hub with 6 phase wires

Nobody wrote yet if it is more powerfull than qs 3kw v3 but lets assume it is and it's worth spending extra money on second Sabvoton controller.

How two 3 phase motors can work with two separate controllers without synced phases. Folks are saying that you just hook up each motor separate controller and use one throttle and all it goes well.
 
minde28383 said:
[
First controller - 100A batt / 250A phase; second controller - 100A batt / 250A phase

Constant 2x100A x 72v = 14.4kw
Phase 2x250A x 72v = 36kw

Only the battery current & voltage above is correct. The phase current will be 250A *at some time* but only at a lower voltage. The controller is a buck converter, it can only increase phase current above battery current if the voltage is less. Power input-to-output must remain constant (less controller losses [~1%]). It can't buck if the output is also 72V!
 
minde28383 said:
How two 3 phase motors can work with two separate controllers without synced phases. Folks are saying that you just hook up each motor separate controller and use one throttle and all it goes well.
Right. In most cases you can just common the grounds between both controllers and send the throttle signal to both. Works best with a torque (current mode) throttle, but will work OK with a voltage mode throttle as well.
 
Yes... and then the offset of the separate Hall sensors does the timing of the two 3 phase sections of the windings. Its a totally different thing than trying to sync up two controllers into one set of 3 phase windings.. then they have to be sending out exactly matched power pulses, and when they need to be master/slave setup. Otherwise you get one trying to fire into a winding at the same time the other is trying do do the reverse polarity in the same winding. But with the 2 independent 3 phase interlaced that never happens.

I'll try to post some better pics later of the actual wiring hookups on my 6 phase from the previous picture while I have it open..
 
Punx0r said:
minde28383 said:
[
First controller - 100A batt / 250A phase; second controller - 100A batt / 250A phase

Constant 2x100A x 72v = 14.4kw
Phase 2x250A x 72v = 36kw

Only the battery current & voltage above is correct. The phase current will be 250A *at some time* but only at a lower voltage. The controller is a buck converter, it can only increase phase current above battery current if the voltage is less. Power input-to-output must remain constant (less controller losses [~1%]). It can't buck if the output is also 72V!

Hi Mate,
This is first time I see someone using word "buck" converter when talking about controllers. And mentioning such thing like, - "increase phase current above battery current if the voltage is less". I know just principal that battery voltage drops more the more current you draw from battery the longer the time you do current draw. Never thought in a way that controller is DC-DC converter.
What did you exactly mean by, - "The controller is a buck converter, it can only increase phase current above battery current if the voltage is less"? Did you wanted to say that I won't have 72v during heavy load times?, eh this is just my guess.

If not wikipedia I woudn't even know what "buck" is, ie.:
A buck converter (step-down converter) is a DC-to-DC power converter which steps down voltage
 
Voltron said:
Yes... and then the offset of the separate Hall sensors does the timing of the two 3 phase sections of the windings. Its a totally different thing than trying to sync up two controllers into one set of 3 phase windings.. then they have to be sending out exactly matched power pulses, and when they need to be master/slave setup. Otherwise you get one trying to fire into a winding at the same time the other is trying do do the reverse polarity in the same winding. But with the 2 independent 3 phase interlaced that never happens.

I'll try to post some better pics later of the actual wiring hookups on my 6 phase from the previous picture while I have it open..

There are people who show you that you know nothing. This is very helpful and revealing. Thank you.
Pics would be much appreciated.

Here is not so short question from my share interest.
Why someone would choose to run such 6 phase motor which requires two controllers when mostly nowadays there are powerful 3 phase motors which are driven by only one powerful controller, be it sin or improved new design trapolized.
Some time ago when sin controllers where not so present EV builders might wanted to use 6 phase due smoother and more efficient ride with two trapolized controllers; and some people say because of two cheaper controllers instead one powerful, but why nodays to make such 6 phase setup. Please note, I'm not claiming one is better than another, but I am interested about your opinion about two controller setup.

