# Torque of motor vs operation as generator

While trying to work out which motor to use for a wind turbine project, I'm coming across torque numbers for DC motors. I'm mainly looking at a BLDC motor right now.

Clearly, for a smallish turbine that I'm aiming for, I'm looking for low startup torque.

So, is there any relation between motor torque and/or wattage with its generator torque? Or more in general, how does one decide which motor is best for a given set of blades?

• Does this answer your question? Are permanent-magnet motors equally (in)efficient as motors vs. as generators? Dec 24 '19 at 23:16
• Define 'smallish'. Dec 24 '19 at 23:47
• I think not. I presume the torque produced as a motor varies on current, but I'd like to know if I can draw any conclusions from torque or power to determine the torque needed to get the shaft spinning. Dec 24 '19 at 23:48
• Smallish as a little project where I have turbine blades around 50cm or less. Dec 24 '19 at 23:50
• Yes and No. You need an MPPT impedance matching controller for Wind + Generator + battery Look here for motor curves BLDC project Dec 25 '19 at 0:35

So, is there any relation between motor torque and/or wattage with its generator torque?

Generally yes. To a pretty good first approximation, the torque that a DC motor gives for a given current is equal to the torque that same machine will need to absorb to deliver the same current as a generator. The biggest difference will be the machine's friction torque, which is always a loss either way.

Clearly, for a smallish turbine that I'm aiming for, I'm looking for low startup torque.

For an ideal DC machine, as long as there is no current, there is no torque. So in that sense "startup torque" is entirely under your control, or at least under the control of the electronics you use.

Typically what's not ideal about a DC machine is the friction torque and cogging torque. These are often not well specified in datasheets (and aren't specified at all if you're buying a motor for hobbyist or consumer use, such as an RC plane motor).

Or more in general, how does one decide which motor is best for a given set of blades?

You find the speed and torque at which the blades are most efficient, and you find a machine that will absorb that torque at that speed -- or you choose to drive the motor through a set of gears (probably step-up) to spin the motor at a different speed, while absorbing the same power.

Then you figure out the speeds and whatnot for when the wind conditions aren't ideal (most especially you check what happens if the turbine goes overspeed), and you deal with those.

• Cripes. Once again, very well covered. I was recalling my years, decades ago when considering designing an ultralight, studying the old NACA (pre-NASA) airfoil books. These provided thousands -- and I mean many thousands -- of plans that showed cross-section and twist for various propeller blades and loads of data on each as well as wing plan forms and lift and drag tables. And before I could sort this in my mind to consider what to write.... Well, I think I'll just lay back now and rest.
– jonk
Dec 25 '19 at 1:02

The minimum torque required to get the shaft turning is essentially the bearing friction. With a simple brushless DC motor, reluctance torque must also be considered. Reluctance torque is the torque that tends to hold the rotor in a specific position when there is no electrical load connected to the generator.

To start the generator with an electrical load connected, you need to determine at what voltage the voltage converter connected to the output begins to operate and supply power to the load.

For minimum speed operation, you must to determine the losses that must be supplied or the point at which the input mechanical power exceeds the losses. That may be pretty difficult to do analytically. It may be easier to test the motor and electronic controller.

• This is the first time I've seen it called "reluctance torque" -- you'll see it called "cogging torque" in datasheets, if the manufacturer will fess up to it at all. Dec 25 '19 at 5:02
• @TimWescott A Google search of "reluctance torque" is quotes brings up lots of usage. Dec 25 '19 at 14:52
• I knew what you meant -- it's just the first time I've seen it, and I've worked around motors all my professional life. Maybe it's a generator thing? At any rate, it's clearly a synonym for cogging torque, and that's what you see in Pittmon and other motor manufacturer's data sheets. Dec 25 '19 at 20:21
• I think it may be a market segment thing. I've been working with motors all of my professional life and never heard of Pittman before. Also mostly worked with 3 phase, 480 volt motors rated 5 Hp and larger. I think there may be a distinction made between cogging torque and reluctance torque in some literature, but I can't explain why I have that idea. Dec 25 '19 at 23:55
• Oh, that's it -- you work with great big dirty things and I work with itty bitty clean things. Pittman makes permanent magnet DC motors; most of their line would be teeny next to the stuff you work with. Dec 26 '19 at 4:32

Motor fundamentals:

• torque relates to current
• angular speed relates to voltage
• mechanical power is torque times angular speed
• electrical power is current times voltage

It works both ways for motor and generator mode.

The more blades your turbine has, the faster it will rotate for a given wind speed.

The power profile of a Wind Turbine vs wind speed does not match the power profile of a DC generator in RED below and therefore if it were loaded, would stall the turbine.

The generator will have some frictional load but nothing else unless current=torque is drawn. Maximum Power Transfer Theorem applies to both mechanics and electrical systems which states that the Impedances must be matched at all times to transfer maximum power.

For PV solar cells which act as current sources with an open cct voltage a, Voc and a short circuit current Isc, that impedance is Voc/Isc=Zpv.

For a battery the maximum power is not safe for more than xx seconds at a time as this creates too much outgassing and temp. rise. For a car battery the CCA rating is 30 seconds like in the coldest weather with a hardened oil friction engine turn an engine slowly up to the CCA current rating where the voltage drops from 12 to 7V during rotation at the CCA rating sustained for only 30 seconds.

For a wind generator, it has no torque to drive a generator until a threshold RPM then that has a torque and power curve vs RPM that is more quadratic.

The power drain from a battery charger depends on CC or CV so it is more flat power during CC mode or slightly rising then declining in CV mode as current declines.

So a Maximum Power Point Tracking (MPPT) is need that hunts or has a smart algorithm to regulate the load to match the product of the Turbine * DC Generator * Battery Charger. I suspect this looks like a 0 curve up to x RPM then a steep linear curve up to the RPM that matches the desired wind speed to max power of the generator. Inertial or friction speed brakes are need to govern over-speed or using the generator into a short circuit if not power is needed.

This is a custom Engineering challenge to define each of these interface specifications to match the Impedance of the load to satisfy both the Turbine and the batteries. The result is specifications that define each interface.

## edit

It sounds more complicated than a basic solution.

If the DC voltage out is matched to the battery charger at some adequate RPM then the charger will regulate the turbine RPM to that speed. It may not be optimal power but at least it will start no load until it reaches the charger Vmin enable threshold.

Then a means to enable higher RPM with wind speed is needed.

If the generator has lower impedance than the load, the RPM will increase. If the SMPS charger has lower impedance then the RPM will decrease.

p.s.

My 1st design was a VLF Doppler GPS used in Arctic Ocean floating-automated weather station in 1976 with a 750W vertical axis Wind Turbine that operated at 5 MPH start speed.