Connecting two BLDC motors in series?
- Thread starter Papa
- Start date
Link
1 MW
In theory, if you were able to weld the axles together, they could probably be run in a sort of weird series. Can't be run in series without sharing an axle or whatever because their phases would need to be on at different times.
Bummer. The motors need to be mechanically isolated for this build - one on each rear wheel. During turns, the wheels will be rotating at different speeds. I have two BLDC's and would dearly love to use them for this app. As I see it, and please correct me if i'm wrong, the downside to using a separate controller for each motor, is the conflicting loads created during turns - the inside wheel drawing higher amperage - pushing the system into inefficient territory.Link said:
Link
1 MW
Papa said:
Not really a problem, I hear. It just feels like it has...what was it called...a limited-slip differential or something. You're not going to be drawing much power at low speeds, and you're not going to be doing 5ft radius turns (where the difference in RPM might be a small problem at high power) at 20mph.
Only major downside is the cost of another controller. :?
I'm not so sure, Link. This trike has 33f/66r weight distribution (close to the same as the HandiZap conversion I built. Needless to say, front wheel traction gets iffy on gravel during hard U-turns. The Zap, on asphalt, can also do a U-turn just inches longer than its 55" wheelbase - meaning the inside wheel is nearly motionless during tight radius U-turns.
That said, seems my electronic "posi" would likely slide the front wheel under graveled conditions... or push up the amps on the inside wheel on asphalt. Dunno
That said, seems my electronic "posi" would likely slide the front wheel under graveled conditions... or push up the amps on the inside wheel on asphalt. Dunno
RLT
10 kW
Well, even if you could overcome the brushless sync on one controller problem, you still wouldn't want to run them in series. In series, they act like a FULL slip differential.
You would do it in PARALLEL, to get a limited slip differential type drive.
At least as far as my BRUSHED 2WD build experiences show, the only way to get full time all wheel drive is to have all motors have their own isolated controllers and battery packs. They can't even share a single throttle. I have no idea as to why this is the case, but my real-world experience has shown it to be the case, at least with the motors, and (BD36) controllers I tried it on.
You would do it in PARALLEL, to get a limited slip differential type drive.
At least as far as my BRUSHED 2WD build experiences show, the only way to get full time all wheel drive is to have all motors have their own isolated controllers and battery packs. They can't even share a single throttle. I have no idea as to why this is the case, but my real-world experience has shown it to be the case, at least with the motors, and (BD36) controllers I tried it on.
Link
1 MW
Like I said, you won't be doing 0 radius turns at speed, so the amount of power going to the wheel is going to be small and can be totally burned off without ill effect.
Can't really comment on RLT's 2WD thing, but they shouldn't have to have separate packs. Separate throttles, perhaps, as the motors will likely have a slightly different voltage constant (since they can't be made exactly alike) and one will be trying to take more of the load, but all the pack is is an indifferent power source. Would be sort of like two gas tanks in a car with two engines; unnecessary. :?
Can't really comment on RLT's 2WD thing, but they shouldn't have to have separate packs. Separate throttles, perhaps, as the motors will likely have a slightly different voltage constant (since they can't be made exactly alike) and one will be trying to take more of the load, but all the pack is is an indifferent power source. Would be sort of like two gas tanks in a car with two engines; unnecessary. :?
RLT
10 kW
Link said:
Heck, I don't understand it; Fechter? and a couple of the other good electronics guys couldn't explain why it was happening on my first build configuration.... but nonetheless, that's the way it worked. maybe it was just some idiosyncrasies of the WE BD36 controllers, but when I blew one of my controllers and cut the power lines to it, it still spun both motors (with the bike off the ground) and the only common connection was the throttle. You could stop either motor by hand or brake, with no noticeable problem; and if you stopped both motor by hand or brakes, the controller wires would get hot quickly. Totally weird, but true, nonetheless.
Well, I just finished connecting two brushed PM motors in series, and they peformed as I suspected;RLT said:
Both ran the same RPMs when unloaded. When I grabbed the armature of one motor, and eventually stopped it, the other motor sped-up. Emulating the perfect, electronic differential - or so it first appears. Am I missing something here?
I had planned to do just that, but as I said eariler, I was hoping to achieve a similar condition with brushless - knowing full well it many not be easily accomplished, if at all.
RLT
10 kW
Hmm... it wasn't REALLY easy to stop the first motor (beyond overcoming centrifugal force)?
