# Decreasing Motor Torque Requirement With a Gearbox

• crooper
In summary, the goal of this conversation is to find a way to rotate a 500kg(1100lbs) load to 90° in 1.5 seconds. The inertia of the load makes it hard to accelerate and decelerate, so a gear system was designed but the motor still needs too much torque. There is a possibility to decrease the motor torque need with a different gear system or different acceleration, or a special servo motor.

#### crooper

hi,
I'm working on a rotational system, and my goal is to rotate a 500kg(1100lbs) load to 90° in 1.5 seconds. Inertia of load makes system hard to accelerate and decelerate. I designed a gear system but still motor needs too much torque.

Is there any possible method to decrease motor torque need? A different gear system or different acceleration? or a special servo motor?

The minimum torque motion profile is to accelerate at a constant rate for 45 degrees, then decelerate at a constant rate for the remaining 45 degrees. That motion profile is also described as "triangular velocity". That is the absolute minimum torque to make your 90 degree move in the prescribed time.

The peak velocity is at the 45 degree point. The best gear ratio will put the motor right at maximum RPM when at peak velocity. Motors are rated for RMS (continuous) torque, and for peak torque. If there is a rest period between moves, and the motor is anything other than a stepper motor, you can run the motor at its peak torque.

If your move is using the motor peak torque, and reaching the motor peak RPM, you have the correct gear ratio and the smallest motor that will do the job.

If you need more power, take a look at what is available for servo motors from one manufacturer: https://literature.rockwellautomation.com/idc/groups/literature/documents/sg/knx-sg001_-en-p.pdf. Take a look at their HPK series motors if you want to move your load really fast.

Note that the linked document is only their selection guide. It does not tell you everything needed to fully specify a motor and drive. That's why the torque and RPM are normally calculated by a mechanical engineer, and the motor and drive and programming are done by an electrical engineer. Small little loads at low speeds can be moved by an Ebay stepper motor. Your load is far beyond that. You need a real servomotor.

Asymptotic, crooper, berkeman and 1 other person
crooper said:
and my goal is to rotate a 500kg(1100lbs) load to 90° in 1.5 seconds.
Yoiks! That's a fast rotation of a pretty heavy object. Can you say what the application is?

Also, how often do you need to do this? Do you need to be able to rotate it back? If the duty cycle is low, you may be able to use a stored energy system that has a lower power input requirement spread out over the machine's cycle time...

crooper
Does it have to be a motor?

I am thinking of a pendulum like contraption that uses gravity to accelerate and decelerate.

I could also visualize pneumatic pistons, one to accelerate and an opposing one to decelerate.

Does it have to decelerate to zero, or can it bang into a stop?

Must the 90 degree rotation be precise?

As @berkeman said, is this a one-time motion or repetitive? If repetitive, does the return stroke need the same 1.5s profile?

crooper
Be very careful about using actuators for moving heavy masses. I know of a case (I was not personally involved) where the engineer need to rotate a mass of about 2000 lbs 180 degrees. He specified a hydraulic rotary actuator with an internal cushion at the end of travel. He properly calculated the torque to accelerate and rotate, but did not think to calculate the deceleration torque. The machine got to the customer, at which point the short deceleration distance caused deceleration torque high enough to shear off the bolt circle attaching the mass to the actuator. A full ton of machinery fell to the floor. I don't know what happened after that. I do know that particular machine failed in the marketplace.

Servomotors are good for moving heavy masses. I helped in the design of a machine that moved a 900 lb mass vertically about 3 feet at accelerations of 500 in/sec^2 (12.7 m/sec^2). Several times per minute, 24 hours per day, for years on end. That's a 36 inch move in 0.5 seconds. The motor was one of the larger Allen-Bradley MPL series motors. It is necessary to carefully calculate the regenerated energy from the deceleration, and specify a motor and drive combination that will properly handle it.

JBA, Asymptotic, crooper and 2 others
Thanks for replies.

My goal is same as in here ; see in the videos first second: Car is rotating shelf in less then two seconds. I was wonderig how they overcome inertia of the mass.

I need to find out a servo(bldc, stepper or other) motor or gear design. At least how these cars rotating shelfs so fast. By the way, it's not a school project, will be industrial project. After finding solution, I'm going to calculate budget.

Last edited:
berkeman
Are you turning the load? How does the load get on and off the turntable? Are you moving the load/turntable assembly? Are there other things happening at the same time? Are people involved, or is this fully automated? We need more information to help you.

Turning a heavy load is one thing. Integrating it into a larger system is far more than finding a motor, and certainly far beyond something cobbed together from Ebay components.

crooper
ok, video is an example but logic is same. Think like that : A worker puts the shelf to platform with forklift. platform/turntable has to rotate the shelf load with a gear mechanism and a motor. Look at the video's first second: Car is stopped, fixed on ground, 300kg mass is rotating fast about 2 sec to turn 360 deg. What kind of mechanism can be? can Motor handle inertia of mass with simple gear?
When I do some math, My motor shoud have torque of 300Nm which is too much.

Also I'm trying do design insdustrial thing, can be machined parts, special parts as well. Goal is not to collect e bay components and assembly. But ofc, budget is important.