So as my motor has two sets of halls; and it is designed to be run by separate controllers, ie. - "offset of the separate Hall sensors does the timing of the two 3 phase sections of the windings." Luckily it was not designed and and meant to be run by synced controllers as I was starting to believe very recently.
So basically these two motors does not need to know about each other position and power state because both motors act only and only according their hall sensors. And they are not interfering each another because halls are in proper places by design. Correct?
 
I found Steve Cesario on Youtube. He runs the same but one inch bigger hubmonster in ebike frame. Says he does 32kw with two Adaptos. Who knows if he is here in ES?
https://www.youtube.com/watch?v=LXpgRH35bg8
 
minde28383 said:
I found Steve Cesario on Youtube. He runs the same but one inch bigger hubmonster in ebike frame. Says he does 32kw with two Adaptos. Who knows if he is here in ES?
https://www.youtube.com/watch?v=LXpgRH35bg8
Already found him in ES. His nick: steveo
 
minde28383 said:
This is first time I see someone using word "buck" converter when talking about controllers. And mentioning such thing like, - "increase phase current above battery current if the voltage is less". I know just principal that battery voltage drops more the more current you draw from battery the longer the time you do current draw. Never thought in a way that controller is DC-DC converter. What did you exactly mean by, - "The controller is a buck converter, it can only increase phase current above battery current if the voltage is less"? Did you wanted to say that I won't have 72v during heavy load times?, eh this is just my guess.
Buck converters are converters that efficiently step down voltages.

You can step down voltage with a resistor - but to do that you have to dissipate the rest of the voltage as heat. Let's say you want 5 volts/1 amp out, but you have 10 volts in. You could use a resistor to drop that down (and in fact that's how linear regulators work.) But if you want 5V 1A (5W) out, that means you have to also dissipate 5 watts in a resistor or other device somewhere. (10 watts in, 5 watts out, 5 watts dissipated as waste heat.) Very inefficient - and hot.

A buck converter is more efficient. Instead of a resistor it uses a switch, a diode and an inductor. It opens and closes the switch rapidly and the inductor "smooths out" the choppy voltage that results. (Also generally uses a capacitor for filtering, but that's not essential.) Now if you want 5 volts 1 amp out you only need to supply 10 volts at half an amp - much more efficient.

Now, a controller for a 3 phase BLDC motor is not designed to be a buck converter. But it has all the same parts. It has a switch (the FET) a diode (the body diode of the opposing FET) and an inductor (the motor winding.) So if you do PWM drive of the FET - which most controllers do - you have, effectively, a buck converter. This is great, because it's more efficient. And since you get the effect of a buck converter, it's possible to get 1 amp through the winding when the motor is moving slowly (=lower voltage needed) when you are drawing half an amp from the battery. That means more torque at low speeds.
 
billvon said:
minde28383 said:
This is first time I see someone using word "buck" converter when talking about controllers. And mentioning such thing like, - "increase phase current above battery current if the voltage is less". I know just principal that battery voltage drops more the more current you draw from battery the longer the time you do current draw. Never thought in a way that controller is DC-DC converter. What did you exactly mean by, - "The controller is a buck converter, it can only increase phase current above battery current if the voltage is less"? Did you wanted to say that I won't have 72v during heavy load times?, eh this is just my guess.
Buck converters are converters that efficiently step down voltages.

You can step down voltage with a resistor - but to do that you have to dissipate the rest of the voltage as heat. Let's say you want 5 volts/1 amp out, but you have 10 volts in. You could use a resistor to drop that down (and in fact that's how linear regulators work.) But if you want 5V 1A (5W) out, that means you have to also dissipate 5 watts in a resistor or other device somewhere. (10 watts in, 5 watts out, 5 watts dissipated as waste heat.) Very inefficient - and hot.

A buck converter is more efficient. Instead of a resistor it uses a switch, a diode and an inductor. It opens and closes the switch rapidly and the inductor "smooths out" the choppy voltage that results. (Also generally uses a capacitor for filtering, but that's not essential.) Now if you want 5 volts 1 amp out you only need to supply 10 volts at half an amp - much more efficient.

