What are the advantages and disadvantages of gear reduction?

[onloop]

If i want to minimize the stress/load on a DC brushless motor, Can this be achieved using gearing reduction?

SIMPLE EXAMPLE.

Say I have a fairly low 150 KV motor with gearing ratio of 2:1 (15T drive-30T driven)

VS

A motor say 300 Kv with a gearing ratio of 3:1 (13t drive -39t driven)

> Lets assume <
A. Both these setups have the same top speed which is suitable for the application.
B. Both have fairly similar torque output at the wheels. (based on the fact they can climb the same hill, in the same time, carrying the same load)
C. Both setups are at / or close to their maximum/optimal performance.
D. Both motors have the same construction / same bearings etc.

QUESTIONS:
1. Will the greater amount of gearing reduction mean that the 300kv motors have less load / stress put on them - therefore is it a better solution for longevity & minimising motor stress?.
2. Would there be higher spikes of amps pulled from the 2:1 setup as it is getting 50% of the load vs the 3:1 setup that only gets 33% of the load.
3. Does my logic in question 2 make any sense?
4. Is it perhaps a better idea to use physically larger motors and set them up to be "underperforming" if you want better longevity?


NOTE: the real world application is an electric skateboard with dual motors.


THANKS

 

[ARod1993]

The short answer is yes, you can reduce the load on a motor by gearing it down. Basically, the KV of a brushless motor is calculated as the reciprocal of the motor's torque constant, given in Nm per amp of current; a motor with KV 150 requires 150 amps through the windings (more or less) to put out a torque of 1 Nm at the shaft; gearing down the motor output (at the cost of speed) will magnify the motor torque at the wheel, and therefore producing the same torque at the wheel requires a lower current through the motor windings. You can use this knowledge to define the concept of "KV at the wheel," or the speed and torque of the driven wheel given a set voltage and current through the motor windings; generally, the system with the lower KV at the wheel will require less current to achieve a given amount of torque, but will also require a higher voltage to achieve the same top speed.

Given your examples, a 150 KV motor geared down 2:1 at the wheel will give your longboard 75KV at the wheel; that means that to produce a full Newton-meter of torque at the driven wheel will require you to put 75 amps through the motor.That torque might be overkill for a longboard; I have no idea because I'm more of a scooter and bike guy, so my systems tend to be between 10 and 40 KV at the wheel). By contrast, the 300KV motor geared down 3:1 has a rating of 100KV at the wheel, and so getting that same Newton-meter of torque will require pushing 100 amps through the motor (25% more current), but you'll be able to go 25 percent faster at the same voltage with this system than you will with the 150KV system. As far as your questions are concerned:

1) Most likely not; a 300KV motor is likely to be a lot smaller in construction and less powerful electrically than a 150KV motor. It might be seeing 50% less of the overall load than the 150KV motor, but the 150KV motor is most likely far better equipped to handle it (and will need half the current to push the same load as the 300KV motor). If you want to use a small, high KV motor you're going to have to gear it down pretty far to make this work; see the saga of eNanoHerpyBike for a treatment of this phenomenon; namely, if you gear down a high-KV motor by a lot but not enough you're going to have an unrealistically high KV at the wheel, giving you an unreachable top speed and fairly little usable launching torque.

In the case of eNanoHerpyBike, Charles used an 880KV motor with an 8.5:1 gear reduction, giving him 103.5KV at the wheel. Given that this was on decent-sized 8-inch pneumatic wheels, that gave him a top speed just north of 60mph, but no launching torque and a huge propensity to burn power heating up the windings; since he was using a "200A" airplane ESC under forced-air cooling with this, it lasted only a little while before the controller blew up. By contrast, the 500KV helicopter motor he put in after blowing the controller gave him about 60KV at the wheel, which meant a top speed of 30mph or so and a fairly scary amount of launch torque.

I think that covers most of my answers to (2) and (3) as well, so onto (4):

Yes. A big motor running at a high voltage through a high gear reduction is going to draw less current and, in general, be decidedly happier as long as you don't exceed the breakdown voltage of (or otherwise damage or melt) the wire insulation or something like that. You're better off taking a slightly lower top speed (or running a slightly higher voltage) on your vehicle and being able to haul ass without drawing a ton of current than you are doing the inverse.

