What's so bad about low impedence?

nowholeself · Feb 28, 2020

So I sent ASI an email about using their controllers with the Neumotors 80xx series engines, adapted with hall sensors. They responded that the motor is "one of those low inductance/impedence" motors, and indicated that it would be difficult, if not impossible, to make it work with their controllers. What sort of controller should I be looking at? And what's so bad about low impedence?

Lebowski · Feb 28, 2020

Nothing really... you may need a bit higher PWM frequency than normal for improved efficiency but other than that a low inductance motor is no problem for a proper FOC controller with phase current control.... a cheap 6 stepping china controller however will have problems with low inductance due to inprecise control and the resulting high currents.

district9prawn · Feb 28, 2020

ASI are pretty conservative with their ratings. I ran their bac4000 controller with the 100kv wind of the 8057. Did the auto tune and set up as per the manual and everything worked.

spinningmagnets · Feb 28, 2020

The low impedance isnt bad by itself, but they typically have inductance issues.

Motors can absorb lots of inductance and voltage spikes, but controllers typically are sensitive to that.

nowholeself · Feb 28, 2020

district9prawn said:
ASI are pretty conservative with their ratings. I ran their bac4000 controller with the 100kv wind of the 8057. Did the auto tune and set up as per the manual and everything worked.

Your thread was actually one of two that encouraged me to try out the Neumotors engines. I was especially excited to see your use of the ASI with them, since I had already had such excellent results with the BAC800 on my BBSHD. Granted, the engine I ordered - the 8025 - is one of the smaller 80xx models. But the kv and inductance aren't wildly different. It's 150kv and 0.0177ohms. I'll probably just go for it.

Toorbough ULL-Zeveigh · Feb 28, 2020

well, zero ohms is ideal.
so naturally a controller/inverter manufacturer is gonna say that, it's hard on the equipment.
it makes his job more difficult to design as well as for more warranty claims having to drive/transfer power into lo-ohms.

why do you think speaker impedance keeps getting lower.
when i started, typical speakers were 64 ohms to match the amp's output impedance of the tubes.
some car radios had 32 ohm with a matching transformer; ff to today there are some hi-end 2 ohm systems.

the typical design impedance of all things inductive at any given point in history is limited by the state-of-the-art in the electronics of the day.
but why should you believe me, i'm just taking shit you say?

liveforphysics said:
If we all had controllers with magic fets that had 10uOhm RdsOn, and no ESR caps, we would all be running 1 turn motors.

too bad liverphysics don't hang around here no more i would tell him (since he's seems to be unaware) about the magic Ge based P-channel FETs.

(aww, he probably knows by now. he noes everything)

JackFlorey · Feb 28, 2020

Toorbough ULL-Zeveigh said:
well, zero ohms is ideal.

Right. But they are talking about impedance, not resistance. Two different things.

why do you think speaker impedance keeps getting lower.

Because people used to drive speakers from radios plugged into the wall (320 volts peak.) Now they are running them from lithium ion powered devices (as low as 2.5 volts) and cars (as low as 9 volts.) You need the lower impedances to get adequate volume at those voltages.

the typical design impedance of all things inductive at any given point in history is limited by the state-of-the-art in the electronics of the day.

It's limited by their power supplies. P=V^2/R.

Toorbough ULL-Zeveigh · Mar 2, 2020

JackFlorey said:
But they are talking about impedance, not resistance. Two different things.

are they really?
or is resistance along with reactance the two sides of the same impedance coin?

wikipedia said:
When a circuit is driven with direct current (DC), there is no distinction between impedance and resistance; the latter can be thought of as impedance with zero phase angle.

car radios in the 50's into the 60's with 32 & 16 ohm speakers ran on 12V same as today so that can't be right.
what you failed to address, why then does Lukeforphysics have a stiffie for a one turn minimum reactance/impedance motor?
why wouldn't you want a more efficient motor with no impedance of any kind, resistance or otherwise?

It's limited by their power supplies.

which in turn is limited by the semiconductors available.
i'd really like to know what sort of semi's would you use to drive a zero ohm superconductive motor.

flat tire · Mar 2, 2020

Short answer, the current and speed control algorithms on ASI are not designed for motors with extremely fast current ramp times (low inductance). You can get some degree of functionality, but there will be issues.

Ask anyone trying to tune a CYC X1 for rapid response on ASI. Yeah, right. The tunes I've seen have relatively slow throttle ramps on the order of hundreds of ms, including the ERT (pos) tune. If you want, say, a 10ms throttle ramp time the current and speed regulator will have issues particularly in a non direct drive application.