Now, a controller for a 3 phase BLDC motor is not designed to be a buck converter. But it has all the same parts. It has a switch (the FET) a diode (the body diode of the opposing FET) and an inductor (the motor winding.) So if you do PWM drive of the FET - which most controllers do - you have, effectively, a buck converter. This is great, because it's more efficient. And since you get the effect of a buck converter, it's possible to get 1 amp through the winding when the motor is moving slowly (=lower voltage needed) when you are drawing half an amp from the battery. That means more torque at low speeds.

That is the most cogent explanation of a (step-down) converter I've ever read. Thanks.

But ... you knew it was coming right? ... what about a step-up converter?

Eg. 36v to 48v
 
billvon said:
minde28383 said:
This is first time I see someone using word "buck" converter when talking about controllers. And mentioning such thing like, - "increase phase current above battery current if the voltage is less". I know just principal that battery voltage drops more the more current you draw from battery the longer the time you do current draw. Never thought in a way that controller is DC-DC converter. What did you exactly mean by, - "The controller is a buck converter, it can only increase phase current above battery current if the voltage is less"? Did you wanted to say that I won't have 72v during heavy load times?, eh this is just my guess.
Buck converters are converters that efficiently step down voltages.

You can step down voltage with a resistor - but to do that you have to dissipate the rest of the voltage as heat. Let's say you want 5 volts/1 amp out, but you have 10 volts in. You could use a resistor to drop that down (and in fact that's how linear regulators work.) But if you want 5V 1A (5W) out, that means you have to also dissipate 5 watts in a resistor or other device somewhere. (10 watts in, 5 watts out, 5 watts dissipated as waste heat.) Very inefficient - and hot.

A buck converter is more efficient. Instead of a resistor it uses a switch, a diode and an inductor. It opens and closes the switch rapidly and the inductor "smooths out" the choppy voltage that results. (Also generally uses a capacitor for filtering, but that's not essential.) Now if you want 5 volts 1 amp out you only need to supply 10 volts at half an amp - much more efficient.

Now, a controller for a 3 phase BLDC motor is not designed to be a buck converter. But it has all the same parts. It has a switch (the FET) a diode (the body diode of the opposing FET) and an inductor (the motor winding.) So if you do PWM drive of the FET - which most controllers do - you have, effectively, a buck converter. This is great, because it's more efficient. And since you get the effect of a buck converter, it's possible to get 1 amp through the winding when the motor is moving slowly (=lower voltage needed) when you are drawing half an amp from the battery. That means more torque at low speeds.

That's so cool to find out correlation between buck converter and controller we use in our ebikes.

Thinking over and over and it might be I comprehended something. I might understood what you wrote in the sentence, - "The controller is a buck converter, it can only increase phase current above battery current if the voltage is less. Power input-to-output must remain constant." It sounds as if saying, - there is no free energy. It does for me.
So my previous calculation by multiplying those constant batt 72v volts by 250a phase is incorrect.
Power does not appear magically high without volts decrease, - "Power input-to-output must remain constant."

So what relation between phase amps increase and volts decrease. What volts we get out of 72v batt when throttle is fully open. Assuming controller needs to decreases volts to raise phase amps.

So if my controlller is rated at 100A battery amps than max power I can get is:
Constant power: 2x100A x 72v = 14.4kw
Phase power: 2xphase_Amps_X x Variable_voltage = 14.4kw (becasue "Power input-to-output must remain constant." )

Nevertheless why do people multiply battery voltage with phase amps expecting to get max setup power?
 
hubmonster
Sent: Thu Oct 19, 2017 6:08 pm
From: minde28383
To: steveo
Hello Steveo,
I got hubmonster but in 12inch rim. Only recently found out why it needs two controllers. It's two 3 phase motors inside; two sets of halls. My thread about it: viewtopic.php?f=30&t=91082

Last two controllers you use for this hubmonster are Adapto max? Correct?

thats correct!