 

[onloop]

Thanks for the detailed answer. It really helps a lot.

However lets assume that:
> voltage cannot be increased. both systems powered from 6S lipo.
> Both motors are identical size 50mm brushless DC. Just different KV.
> Dual ESC can handle 120amps
> But preferably the system should max at 60-70amps

The variables i can work with are:
> 50mm Motors custom made at any KV between 150 & 300
> Use any gear ratio within this range: min size of drive pulley is 13T. max of driven is 40T

So what do you think now?

 

[ARod1993]

Thanks for the detailed answer. It really helps a lot.

However lets assume that:
> voltage cannot be increased. both systems powered from 6S lipo.
> Both motors are identical size 50mm brushless DC. Just different KV.
> Dual ESC can handle 120amps
> But preferably the system should max at 60-70amps

The variables i can work with are:
> 50mm Motors custom made at any KV between 150 & 300
> Use any gear ratio within this range: min size of drive pulley is 13T. max of driven is 40T

So what do you think now?

That's going to depend a lot on your board wheel radius, desired top speed, and purpose for your board (is this for doing long runs on the flats? Is it for downhill longboarding? Do you want to be able to reliably climb hills with it?). On the one hand, if you're running 200mm pneumatic wheels on your board, you don't really have to worry so much about slip all that much; in that case going with 150-160KV and a 3:1 reduction will probably serve you fairly well. 6S LiPo would give you about 22.5V nominal, with about 24-25V fully charged, so your theoretical max speed becomes (150 rpm/V)(22.5V)(1/3)(2*pi*100mm/1 rev), which works out to about 26-27mph on the flats. Similarly, at 70A and 50-53KV at the wheel, this system would give you about 1.3 N*m of torque at the wheel, which translates to a fairly significant amount of torque and probably a halfway decent board for hill climbing with.

By contrast, if you run 75mm skateboard wheels on the board, then you're looking at a wheel with 35-40% of the radius of the setup above; that wheel will take the same amount of torque from the motor and convert it to somewhat more than twice the amount of linear force as the 200mm wheel will produce. Basically, at that point you want to increase your KV at the wheel by a factor of about 2.5 or so to stay in the sweet spot of the board (about 25-30mph cruising on the flats, with a half-decent but not stellar amount of torque available for hills), so running something at about 200KV and reducing it about 1.5 to 1 or 1.6 to 1 will give something like 125/133KV at the wheel, which should work well with skateboard wheels at that range. If you want a dedicated hill climber that does no more than 15 or so on the flats, then running about 180KV or so with a reduction around 2.6 or 2.7 to 1 on 75mm skateboard wheels will give you 65-70KV at the wheel and a lot of torque (assuming that doesn't deform the wheels or exceed their torque capacity).

 

[onloop]

Once again thank you for the detail. It is really helping me gain a deeper understanding on the topic.

Actually my goal is a good all round system that offers.
> max torque before causing problems due to loss of traction and without sacrificing top speed. Read; rapid controllable acceleration
> good hill climbing speed, 60-70% of top speed sounds fair
> efficient use of energy, through correct gearing & motor KV & not pulling more amps then necessary
> top speed on flat ground of 40-45km/h (24-28mph)


The current config that i am operating with is:
> 270kv 50mm motor. (dual motors - does this change the method of calculating torque at the wheel?)
> 83mm wheel diameter
> 2.4:1 (15-36t)
> 6s lipo
> recent top speed was recorded at 40km/h (approx 25mph) this speed is ideal.

this setup is great for me 95kg., but i believe it may be running near its limit, especially when the load is increased by a heavier rider / or load increase from very steep hill. I need the system to be super reliable even with a large rider say 110kg.

after doing long hill climb outer motor temp is probably reaching 60deg C.... its hot to touch.... but after riding at top speed temp is fine.

if you have the time you can view some recent testing that i have done comparing different KV motors. Actually it was the process of performing these test that has sparked my interest in DC motor performance theory.

