Worm drive gearbox "coasting" problem :-( 1

[bithkits]

At work we have an electric sliding gate motor under development. I inherited the project and unfortunately extremely little information is available to me. It uses a worm drive system with the following spec's:

Worm (bright mild) - single start, matched profile to worm wheel.
Worm wheel (PhosphorBronze1) - helix angle = 4.398deg; no. of teeth = 36; module = 2.
Electric motor - DC12V, 50W, 3000rpm max.
Gate mass = 350kg (approx)
Output to gate - rack and pinion, Nylon6

When the gate is stationary it is not possible to turn the gearbox shaft by hand or by pushing the gate (this is desired for this worm drive, so this is fine).

The problem we are having is the following:
When the gate is moving and the motor shuts off, the gate will continue moving under its own momentum. Also, if the gate is pushed by hand while it is moving and the motor has shut off, it can continue moving indefinitely.

Is there a way to increase the "drag" on the system or possibly some other solution? Are the gear tooth profiles correct or would a different helix angle help?

Ideally, the gate should slow down and stop soon after the motor has been shut off. The electronics guy tells me that using the motor to "brake" by applying reverse torque is not an option. But to be honest, I am not entirely sure if the problem lies with the gears. The original spec's required grease, but 75W90 oil is now used. Also, there used to be a steel coupling between motor and worm, but the worm is now directly attached to the motor shaft.

I have little experience with worm drives so any information will help. Thank you kindly!

Adriaan.
I am an Engineer/part time student (Mechatronics) from South Africa.
Advice from lecturer: "Be warned - when you go into industry your boss will give you a thousand things to do and he wants them done yesterday!" So far he is right...

 

[DesignerGuy16]

Lot of things you can do. A VFD is a option (and yes, VFD's can "brake", sensors to control the motor so you can control the "coasting", an electromagnetic brake on the motor, hard stops, etc.

James Spisich
Design Engineer, CSWP

 

[MikeHalloran]

OR,
if the motor is controlled by relays, it should be relatively simple/cheap to set them up so that a braking resistor connects the motor terminals when power is removed.


Mike Halloran
Pembroke Pines, FL, USA

 

[Compositepro]

An electric brake on the motor would be my choice. Dynamic breaking with a resistor does nothing when the motor is stopped or slow. Electricity that turns the motor turns the brake off. These are often sold prepackaged as a brake motor. The brake can fit on the fan end of the motor.

 

[MikeHalloran]

OP stated that the worm won't backdrive from a standing start, so I think a motor brake would be overkill, unless the braking resistor is not effective enough in deceleration.






Mike Halloran
Pembroke Pines, FL, USA

 

[BrianE22]

You need something to deaccelerate the motor to a low enough speed so that the coefficient of friction in the gearbox jumps up.

You already have a low helix angle. For self-locking you need the coefficient of friction to be less than the tangent of the lead angle:

mu <= tan (lead angle)

 

[drawoh]

bithkits,

The ability of a worm drive to backdrive is based on its geometry, mostly the lead angle. My gut feeling is that you are on the margins of being back-driveable. If you went to a worm drive with a smaller lead angle, your problem would be reduced or eliminated.

This all should be explained in your machine design textbook from college.

JHG

 

[bithkits]

(I have a gut feel that the ramp profiles the PWM controller is implementing are causing the problem, but I don't know enough about embedded electronics to raise the issue until I have exhausted all other causes. However logic tells me that a worm drive that locks from a standing start can only move if the worm is rotating, therefore the motor is rotating, therefore the problem lies with the motor or electronics. Apparently during power cut off to the motor only the ground is lifted by the one relay, while the V+ remains connected - hmmmmmmm maybe there is another earth somewhere)

Lot of things you can do. A VFD is a option (and yes, VFD's can "brake", sensors to control the motor so you can control the "coasting", an electromagnetic brake on the motor, hard stops, etc.

I spoke to the electronics guy and apparently PWM is currently used. Then he mumbled something about relays not coping and some other stuff that wasn't explained well. Figures... Thanks for the suggestion though!

OR,
if the motor is controlled by relays, it should be relatively simple/cheap to set them up so that a braking resistor connects the motor terminals when power is removed.

Interesting! The electronics guy at work did not like the idea but I am going to try that myself anyways. Thanks!

An electric brake on the motor would be my choice. Dynamic breaking with a resistor does nothing when the motor is stopped or slow. Electricity that turns the motor turns the brake off. These are often sold prepackaged as a brake motor. The brake can fit on the fan end of the motor.

Unfortunately the addition of a brake is not viable since the design has gone very far with years of development behind it. But I will look into such devices in any case for future products, thank you.

You need something to deaccelerate the motor to a low enough speed so that the coefficient of friction in the gearbox jumps up.

You already have a low helix angle. For self-locking you need the coefficient of friction to be less than the tangent of the lead angle:

mu <= tan (lead angle)

Is this self-locking from a standing start or even when moving? I will investigate further. I assume a higher coefficient of friction will lead to increased wear patterns and lower efficiency since worm drives are a sliding gearset? Thanks anyway!

bithkits,

The ability of a worm drive to backdrive is based on its geometry, mostly the lead angle. My gut feeling is that you are on the margins of being back-driveable. If you went to a worm drive with a smaller lead angle, your problem would be reduced or eliminated.

This all should be explained in your machine design textbook from college.

Thanks for the advice. A lead angle change may be the quickest fix at the moment. I studied my degree at University, not college and it was very theoretical - we never dealt with any worm drives, only spur and epicycloid very very briefly - dealt more with gear ratios than anything to do with tooth profiles/lead angles/pressure angles etc... Luckily I found some good literature on gear design.

Adriaan.
I am an Engineer/part time student (Mechatronics) from South Africa.
Advice from lecturer: "Be warned - when you go into industry your boss will give you a thousand things to do and he wants them done yesterday!" So far he is right...

 

[MikeHalloran]

If you're already using a PWM controller, there _may_ be a simple way to add a braking resistor. ... but you can't just slap it on anywhere. Get the sparkys to help you.



Mike Halloran
Pembroke Pines, FL, USA