I got one Sabvoton 72150. Don't know if it's worth to buy second the same or go cheeper way and buy two less powerfull controllers? Whad do you think.

despite some firmware issues i had with wheel size setting in the adaptto and the adaptto displaying the correct speed.. and affecting speed limiter ... watch my yt videos on my hubmonster bike... adapttos are good controllers.. sevcon controllers are ghe best.. but the avg consumer can not program them...

unless you want wheelie power on your scooter .. an adaptto midi x 2 should be plenty of power...and will still put your battery to the test...

i dont know which sabvoton you have... older retangle one or newer box style version.. i had one .. it was blue... with a thick aluminum heat sink on bottom (square one) . the newer gen is a solid controller... it has phase and battery current sensors built in this makes it difficult for it to self destruct. its a 24 fet sine wave with 4115 mosfets.. i tested mine with my x6.. solid controller .. it peaked around 264 phase under heavy burnout loads with my trike..

...


if your strapped for funds go sabvoton.. you should be happh with performance as well as money in you wallet..

if you want baller factor.. and are ok with firmware problems... glitches then adaptto

remember if you have dual adapttos thats 2 times the glitches to deal with when setting up each controller via the display

You could anwer here or directly to the topic or to your topic. Do you have topic about that? I can ask there for rest of us to see.

You mentioned that you plan:
motor cooling will be in the future.. and a front fork .. and wheel.. my bike really needs it!

i got my front fork and wheel on.. i just need to properly shim the fork as its rubbing the bottom of the head tube...


Bicycle fork is not enough for this hubmonster?

i went marzzochi shiver fork here
 
Steveo, thank for reply.

I'll add what I know too.

Regarding Sevcon.
Sevcon is not just very robust ant complicated to program and can run almost any motor if programmed by skilled person, but also needs programming cord which costs hundreds; that's why I sold this controller.

Regarding Sabvoton.
Both new sabvotons rectangular and square shape controllers have the same protection functions. Only funtional differences between them are following:

-Square controller is more suited to hidden places like box/compartment of the scooter due its shape; also it can attach additional cooling to its radiator, to original radiator plate under controller; water cooling is an option too. So basically you hide controller in box under seat and then deal with lack of moving air which would appear naturally when EV moves. But sometimes EV needs to be clean outside of hanging boxes thus hidden inside may be a must.
-Rectangular controller is more suited for ebikes due it's rectangular shape; it sheds heat better not just because it is mounted usually outside ebike but because it's differently built internally. Nevertheless it does not have option to attach additional radiator by default.

I think square Sabvoton controller is copied from Sevcon. When you look at it it reminds you Sevcon but when lid is open it you know where it came from.
 
minde28383 said:
Thinking over and over and it might be I comprehended something. I might understood what you wrote in the sentence, - "The controller is a buck converter, it can only increase phase current above battery current if the voltage is less. Power input-to-output must remain constant." It sounds as if saying, - there is no free energy. It does for me.
So my previous calculation by multiplying those constant batt 72v volts by 250a phase is incorrect.
Power does not appear magically high without volts decrease, - "Power input-to-output must remain constant."

So what relation between phase amps increase and volts decrease. What volts we get out of 72v batt when throttle is fully open. Assuming controller needs to decreases volts to raise phase amps.

So if my controlller is rated at 100A battery amps than max power I can get is:
Constant power: 2x100A x 72v = 14.4kw
Phase power: 2xphase_Amps_X x Variable_voltage = 14.4kw (becasue "Power input-to-output must remain constant." )

Nevertheless why do people multiply battery voltage with phase amps expecting to get max setup power?

Bill explained the operation of the controller better than I could have :)

Yes, it is wrong to multiply phase amps by battery voltage. If calculating power from phase amps you must also measure phase voltage, not battery voltage.

Remember that as motor speed increases, so does BEMF. Because of this, the current that can flow decreases (to near zero at motor no-load speed). At lower motor speeds motor impedance is so low that current must limited by restricting the voltage (via PWM duty).

A member here once gave an example of his powerful ebike controller that at low speeds could deliver ~600A phase amps while drawing only ~6A from the battery.
 
Buk___ said:
But ... you knew it was coming right? ... what about a step-up converter?
Good question - but it gets more complicated.