THE BRUSHLESS OUTRUNNER MOTORS I TESTED ARE:

NTM PROPDRIVE 5060 270KV
VS
SK3 6354 245KV MOTOR

> heres the vid: https://www.youtube.com/watch?v=ioFqJPht2kM
> here is the data: https://docs.google.com/spreadsheets/d/1hnI3sXMS3NYDerEMLpPYJY9Se4gTGP6e4FjirWgXJUY/edit


so based on this additional information above can you offer anything further to your previous information....


FROM WHAT I CAN GATHER SO FAR BASED ON YOUR INFORMATION< going KV of around 200 then gearing say 15-24T (1.6to1) is a more efficient approach that will draw less current & put less stress on the motor VS. higher KV with higher ratio of reduction - which will want to pull more amps when under increased load.

 

[ARod1993]

This also depends on whether the current rating on that dual ESC is 120A peak and 60-70A continuous per channel or whether that's the totalallowable current across both channels. If you're able to pull 60-70A continuous per channel, then you're in very good shape, because on 83mm skate wheels, two motors pulling 70A each on matched systems that run about 125-ish KV at the wheel will let you pull a little over a Newton-meter at the wheel (twice what I estimated before, because you're actually putting twice the current into the overall motor system and will get twice the torque back out) while still giving you a top speed around 25-30mph. If your controller is only able to pull 60-70A continuous summed across both channels, then my old estimate still holds. Your motors will be pretty happy never seeing more than about 60A momentarily and 30-35A continuous, but your total torque will not be super amazing because you're only going to get about half a Newton-meter at the wheel (it'll just be a quarter Newton-meter at each of two wheels).

That said, on very small diameter wheels (i.e. skateboard wheels), assuming you don't exceed their torque transfer capability and skid or mash them, a fairly low torque will translate into a significantly higher linear force at the wheel; you'll still be able to pull up some reasonable hills without needing a ton of torque. If you're specifically concerned about running too much power through the motor, then that may not work so well because you're already running 112KV at the wheel (which is lower than the 125-133KV at the wheel than my recommended system provides, and therefore sucks less current into the motors). If you want to hold a top speed around 25mph without heating your motor up too much on hills (although I don't think that 65 deg C is necessarily worrying temperature yet), you'll need to pull the KV at the wheel down to about 90-100 and upgrade to 7s LiPo.

 

[spinningmagnets]

Is the system voltage is constant for both examples, the higher Kv motor with the greater reduction will have a higher magnet speed. That will result in the motor drawing fewer amps for the same acceleration. Amps cause the majority of the heat that can build up.

When you are starting from a stop, there will be a temporary peak amp-draw based on a balance between the load holding you back, being pulled by the volts and amps. After you are up to a fairly constant cruise speed, the volts stay the same, but the amps drop off to a lower level that is just adequate to maintain the load.

A larger motor will have more copper mass, which will help it absorb a greater peak amp draw...which will then cool off to just warm during the cruise phase...but of course you mentioned that the 50mm diameter was a fixed parameter for you.

 

[onloop]

the higher Kv motor with the greater reduction will have a higher magnet speed. That will result in the motor drawing fewer amps for the same acceleration. Amps cause the majority of the heat that can build up..

Hi Spinning Magnets, thanks for chiming in.
However now i am getting a bit more confused...

your saying that greater reduction may be beneficial if trying to minimize amp-draw which is responsible for the heat build up which is eventually what causes motors to melt the wire insulation & short out?

So this is my question...

Is it better (for component longevity & health & heat minimization & avoiding motor fires etc) to get your torque from higher KV motors with higher gearing reduction OR just utilizing the increased torque that comes from lower KV motors.

NOTE: My dual 120amp esc is two 120amp singles joined together, so max 120amp per channel. I want the constant to be kept around 60

 

[onloop]

It appears i have some conflicting theories... anyone else want to chime in?

 

[thepronghorn]

1. There is no "increased torque" from a lower kV motor. A lower kV motor only has a higher torque per amp, so it makes more torque for the same phase current than a motor with higher kV. However, provided copper fill is the same (which is a reasonable assumption), then the lower kV motor will have greater resistance in its windings which will balance out the lower current it needs to make a certain torque, giving it exactly the same torque rating as the higher kV motor.