Your limited description does not give us enough information to help you further. Go back and read EVERY.SINGLE.WORD of the previous posts. Pay special attention to gear ratios. Then list exactly what your system needs to do. There is a huge cost difference between putting a motor on an existing turntable, and a system to load, turn, move, and unload the turntable. When I say huge, I mean more than a factor of 10. Like several times more than a factor of ten.

A budget price requires that you have costs for major components, such as servomotors and gear reducers. Here are a few gear reducer manufacturers to get you started: Boston Gear, Stober, Alpha, Falk. Note that Stober reducers are completely different from Falk, so you cannot just look at one manufacturer. If you plan to make the turntable, and support the turntable with a turntable bearing, you had better check the price and delivery of the bearing first. And the cost of machining the large parts. After carefully calculating overturning moments on the bearing, and bearing life for your selected bearing.

A budget price requires a concept design. That consists of a sketch (a pencil sketch is good enough) that shows the major components, and a written description of loads, speeds, duty cycle. What major components will be purchased? What will be fabricated?

Will it turn once per hour, per minute, or every 4 seconds? How many loads per year? How many years does it need to last? How durable does it need to be? If loaded/unloaded by forklift, will it be designed to survive a forklift running into it? Are there people in the area? If so, how will you make it meet safety codes? How automated will it be? Will it be integrated into an existing control system?

I have over 20 years experience designing exactly this type of machine, and a minimal budget estimate would take me 20 to 100 hours depending on the scope of the project. It will take you several times longer.

Or you can find a machine building / system integrator / automation company that specializes in this type of work, and ask them for a quote. If you do that, you will need a detailed system specification so that they know exactly what to bid on.

crooper, Asymptotic, anorlunda and 2 others
jrmichler said:
Your limited description does not give us enough information to help you further. Go back and read EVERY.SINGLE.WORD of the previous posts. Pay special attention to gear ratios. Then list exactly what your system needs to do. There is a huge cost difference between putting a motor on an existing turntable, and a system to load, turn, move, and unload the turntable. When I say huge, I mean more than a factor of 10. Like several times more than a factor of ten.

A budget price requires that you have costs for major components, such as servomotors and gear reducers. Here are a few gear reducer manufacturers to get you started: Boston Gear, Stober, Alpha, Falk. Note that Stober reducers are completely different from Falk, so you cannot just look at one manufacturer. If you plan to make the turntable, and support the turntable with a turntable bearing, you had better check the price and delivery of the bearing first. And the cost of machining the large parts. After carefully calculating overturning moments on the bearing, and bearing life for your selected bearing.

A budget price requires a concept design. That consists of a sketch (a pencil sketch is good enough) that shows the major components, and a written description of loads, speeds, duty cycle. What major components will be purchased? What will be fabricated?

Will it turn once per hour, per minute, or every 4 seconds? How many loads per year? How many years does it need to last? How durable does it need to be? If loaded/unloaded by forklift, will it be designed to survive a forklift running into it? Are there people in the area? If so, how will you make it meet safety codes? How automated will it be? Will it be integrated into an existing control system?

I have over 20 years experience designing exactly this type of machine, and a minimal budget estimate would take me 20 to 100 hours depending on the scope of the project. It will take you several times longer.

Or you can find a machine building / system integrator / automation company that specializes in this type of work, and ask them for a quote. If you do that, you will need a detailed system specification so that they know exactly what to bid on.
Ok, you are looking at the big picture. Let's make it easy. I am designin turntable of this agv as a subsystem. I need to know ways to reach the goal: turn the table 90 degree in 2 secs with load. Motor will accelerate 1 seconds then decelerate in 1 second. I will put a 1:6 ratio gear system between motor and turntable then I needed 160Nm torque which is much.
Is there any other design solution to decrease motor torque? Using planetary gear may be?

crooper said:
goal is to rotate a 500kg(1100lbs) load to 90° in 1.5 seconds.

crooper said:
When I do some math, My motor shoud have torque of 300Nm

crooper said:
turn the table 90 degree in 2 secs with load. Motor will accelerate 1 seconds then decelerate in 1 second. I will put a 1:6 ratio gear system between motor and turntable then I needed 160Nm torque
These numbers are not consistent, so let's start at the beginning. Make a rough sketch of your load, and show us how you calculated the polar moment of inertia about the axis of rotation (which may or may not be the center of the load). Then show us your calculations for the torque to rotate the load, and the maximum speed (RPM) of the rotation. Specifically state any assumptions that you make.

my goal is to rotate a 500kg(1100lbs) load to 90° in 1.5 seconds.

This would be only the first question I would ask. Remember that work is defined by a mass moved over a distance in a specific time. So the bottom line is; no matter how you get there if you move the same mass the same distance in the same time you will need the same effort applied.
Think along the lines of reducing the input torque by gearing will also reduce the rotation rate, so then you need a motor that is faster to get back to the same time (or is that really required). Also having a wide flat load distributed over the table will be more stable but stacking it in a narrow column reduces the inertial effects but can make it very unstable (but maybe it could be supported in some other manner).

The second question that really needs to be answered is; What part of your formula is open to be changed and if none, what other parameters can we play with to achieve the goal?