For a boost converter, you need an inductor, a switch and a diode again. But this time you connect one side of the inductor to the power source, and use the switch to short the other side to ground. Current starts flowing through the inductor. Inductors have the characteristic that they will allow current to change only slowly - they resist rapid changes in current. So that current slowly climbs - 0 amps, 1 amp, 2 amps.

Then you open the switch. Now the current has no place to go. Remember that inductors resist rapid changes in current. Without a diode, what happens is that the voltage rises so fast and so high that it exceeds the voltage rating of the switch. (This can get to thousands of volts.) The switch fails, and the current flows through the (now ruined) switch.

But let's say you give the current another place to go. For example, in an Otto cycle engine, they use a system like that to create a spark. When the switch opens, the current discharges through the spark plug, creating a spark with those thousands of volts. In a boost converter, they use a diode to steer that high voltage to the output. So if you have a 5V input, you can get 10V (or more) to the output through that diode.

In a motor controller such a current path isn't created; there's no point. But there is one exception. When the motor is spinning (i.e. a bike is coasting) the windings are generating a voltage, referred to as back-EMF. In other words, it's acting like a generator. Normally, with the controller off, the voltage isn't enough to make it back through the diodes, so the motor just freewheels. If you go faster and faster (say you go down a hill) then the EMF builds up. Once it's higher than battery voltage (i.e. you are spinning it faster than its base speed) then the voltage forward-biases the diodes and the system starts regenerating.

But let's say you are at speed much lower than that, say half of base speed. Now you turn on ALL the bottom-side FETs. This shorts out the windings. Windings are basically inductors, and these inductors have their own built-in voltage source (the EMF from the motor's speed.) So the current starts to build. Then you open all the switches. Now the current is looking for some place to go, and it sees it - through the top diodes in the controller and back to the battery. You have effectively created a boost converter. That's how most regen systems work, and why they can work at less than base speed.
 
minde28383 said:
So what relation between phase amps increase and volts decrease. What volts we get out of 72v batt when throttle is fully open.
72 volts. At wide open throttle there is no (or almost no) PWM effect; the battery is connected directly to the coils one phase at a time. So total phase amps is very close to battery amps, and phase voltage is very close to battery voltage.
 
All you posts were very supporting. During last days I was busy doing some wiring (installing contactor etc) for my scooter and will do few tests outside. At the moment I plan to ride it only with one controller (only one 3 phase motor), possibly register it and get plates to make it road legal, and if I like how scooter rides than I'll get second Sabvoton to unleash its (6 phase hub) whole potential.
 
minde28383 said:
MadRhino said:
From the pics, your controller is looking like it is real 6 phase, not 3 pairs.
You will need two 3 phase controllers to run it. One will run, but only part of its capability and possibly cogging in acceleration.

Tried quite a few times to run pair of three phase wires (one test with three wires, and second test with another three wires) to run by three phase controller and it runs smooth with wheel lifted in the air. It's like running with half motor and dragging dead weight.
Idea of acquiring some obsolete six phase controller is not appealing. I rather would rewind hub to be driven by regular three phase controller.

I think what you got there is worth keeping in original state. 6 phase motors are hard to come buy, and many have tried to find a source for them without luck. If you search for 6 phase here on the forum you will find the hubmonsters from John in CR. He have run serious power into his 6 phase motors. Running 2 x 3 phase controllers seems to split the load onto two controllers rather then one. Meaning you get serious amps into motor, while not pushing the controllers borderline meltdown.

It seems that people who have run multiple motors either in a dual/triple installation or like your 6 phase, real life is that no extra synchronization is needed. No complicated slave/master. Just hook up each controller like you would any 3 phase controller to any regular hub. They will work straight out the box is my guess.

If your motor is like the hubmonsters your motor is like you said earlier a two times 3 phase motor. In essence two motors in one carcass. Cherish what you got and take this motor to the max with dual controllers. Bonus is that if you ever melt down a controller you can ride home with a single controller.

Iirc using two controllers to run the 6 phase will give you a smoother running motor too.
 