2. In order to get the most power out of a motor (or to keep it the coolest for a certain power output), you want to spin the motor faster (with a higher kV or more volts) and gear it down accordingly. Thus for max power out of the motors in your specific application, you want to select the largest gear ratio you can (dictated by minimum and maximum pulley sizes, wheel size) then spin the motor at an appropriate rpm to get the top speed you want.

 

[onloop]

the lower kV motor will have greater resistance in its windings

Why/how? is it the size of the wire? or the amount of voltage?

 

[Miles]

the lower kV motor will have greater resistance in its windings

Why/how? is it the size of the wire?

Yes. If you halve the Kv (double the Kt), by doubling the number of turns in the winding, you not only halve the cross sectional area of the wire you also double its length. So, it doesn't matter how many turns you use, the copper losses for a given amount of torque stay the same...

 

[spinningmagnets]

There is a set amount of airspace in a given motors' stator to wrap wire around the stator-teeth. You can use many turns of thin wire, or a few turns of fatter wire (and of course some combination of wire thickness and turns in the middle of the possible range).

If you cut out the stock wire, and then re-wire it with a different diameter of wire, you have some control over the number of turns and the thickness of wire.

If the stock motor had thick wire, and you re-wind it with thinner wire, but use the same number of turns, there will be less copper mass. That has a lot of downsides, so nobody ever rewinds and uses less copper...fill the stator teeth with as much copper as will fit.

Fewer turns will have a high Kv, and high RPMs when using few volts. A lot of turns results in a low Kv, meaning low RPM's per volt that you apply...sometimes called a slow-wind motor.

This explanation is complicated a little by "winding in hand"..which means that: if you take four skinny wires and solder their ends together, then wrap the bunch around a stator-tooth, the result is that the bundle will act like a single fat wire. You might ask why someone would do that...it's my understanding that winding-in-hand is much easier on the fingers of the factory workers who have to wind motors all day, and also from a performance perspective: fat wires leave some airspace between all the wrapped wires, and lots of skinny wire can really pack as much copper as possible around the stator tooth.

You also asked about drawbacks: a direct drive motor is simple (and can be very quiet), slap it on your bike or stand-up scooter and it just runs.

Use a non-hub with a single stage of reduction and it can shred pants, chains/belts can break or wear out. Try to dump extra high power into a non-hub and the reduction becomes the weak link (a Cromotor can take 6,000W, but the Cro-bicycle pedals use bicycle chain because they only see about 200W of leg power).

If you want to use a smaller motor, you may need to raise its RPMs and use more reduction. You can raise the RPMs by using higher volts on the same motor, or swapping to a motor with a higher Kv while using the same volts.

If your motor is getting too hot to touch, you need to increase its RPMs, or use a larger motor (or both?).

Something that's not often mentioned is that...if your current system is drawing max amps from the battery, the battery may only last half as long as it might have otherwise if used on a system with lower overall amps draws, and lower peak amp-draws. High amp-draws make a battery hot, and battery heat sucks extra life out of it and makes it die earlier.

 

[onloop]

I feel like i need to summarize for clarity. Please correct me if i am wrong with my points.

1. For maximum power out of the motors (& longevity of motors?) One should utilize the highest KV motor combined with the maximum possible gearing reduction that allows you to reach your desired top speed.

2. Using max KV motors with MAX reduction is also the most efficient in terms of energy consumption, as there will be lower overall amps draws, and lower peak amp-draws.

3. Physically Bigger motors would be better because they should have larger area of stator allowing larger copper mass. i.e A Bigger motor with the same KV as smaller motor could use thicker wire & have less resistance which is better? less losses = more efficient with same volts?
> does less resistance actually make much difference in the real world, ie would it result in battery last longer?
> if so at what point are the returns diminished? bigger motors weigh more / cost more etc.

If someone could now explain the formula to determine what KV motor i should build for best efficiency & max power output based on:
> max speed 45km/h
> wheel diameter 83mm
> max available reduction (2.769 : 1)
NOTE: this system has two motors.. I guess that just equals more torque at the wheels.