Every time I have tried putting two motors in series, it only took two fingers pinching the motor shaft, to stop one quite easily, even when they were up to speed .... I have tried it on 250 watt currie bike motors and on 3/4 horsepower treadmill motors.
To get a better idea as to what will happen, try this: with the power off and the motors at rest, prevent them from turning, then apply power. See what happens, then release them. and intermittently apply braking force to either or both motors at random intervals. Then try the same thing with them in parallel. See which seems to work better for you. I'm thinking you will see more applicable power, but with enough slippage or differential action to prevent problems in your application, with them in parallel. But if I'm wrong... well, it won't be the first time today that I was wrong about something I was pretty sure of.
Every time I have tried putting two motors in series, it only took two fingers pinching the motor shaft, to stop one quite easily, even when they were up to speed .... I have tried it on 250 watt currie bike motors and on 3/4 horsepower treadmill motors.
To get a better idea as to what will happen, try this: with the power off and the motors at rest, prevent them from turning, then apply power. See what happens, then release them. and intermittently apply braking force to either or both motors at random intervals. Then try the same thing with them in parallel. See which seems to work better for you. I'm thinking you will see more applicable power, but with enough slippage or differential action to prevent problems in your application, with them in parallel. But if I'm wrong... well, it won't be the first time today that I was wrong about something I was pretty sure of.
Link
1 MW
RLT said:
Yes, it's quite easy to stall one without them moving anything because the other one's BEMF makes it so that the stalled motor won't be able to draw any more power than the other one does at no load. Try stopping them both at the same time...
With two motors in series, the BEMF the controller/battery sees is the total BEMF of both motors. So, when both are at no load, they both produce half of the batteries voltage and run at half speed. When one is stalled, the other motor speeds up because it's getting full battery voltage across its terminals.
Yes, you can run two brushed motors in series or parallel with a single controller. In series it will act like a differential. In parallel it will act like a limited slip differential.
With brushless motors, there is no way to run two motors off one controller without welding the shafts together. You pretty much have to run the two controllers in parallel. In a really tight turn, the inside motor might go into current limiting, but it should not be a problem.
With brushless motors, there is no way to run two motors off one controller without welding the shafts together. You pretty much have to run the two controllers in parallel. In a really tight turn, the inside motor might go into current limiting, but it should not be a problem.
rkosiorek
100 kW
you can do the brushless with 2 controllers. you just need a way of connecting single throttle to both controllers.
View attachment 1
this also includes a few adjustments to balance the speed of both motors and to make sure they both start turning at the same time.
rick
View attachment 1
this also includes a few adjustments to balance the speed of both motors and to make sure they both start turning at the same time.
rick
diver
1 kW
I'm going to run two eteks in my trike in parallel off one controller just for the fun of it. that is my winter project and i cant wait to get started on it.
rkosiorek
100 kW
etek's are brushed. so no problem.
why not try the Kill-a-cycle thing of running them in series for max torque at start up and switch to parallel for max speed? just need a big switch or relay to do that.
rick
why not try the Kill-a-cycle thing of running them in series for max torque at start up and switch to parallel for max speed? just need a big switch or relay to do that.
rick
Papa said:
With brushless motors, if one tries to go faster than the other (outside wheel during turn), it will simply freewheel. Even with brushed motors in parallel, it's not like they're locked together. For normal turning speed differences, I don't think you'd even notice the drag. At low speeds, like tight turns, the "lockup" would be minimal and it would behave more like a normal differential.
johnrobholmes
10 MW
I see no problem with running two BLDC motors together, but they would need separate controllers at least. The turn radius and tire scrub wouldn't be an issue at all IMO. I have never tried the setup, but my experience with dual motor systems in ground RC reinforces this. Individual throttles may be very fun though 
If the build had 4 wheels and 50/50 weight distribution, I'd try it. Unforunately, this is a trike with only 33% of the laden weight on its single steer axle. My previous trike build (with diff) exhibits mild understeer at speed, and u-turns on gravel isn't exactly textbook - posi would no doubt exacerbate both conditions. As such, with brushless, I'd feel much better if both motors electronically communicated with each other. Maybe there's a way, maybe not, dunno.johnrobholmes said:
frodus
10 kW
diver said:
It'l work if they're the brushed type..... if they're BLDC, they won't work with one controller. A controller has only one control loop.
frodus
10 kW
Mathurin said:
why would you use 3 controllers, you need 1 per motor. You do NOT put a controller in front of a controller. They control themselves. You would have to split the throttle input by tying the throttle groung together.