Btw, is there anyone that has taken one of these 6 phase motor to a dyno? Or done any simulation of 6 phase vs 3 phase?
2 controllers most likely will dump more current into a 6 phase from standstill so I guess acceleration should be quicker. But then again, quicker then what? What motor would it make sense to compare these 6 phase motor against? A 205, dual width magic pies, 273? Or dual RC motors like what Matt builds?

Over the years there have been a few threads about 6 phase motors here, I think I've read them all. Yet I am still not sure if the benefits or perceived benefits of a 6 phase motor is possible to quantify. I've tried to google and yet I can't say I understand it all. I am still so fascinated by them that I think the only thing left is to get my hands on one of these and test it in real life and see what they can do.

John in CR has put effort into cooling. Another 6 phase enthusiast milled down the carcass to shave some weight off. For all the effort people put into these motors they must be special in some way? And if there are true and important advantages to run 6 phase over 3 phase why don't we see more manufacturers do 6 phase motors? Is because 3 phase is "good enough", or that the assembly and machining will be way more costly so much so that people would not pay extra for those benefits? Assuming there is benefits.
 
I think what makes it special is the ability to run lots of power on cheaper controllers, which is less important with the modern generation of controllers, and the high efficiency due to the thin laminations, and the low phase to phase resistance keeping the heat losses down. So not just its raw power output, but the efficiency at high power that was a selling point.
 
Voltron said:
I think what makes it special is the ability to run lots of power on cheaper controllers, which is less important with the modern generation of controllers, and the high efficiency due to the thin laminations, and the low phase to phase resistance keeping the heat losses down. So not just its raw power output, but the efficiency at high power that was a selling point.

Pretty excellent summary of the benefits right there. There's no inherent magic of running 6 phases, in some respects it's worse (needing two controllers, double up on wiring etc) but given the limitations of controller technology up until relatively recently this was the most achievable way of obtaining highway capable performance.

I've got a hub monster reconfigured for use as a mid drive. Geared 2:1 to the rear wheel it's capable of some pretty staggering performance with extreme efficiency. MUCH more efficient than a QS 273.

As an example situation, hitting full throttle from a roll of 50 km/h or 30mph, delivering enough phase current to result in 20kw of *output* at the tire, the hubmonster mid drive requires an input of 23,180 watt, efficiency of 86.4%. QS 273 40mm stator requires 29,420 watt, efficiency of only 68%. Under these circumstances they will both accelerate just as hard, but one places much more stress on the battery, requires a much more substantial controller (though only one) and most importantly would become a molten puddle very quickly. Because of the very low iron losses, you can spin a hub monster much, much faster than a conventional hub design - if you've got a small tire or the ability to reduce RPM this is a huge win.

Getting 400-500 phase amps on a single controller was up until recently quite expensive and very sizeable. In contrast two controllers capable of 250 phase amps are small, inexpensive and readily available.
 
Today was the day. Winds brought some warmer weather. I've tested my little machine with one Sabvoton 72150.

Battery: 72battery 30Ah
1 x controller:phase battery current - 50A, phase current - 120A
max speed reached little bit over 80kmh. Was afraid to wait for few more km/h. It was first run therefore don't know what to expect from these little scooters on turns.
Scooter weight about 93kg.

was checking constantly my new battery and all wiring for heat issues, found none. Was doing full throttle, brake, full throttle, brake... Will increase Ams and will do more testing next time. The one thing I managed to heat up was rear brake rotor. Wires, motor and phase wires were barely warm.

I could run it all the time with only one controller without knowing that it has double potential.

What are options for cheap controllers you might consider yourself if you had to buy two of them?
 
Finally removed cover see inside.
 

Attachments

  • 1.jpg
    1.jpg
    255.6 KB · Views: 1,829
  • 2.jpg
    2.jpg
    248 KB · Views: 1,829
  • 3.jpg
    3.jpg
    387.6 KB · Views: 1,829
  • 4.jpg
    4.jpg
    262.8 KB · Views: 1,828
  • 5.jpg
    5.jpg
    228 KB · Views: 1,828
Back
Top