 

[spinningmagnets]

I would start with the largest motors that would fit. The larger diameter will give the rotor more leverage, so you get better torque per watt that's applied. Also, larger motors typically come stock with lower Kv's. Then add the largest reduction that will fit (and is buyable without needing custom parts). Choose the highest voltage you can afford, and the highest voltage that will also fit on the power board.

Wheel diameter and Kv of the motor can then be determined, keeping in mind the top-speed that is desired.

During start-up from a stop, the motor will be drawing max amps (heat)....and after a few moments it will have stabilised at a lower amp-draw, which will hopefully give it some time to cool off to a lower steady-state temp. 93C / 200F is too high and motors/controllers will be living right on the edge of survival. 140F during normal operations indicates that the motor + reduction is properly sized for the load that is being applied. A larger motor can also absorb and shed a larger temporary heat-spike without getting so hot that there is a danger to it.

Regardless of the full page of calculations that you might be wrestling with, sometimes it is best to simply research previous builds and just copy a system that is working well in a user-profile that is similar to yours (hills? flat land? top speed?, etc).

If you are on a budget and don't have access to tools, it's rough to pursue your dream system. When faced with two design choices, I encourage you to select the choice that results in a higher magnet speed. When a motor bogs down and the magnets are spinning slowly at full throttle, it will go from bad to worse very fast.

 

[onloop]

I think I now have the knowledge to make better decisions now, thank you everyone...

HOWEVER I DO HAVE ONE LAST QUESTION>
if you review the data from a recent motor comparison test i did. see it here: https://docs.google.com/spreadsheets/d/1hnI3sXMS3NYDerEMLpPYJY9Se4gTGP6e4FjirWgXJUY/edit

the surprise i have is that the 245KV motor performed to about the same level as the higher KV motor with my body weight, however when loaded with extra 10kg the 270KV really struggles.

The theoretical top speed for the 270kv with 83mm wheels & 15-36t pulley should be 44.04km/h - when loaded with 10KG extra -However it does not reach this in fact it could only reach 38.6km/h

The theoretical top speed for the 245Kv with identical reductions as above should be 39.96km/h - when loaded with 10KG extra - it basically achieved top speed, it got to max 39.2km/h

SO MY UNEDUCATED CONCLUSION: this data tells me that the 270KV system is geared incorrectly (or on the absolute max for my body weight &), not higher enough reduction, so it therefore would be pulling higher constant amps, & much higher peak amps? and not running efficiently

DOES THIS SOUND LIKE A FAIR CONCLUSION? clarification of my thoughts will put my mind at ease!

THEREFORE IN THIS PARTICULAR APPLICATION LOWER KV IS ACTUALLY BETTER THAN HIGH KV MOTOR?

 

[fechter]

DOES THIS SOUND LIKE A FAIR CONCLUSION? clarification of my thoughts will put my mind at ease!

THEREFORE IN THIS PARTICULAR APPLICATION LOWER KV IS ACTUALLY BETTER THAN HIGH KV MOTOR?

It would appear that way from the observation.

For any setup, there is going to be a 'sweet spot' for efficiency. If the motor is spinning extremely fast but geared down to the right speed, there will be more RPM related losses in the motor (windage, bearing friction, hysteresis). Calculations can get you in the ballpark but actual testing is best. You want to try to keep the motor as close as possible to the peak in its efficiency curve for the most typical operating condition (usually flat ground, full speed). Most motors have their efficiency peak on the side of being lightly loaded.

Resistance losses in the controller, battery and wiring are a function of current squared, so anything that reduces the current helps there, but those losses are typically small compared to the motor itself. Controller and battery limitations frequently determine the maximum current you can run.

You also should consider losses in the gear reduction. Direct drive is the best in this respect. Generally speaking, the higher the reduction ratio, the greater the losses in the gearing system. Larger diameter gears or sprockets tend to be better than small ones.

A good way to find where the losses are is to see where the heat is being generated. Anything that's getting hot is wasting power.

Bottom line: try different setups and see which one works best. There are a lot of variables and an accurate simulation is hard to do.