Toorbough ULL-Zeveigh · Mar 2, 2020

flat tire said:
Short answer, the current and speed control algorithms on ASI are not designed for motors with extremely fast current ramp times (low inductance).

even shorter answer: slew rate
i think that's why LFP favours lower voltage/low turn count

JackFlorey · Mar 2, 2020

Toorbough ULL-Zeveigh said:
or is resistance along with reactance the two sides of the same impedance coin?

Resistance and reactance are two very different things. Reactance is the complex part of impedance. Resistance is the real part of impedance, usually at zero frequency.

wikipedia said:
When a circuit is driven with direct current (DC), there is no distinction between impedance and resistance; the latter can be thought of as impedance with zero phase angle.

Yes, at zero frequency, then the real part of impedance equals the DC resistance, and there is no complex part (you need frequency to have a complex impedance.)

car radios in the 50's into the 60's with 32 & 16 ohm speakers ran on 12V same as today so that can't be right.

In the 50's and much of the 60's radios were made from vacuum tubes. Vacuum tubes have to run at high voltages (around 200 volts.) In homes this was easy since you can get 320 volts from an AC outlet with two diodes and a capacitor. In cars it was a lot harder.

So they used vibrator tubes to create an AC waveform, then a transformer to boost the 12V to the voltage that tube radios needed (at LEAST 50V, more often 200 volts.) At 50 volts there is no problem running 32 and 16 ohm speakers.

Then transistors came along, and the radio could operate at 12 volts. And for 12 volts you need 8 and 4 ohm speakers to get enough volume out. (Or you need to do the same thing that the vibrator tubes did and boost the voltage much higher, which some high end car audio systems do.)

what you failed to address, why then does Lukeforphysics have a stiffie for a one turn minimum reactance/impedance motor?

Because you can get higher speeds, and often you can get greater overall system efficiency (and a much smaller motor) if you can handle the RPM.

why wouldn't you want a more efficient motor with no impedance of any kind, resistance or otherwise?

You can't have a motor with no impedance of any kind. You'd violate the laws of thermodynamics.

i'd really like to know what sort of semi's would you use to drive a zero ohm superconductive motor.

Same ones we use now. Again, zero ohms DC is not the same as zero impedance.

Toorbough ULL-Zeveigh · Mar 2, 2020

JackFlorey said:
Then transistors came along, ...

what i said in a broader sense to include more than transistors.

You can't have a motor with no impedance of any kind.

no but you can (theoretically) approach it.

You'd violate the laws of thermodynamics.

which is why i asked ...

i'd really like to know what sort of semi's would you use to drive a zero ohm superconductive motor

how is it possible to extract any real work out of this motor?

JackFlorey · Mar 2, 2020

Toorbough ULL-Zeveigh said:
what i said in a broader sense to include more than transistors.

It all goes back to supply voltage. The power you can get out of a load is given by its resistance (at DC) or its impedance (at AC.)

At 32 ohms you can get 4.5 watts out of a 12 volt supply. At 4 ohms you can get 36 watts. So often car audio goes with the lower resistance.

Want more power? You can go to 2 ohms, although now you are going to need seriously heavy speaker wire to carry the currents. Or you could boost the voltage with a DC/DC.

You can't have a motor with no impedance of any kind.

Click to expand...

no but you can (theoretically) approach it.

Not really. An absolutely ideal motor will present an impedance equal to its mechanical power output. Outputting 100 watts? Then if you are driving the motor at 36 volts, you will see an effective impedance of 13 ohms. Even if it uses superconducting wire, frictionless bearings etc etc.

At no load you would actually see close to infinite impedance in an ideal motor. (Assuming one that behaves like a DC PM motor, that is.) But power in would approach zero.

how is it possible to extract any real work out of this motor?

That's not a motor. It's a picture of the Meissner Effect, which allows magnetic levitation over a superconductor.

thepronghorn · Mar 2, 2020

JackFlorey said:
Toorbough ULL-Zeveigh said:

what i said in a broader sense to include more than transistors.

Click to expand...

It all goes back to supply voltage. The power you can get out of a load is given by its resistance (at DC) or its impedance (at AC.)

At 32 ohms you can get 4.5 watts out of a 12 volt supply. At 4 ohms you can get 36 watts. So often car audio goes with the lower resistance.

Want more power? You can go to 2 ohms, although now you are going to need seriously heavy speaker wire to carry the currents. Or you could boost the voltage with a DC/DC.

You can't have a motor with no impedance of any kind.

Click to expand...

no but you can (theoretically) approach it.

Click to expand...

Not really. An absolutely ideal motor will present an impedance equal to its mechanical power output. Outputting 100 watts? Then if you are driving the motor at 36 volts, you will see an effective impedance of 13 ohms. Even if it uses superconducting wire, frictionless bearings etc etc.