Yeah but two brushless controllers with split throttle is a parallel setup. That should work like an speed sensing LSD.
The way I see to get series requires the brushless controllers to function like brushes would, 100% duty cycle, no current limit and no voltage limit. 5v line must be fed from brushed controller to the brushless ones so they can keep switching at low voltages. Power would be modulated using the brushed controller, used like the single controller would be in a series brushed setup. That should work like an open differential.
The way I see to get series requires the brushless controllers to function like brushes would, 100% duty cycle, no current limit and no voltage limit. 5v line must be fed from brushed controller to the brushless ones so they can keep switching at low voltages. Power would be modulated using the brushed controller, used like the single controller would be in a series brushed setup. That should work like an open differential.
317537
10 kW
- Joined
- Oct 19, 2008
- Messages
- 939
2 hubs in series just takes you load resistance to double, you wont sync them with a single controller, you need two controllers and that blows the series thing out of the water. Hmmm.
Cant run the hubs in paralell on one controller as you will have two seperate sensor pulses, unless you join the hubs and add a diff and you will take the load resistance down placing to much stess on the controller.
Need two controllers and two hubs.
Tinkering is fun though
Off topic
Just an Idea!
Replace your fet array... That may work nice. Bypass your fets and take the sigs from the signal emiiters. Hook them up to the gates of a huge fet amplifer array with 300 amp 144 volt power supply and you have a a very very powerfull controller for cheaper than you can buy one for
I really want to get my fets far away from PWM PCB and fuse it off ..
I was thinking of using a modified audio power amp to do this in order to deintergrate the fets. It might work...
1: you have the pwm on its own PCB and regulators, you could use any power regs you want, tap from either main supply, or voltage tap batteries, use a dc to dc step down converter or even a use sepperate supply voltage (best).
2: You can supply any voltage to your fets source without blowing your pwm. As long as the fets are in spec with the applied voltage (source and gate) it will work.
3: You can seal the fet array from weather and use a mother heatsink and have it anywhere in the breeze, and protect your PWM circuit from weather.
Its really audio technolgy with DC offset re-invented for PWM motor application. You have a pre amp for (PWM) and a power amp (Power FET array)
As to why when someone fries a controller they need to be replacing the whole dang controller is beyond me. I can power a 10 k watt power amp with most sensitive expensive DSP effects rack supplying the signals and nothing will blow each other up. Sure I'll blow an amp if I turn it up to much or ill blow my expensive effects rack up if I spill my beer on it.. But very very rarely you will take out everything in one go, unlike our PWM controllers..
Just a thought.
Cant run the hubs in paralell on one controller as you will have two seperate sensor pulses, unless you join the hubs and add a diff and you will take the load resistance down placing to much stess on the controller.
Need two controllers and two hubs.
Tinkering is fun though
Off topic
Just an Idea!
Replace your fet array... That may work nice. Bypass your fets and take the sigs from the signal emiiters. Hook them up to the gates of a huge fet amplifer array with 300 amp 144 volt power supply and you have a a very very powerfull controller for cheaper than you can buy one for
I really want to get my fets far away from PWM PCB and fuse it off ..
I was thinking of using a modified audio power amp to do this in order to deintergrate the fets. It might work...
1: you have the pwm on its own PCB and regulators, you could use any power regs you want, tap from either main supply, or voltage tap batteries, use a dc to dc step down converter or even a use sepperate supply voltage (best).
2: You can supply any voltage to your fets source without blowing your pwm. As long as the fets are in spec with the applied voltage (source and gate) it will work.
3: You can seal the fet array from weather and use a mother heatsink and have it anywhere in the breeze, and protect your PWM circuit from weather.
Its really audio technolgy with DC offset re-invented for PWM motor application. You have a pre amp for (PWM) and a power amp (Power FET array)
As to why when someone fries a controller they need to be replacing the whole dang controller is beyond me. I can power a 10 k watt power amp with most sensitive expensive DSP effects rack supplying the signals and nothing will blow each other up. Sure I'll blow an amp if I turn it up to much or ill blow my expensive effects rack up if I spill my beer on it.. But very very rarely you will take out everything in one go, unlike our PWM controllers..
Just a thought.
johnrobholmes
10 MW
Papa said:
I mean that dual motors won't present tire scrub or traction issues when cornering, except at very slow speeds where the inside tire will try to drive the rig fully. It may spin out if you give too much throttle.
