Worm Gear Failure 2

[propman]

This is my first experience dealing with a worm gear unit. I have a right-angle worm gear box with a 20:1 reduction. The worm is steel (not sure of the spec) and the worm gear is Copper Alloy C92900. Have a problem with a number of the units having the worm gear failing (teeth breaking). Having trouble correcting the problem.

How critical is it for the two gear centerlines to line up? If they do not, could this be contributing to the failure?

What other methods can be used for this gear reduction that also don't allow backdriving the motor? (The box foot print is approx. 2.5" X 2.5" X 4" high; this cannot be increased by much.)

Any help is appreciated.

 

[dvd]

It sounds as though you might be overloading the gearbox with your backdriven load. Have you looked into the loading due to this? I have had similar failures due to sudden loads applied to the gearbox through the output shaft.

My thinking was that by choosing a reducer that couldn't be backdriven I would be able to restrain a load from moving. The problem was that my assumptions on the magnitude of the forces creating the backdriving load weren't right. So take a good look at at the forces involved. In my case I was assuming the velocity of impact of a tractor trailer at the loading dock to be too slow. Everything worked on paper, but the truck driver proved me wrong.

I ended up separating the restraining function from the conveying function so that the gearbox could be sized for the conveyor load, not the impact load.

Other thoughts on your request are to use a brakemotor and helical-bevel gearbox, or a shot-pin to lockout the impact load from the gearbox. The brakemotor probably won't have enough torque to eliminate all backdriven motion though.

 

[plasgears]

Investigate from the mfr whether the worm is HT material. Contrary to plastic worm gear sets, metal worm drives need a hard worm running against a relatively soft gear.

Is the set enveloping, single or double? If so, is it adjustable? An enveloping set would have a higher contact ratio, which would be more durable than an otherwise equivalent crossed helical set.

Also, an enveloping set would not backdrive as easily as a crossed helical set; more contact area. To answer your question, crossed helical is more forgiving in misalignment.

 

[propman]

Thank you both for the responses. They each provided me some insight into my problem.

In response to some of what plasgears said, I have the following: The worm gear is steel, not sure of the HT. But the set is a hard worm with a soft gear. Thought part of the problem is that the gear is too soft and that is why the teeth are breaking.

The set is single enveloping, but is not adjustable. There is a small amount of freedom in the worm along its axis. It was thought that this freedom would allow the two gears to match better.

Since it is a enveloping setup, will have to look at changing the alignment of the gears to aleviate the problem.

 

[Acajun]

This may be a stupid answer, but has the oil grade and oil level been checked? Usually a 460 grade of oil is used in a worm unit. Oil level is also important, too much is not better than the right amount. What is the mounting position of the unit? This will have to be taken into consideration when the oil level is checked.

Also what is the "Service Factor" of the application? One of the strenghts of a worm unit is the ability to absorb high shock loads. If your "Service Factor" is too low, you could be oveloading your unit. Alighnment of the worm and the wheel is also very important.

"Backdrive" is the term used when the output shaft is driving the input shaft, this can be done with low ratio units without any problem. As the ratio increase, it gets harder to be able to transmit this energy. If you are able to, then you will certainly be overloading the unit.

Most Gear reducer manufacturers have what is called a "BackStop" option. You could use this option with a helical bevel unit. The option is usually on the motor, if you are using a gearmotor configuration, on in the "C Face" adapter if you don't want the motor output shaft in contact with the reducers oil.

Another option is to use a "GearBrakemotor" the brake can hold 150% of the motors max torque. If the motor can move it, the brake will hold it.

Hope this helps.

Jp

 

[BobM2]

Is there wear on the teeth before breakage? Misalignment would tend to cause wear problems before tooth breakage problems.

 

[metman]

propman,
BobM2 makes a good point. Someone please correct me if I am off base but it seems to me that if overloading is caused by misalignment, then wear evidence would show immediatley upon assembly caused by forcing the gears into mesh.

Is there a unit with geometry and reduction for your application available with an alloy steel gear instead of bronze? These steel/stainless steel worm vs Steel gear material combinations are available but maybe not in your particular size/ratio.

I agree with dvd's first paragraph and suspect fatigue failure from backloading. A relatively simple materials/heat treat substitution might solve the problem.

 

[Acajun]

Metman I think that you are forgetting the reason a worm unit uses a brass alloy worm to begin with. The ineficientcies of this type of reducer require that one gear in the set must be softer than the other. Eg. a 20:1 ratio with an 1150rpm input speed has an efficiency of 80%. Where do you expect this energy loss to go? Some of it will dissipitate as heat but some will also dissipitate as wear. If you were using an all steel wormgear it wouldn't last very long hence the use of a softer metal.

Jp

 

[plasgears]

This reminds me of another worm gear set that failed, and the analysis showed that it was grossly overloaded.

Also, your design could be overworked (long duty cycle). There's a big difference in durability when the set doesn't have a chance to cool down.

The right lube, as somebody else alluded, is ISO 460. In years past this requirement was specified as cylinder oil, an animal fat compounded oil. Yes, cylinder oil is steam cylinder oil!

 

[propman]

I want to thank everyone that responded to my request. If nothing else, I got some education on worm gears.

We think we have found the problem and a solution.

To give a summary that will cover most of the points that have been brought up:

The unit is used for about 15 seconds at a time, maybe twice a day (not a long duty cycle).

There is not really a back load; the gear is used so that the weight of what is being moved will not back drive the motor and move on its own, so there is no more than what the gear is moving out.

There is some wear on the gear prior to breakage.

We do not oil the gear because of the low usage (maybe 50 cycles a year tops).

What we found out, and duplicated in a test, is that the people we provide the unit for are over powering the motor and unit. They have a 20 amp breaker, but we found out it may not kick for 20+ seconds. We ran some tests on a unit, found when it hits its stops the motor can pull over 70 amps. Since the gear had stopped, the motor was twisting against its base increasing the misalignment and forcing the worm against the brass gear. We broke one of the gears in 4 cycles. Also, the gear we were being supplied with was a lower strength version.

What we have done is changed to a stronger brass gear. Ran a unit to 150 cycles with no (or only slight) sign of wear on the brass gear. Additionally, we are beefing up the motor mount so that we will not get the same twist (and misalignment) along the worm gear.

Again, thank you all for your responses and information.