At no load you would actually see close to infinite impedance in an ideal motor. (Assuming one that behaves like a DC PM motor, that is.) But power in would approach zero.

how is it possible to extract any real work out of this motor?

Click to expand...

That's not a motor. It's a picture of the Meissner Effect, which allows magnetic levitation over a superconductor.

That's only an ideal motor with regard to losses. The ideal motor that lfp was talking about is with regards to internal resistance and operating voltage. His ideal motor wouldn't be driven at 36V, since its impedance would be so low that 100W output only takes 0.1V in or something like that. Continuing with this same example, 3000W only takes 3V, so you can run a 1S battery with a very simple BMS and no HV considerations and get to 50mph easily.

Toorbough ULL-Zeveigh · Mar 3, 2020

JackFlorey said:
It all goes back to supply voltage. The power you can get out of a load is given by its resistance (at DC) or its impedance (at AC.)

well there's the flaw in your thinking.
generally power goes into a load.
hard to believe you actually went to MIT.

interestingly lower supply voltage also comes back round to faster slew.

At 32 ohms you can get 4.5 watts out of a 12 volt supply. At 4 ohms you can get 36 watts. So often car audio goes with the lower resistance.

Want more power? You can go to 2 ohms, although now you are going to need seriously heavy speaker wire to carry the currents. Or you could boost the voltage with a DC/DC.

no, the load ohms is dictated by impedance matching & Maximum Power Transfer Theorem

You can't have a motor with no impedance of any kind.
no but you can (theoretically) approach it.

Not really.

Click to expand...

every improvement in motor efficiency is 'approaching' ideal.
is called asymptotic.

That's not a motor.

it behaves like a motor.
and while i'm curious why it doesn't qualify as one,
rather than continue straying further from the subject,
how about actually answering the OP's question?
it's okay if you don't know.

Lebowski · Mar 3, 2020

JackFlorey said:
Not really. An absolutely ideal motor will present an impedance equal to its mechanical power output. Outputting 100 watts? Then if you are driving the motor at 36 volts, you will see an effective impedance of 13 ohms. Even if it uses superconducting wire, frictionless bearings etc etc.

Just my $0.02 ... the whole impedance story is wrong. The motor generates a voltage (the backemf) in which you are putting a current -> the amoujt of mechanical power you get is voltage times current. I dont see why this would not work with a superconductor.

JackFlorey · Mar 3, 2020

Toorbough ULL-Zeveigh said:
JackFlorey said:

It all goes back to supply voltage. The power you can get out of a load is given by its resistance (at DC) or its impedance (at AC.)

Click to expand...

well there's the flaw in your thinking.
generally power goes into a load.

Generally, with motors, you want to get power OUT of the load (the motor.) If power goes into a motor, but nothing comes out, it would be pretty useless, eh?

no, the load ohms is dictated by impedance matching . . .Maximum Power Transfer Theorem

Not really. If you have zero output impedance (which batteries come close to) you don't really want zero input impedance (infinite current.) You want a load that matches the max current (and voltage ranges) of your battery.

Likewise, if you have close to zero output impedance from an amplifier, you don't want close to zero impedance speakers; currents would be very high but you wouldn't get much power out of the speaker. (since power = current x voltage.)

If you are limited by output impedance, then you might well want to match the load impedance to that. That's more applicable to HF electronics (like antennas) than motors.

how about actually answering the OP's question?
it's okay if you don't know.

Sure. Bad things about low impedance:

High (potentially damaging) currents

If inductance is low, you might hit damaging currents before the switch can open the circuit and interrupt the current

If it's an FM inverter, you might end up running at a very high and inefficient frequency with a low inductance motor. (Note - low inductance = low impedance at higher frequencies.)

If it's a fixed frequency inverter, you might end up running at such a low duty cycle that again you'd see low inverter efficiency.

JackFlorey · Mar 3, 2020

Lebowski said:
Just my $0.02 ... the whole impedance story is wrong. The motor generates a voltage (the backemf) in which you are putting a current -> the amoujt of mechanical power you get is voltage times current. I dont see why this would not work with a superconductor.

Agreed. Back-EMF becomes an issue once the motor is spinning at some fraction of its base speed. Impedance is still an issue, but you are seeing a higher effective impedance, since you see less current flow for a given voltage across a motor winding due to the lower delta V. Eventually, of course, the current becomes zero when you are at base speed, and the effective impedance becomes infinite.

