the battery side.

Nuts & Volts
20 August 2010, 1306
Ok, good i am glad i didnt just make all that up. And good to know about the AC motors. Ill look up the brushless idea.

As for battery switching it sounds like most contollers wont like it.

magicsmoke
20 August 2010, 1429
Hmm, we've gone from a suggestion to a 'not sure if any' to a 'sounds like most won't' in 3 posts.
I think perhaps a trial first!

The only reasons I can see why a controller would not like to have its supply doubled / halved is if the internal regulator's reservoir caps are not capable of 'holding up' the control cctry during 'switch over' (at zero pwm) and whether or not the main caps could handle the current inrush / outrush ( ala precharge resistor). Neither issue insurmountable.

Here's another one, akin to series / parallel motor switching.
If a dc motor winding is done in say two halves, connections to which are made accessible, then you can simply connect these halves in either series or parallel with exactly the same effect as though it were two separate motors. i.e. parallel - high speed / low torque, series - low speed / high torque.

caveat: of course there are details :)

Rob

frodus
20 August 2010, 1500
You're completely right, as long as the circuitry can handle s/p on the batteries its fine, but most controllers have a cutoff voltage (high/low), so if you halved it, it will fault. Also, if you interrupt power while the throttle is on, you can't restart most controllers if its not at 0. There may be other delays in startup as well.

I'm still not getting why anyone would want to s/p switch batteries though. The controller is a power converter. Watts is Watts. For 7200W to the controller, you'd need 72V and 100A or 36V and 200A, still 100A from each battery. Higher voltage means smaller cables too. Just do one series string and be done with it and let the controller do its job, right? Its not a matter of "can it be done", its more of a "why would you wanna do that" :)


As far as series wound motors go, paralleling the windings doesn't work well, if at all. The resistances don't match equally, so the current will go to one side more than the other.

magicsmoke
20 August 2010, 1545
Doh! That's not the first time either that I've fallen in the s/p battery trap :)

As for the switched motor winds, it doesn't matter whether the current shares equally as the resultant forces are additive.
Actually, there's not even any reason to reconnect the windings in parallel, all we really want to achieve is a different number of turns i.e. a tap (or multiple taps). More turns - lower speed, less turns - higher speed.

magicsmoke
20 August 2010, 1601
A little bit of useless while I'm on turns.
The Agni model numbers relate to turns and you can see that their respective rpm/volt range accordingly

model - rpm/V
95 / 71
119 / 58
135 / 50
143 / 47

frodus
20 August 2010, 1610
Doh! That's not the first time either that I've fallen in the s/p battery trap :)

As for the switched motor winds, it doesn't matter whether the current shares equally as the resultant forces are additive.
Actually, there's not even any reason to reconnect the windings in parallel, all we really want to achieve is a different number of turns i.e. a tap (or multiple taps). More turns - lower speed, less turns - higher speed.

I remember a while back talking to the guys at Synkromotive about it, we never tried it, but essentially you want all of the current to go through both field and armature. if one is lower than the other, the motor doesn't run as well. Another issue: Runaway. Series motors can run away, especially if the field winding is weakened.

Oh, and the coils may not like full pack voltage on them either, since they're made to essentially split the voltage between the two coils.

Maybe someone here can try it :)

magicsmoke
20 August 2010, 1614
Must confess, I wasn't even thinking of the wound field types, simply the PM, but now you've mentioned it I'll 'brush up' . No pun :)

frodus
20 August 2010, 1658
Thats what is so great about this forum, we can discuss all of these ideas, learn, share....

I'm not entirely sure how well (or unwell) the motor would work though, might be worth trying. I know John Alder has a series wound motor that he might be able to try out :)

Tony Coiro
20 August 2010, 2203
I hope this hasn't been mentioned, partially because it would be redundant but mostly because it is probably really dumb. There is a way you could parallel/series switch BLDC or AC....... you could do it battery side. In the 2 minutes of thought I have given this, I can't think of any reason this would be a good idea.

GUFF
21 August 2010, 0809
I remember reading something similar to what Rob and Frodus were talking about a long time ago when I was active on the EVList. Attached is a Jpeg of the PDF that describes the theory behind the Series wound motor modification.

magicsmoke
21 August 2010, 1004
That's pretty interesting Guff. Thank's for that.
Have to think about it a bit before making any stupid or otherwise comments :)

Rob