JackFlorey · Mar 3, 2020

thepronghorn said:
That's only an ideal motor with regard to losses. The ideal motor that lfp was talking about is with regards to internal resistance and operating voltage. His ideal motor wouldn't be driven at 36V, since its impedance would be so low that 100W output only takes 0.1V in or something like that.

Even an ideal motor has a motor constant.

I would also note that a motor with a motor constant such that it took 100W input with a voltage of .1 volt across the motor (at some given speed) would be difficult to design for; consider what the controller/wiring/windings would look like if they had to carry 1000 amps.

Toorbough ULL-Zeveigh · Mar 3, 2020

JackFlorey said:
Generally, with motors, you want to get power OUT of the load (the motor.) If power goes into a motor, but nothing comes out, it would be pretty useless, eh?

power in, torque & rpm out is what you want.
nice try cya tho.

no, the load ohms is dictated by impedance matching . . .Maximum Power Transfer Theorem

Click to expand...

Not really. If you have zero output impedance (which batteries come close to) you don't really want zero input impedance (infinite current.) You want a load that matches the max current (and voltage ranges) of your battery.

Likewise, if you have close to zero output impedance from an amplifier, you don't want close to zero impedance speakers; currents would be very high but you wouldn't get much power out of the speaker.

the exact opposite what the theorem stipulates, when in fact you get max power.

If you are limited by output impedance, then you might well want to match the load impedance to that. That's more applicable to HF electronics (like antennas) than motors.

the theorem applies to any form of power transmission.
mechanical, sound, rf, optical, x-ray; makes no difference.

JackFlorey · Mar 3, 2020

Toorbough ULL-Zeveigh said:
power in, torque & rpm out is what you want.

I want power out. YMMV.

the exact opposite what the theorem stipulates, when in fact you get max power.

Yes, the theorem states that. The real world doesn't work like that.

Here's a physics problem for you. A speaker has zero impedance at all frequencies. You feed it 100 amps. What is the power out of the speaker? (Assume any efficiency you like.)

mechanical, sound, rf, optical, x-ray; makes no difference.

It does indeed. If you have a specific output impedance, AND you are not limited by voltage or current, then matching the power in the output impedance and the load gets you maximum power.

However if you tried to do that with a 36V 15AH lithium ion battery (output impedance of .02 ohms or so) you would get 900 amps out. For perhaps a few milliseconds, before either the protection circuit kicked in, a fuse blew or your battery melted down. So you would specifically NOT match the output impedance of the battery. You would instead choose an impedance range that went as low as 1 ohm or so. That would give you a maximum current of 36 amps, which the battery could likely handle. And you'd get 1300 watts, which is a good amount of power for acceleration.

Simplified explanation:
Matching impedances in ebikes = BAD. Choosing impedance to match capabilities of battery = GOOD.

flat tire · Mar 3, 2020

JackFlorey said:
Toorbough ULL-Zeveigh said:

power in, torque & rpm out is what you want.

Click to expand...

I want power out. YMMV.

You're in luck, you guys inadvertently agreed since torque * rpm = .....wait for it.....POWER.

Marlboro man seems to have added some freebase to that sweet virginia tobacco.

JackFlorey · Mar 3, 2020

flat tire said:
You're in luck, you guys inadvertently agreed since torque * rpm = .....wait for it.....POWER.

Yep. And motors just convert it from one form to another (and, often, back.)

Lebowski · Mar 3, 2020

JackFlorey said:
Lebowski said:

Just my $0.02 ... the whole impedance story is wrong. The motor generates a voltage (the backemf) in which you are putting a current -> the amoujt of mechanical power you get is voltage times current. I dont see why this would not work with a superconductor.

Click to expand...

Agreed. Back-EMF becomes an issue once the motor is spinning at some fraction of its base speed. Impedance is still an issue, but you are seeing a higher effective impedance, since you see less current flow for a given voltage across a motor winding due to the lower delta V. Eventually, of course, the current becomes zero when you are at base speed, and the effective impedance becomes infinite.

You are confusing V/I and dV/dI ... impedance is the latter.

madin88 · Mar 3, 2020

@ op, did they tell you a number for inductance?

At which voltage and RPM you would like to use it?

One idea, which probably would make it running, would be to split the delta winding and connect it in star. By doing so you would get about 3 times higher inductance.
The loss of RPM (1,7x less) you could get back with field weakening to some point.

What's so bad about low impedence?

1 W

10 MW

1 kW

Moderator

1 W

1 MW

100 kW

1 MW

100 kW

1 MW

100 kW

1 MW

100 kW

1 kW

1 MW

10 MW

100 kW

100 kW

100 kW

1 MW

100 kW

100 kW

100 kW

10 MW

10 MW

Similar threads