Helical gear failure, is it misaligment or overload? 1

[sokenyou]

Hi Everyone,

I have this failed gearbox, which is pretty small, driven by a 1.5 HP motor at 1760 rpm. It drives a what is called Zero pressure conveyor. It failed and caused some downtime. My Manager said it's not just an age thing, it's an overload but only a section of the gearteeth wore out or chipped (broken). I have looked at shafts and broken parts but I thought overloads would be different. Any help or suggestion will be greatly appreciated.


Thanks,
K

 

[sokenyou]

Sorry, attached is a better picture of the failed helical gear.

Thanks

 

[desertfox]

Hi

Can you get the thing cleaned up and then post another picture please, I can't see anything but dirt.
Also if possible post pictures of the mating worm and focus on the failed faces.

Desertfox

 

[sokenyou]

Hi Desertfox,

Thanks. Have you checked the 2nd attachment? Please also see the attached. It's taken after it's cleaned out of grease.


Thanks!

K

 

[Timelord]

sokenyou,

I don't know how a zero pressure conveyor works, but my first look suggests to me that the gear set is overloaded during a specific part of one revolution. Is the load on the gearbox variable during a single output revolution? Have you examined the entire system looking for an abnormal load i.e. bad bearing, jam etc.?

Timelord

 

[MintJulep]

Overload. Either because the conveyor is being used beyond its rated capacity, or something is jammed, binding, etc.

I'll bet you a dollar that if you talk with operating or maintenance staff you will find that this conveyor has a history of tripping it's motor overload protection.

I'll bet you two dollars that you will find the motor overload setting to be turned up to the maximum possible value, or bypassed completely.

 

[Tmoose]

What is the orientation of the gearbox?
Does it use oil or grease?
Are oils with bronze eating EP additives forbidden?
And what is the operating lube level supposed to be?
Have you had the lube analyzed?

Toward the output end it Looks like kind of milk-shakey as happens with water contamination of oil built with detergents. Then again, the residual lube (?) on the gear and opposite bearing are dry as Christmas 2007's fruit cake. (And I //like// fruit cake).

 

[tbuelna]

From looking at the picture, it appears that the worm still has most of its teeth, while there is a sector of the gear where the teeth have been completely worn away down to the roots. Seems like one section of the gear's teeth experienced enough wear such that there was no longer sufficient contact ratio with the worm to produce a continuous transfer of force as they rotated. The worm teeth simply began to skip over the gear's teeth and eventually wore them away.

 

[tygerdawg]

I'm with Mint. Looks like an overload condition.
I've got some experience with different types of gearmotors in a previous life. Worm drives IIRC have softer gear sets, less efficiency due to sliding contact, and deal better with shock loads. But once the damage is initiated (by too-heavy shocks?), the tooth & flight faces will gall and it starts to degrade fairly quickly. Then there's all the other factors that contribute to a bad situation: poor or perhaps no maintenance on the lubricant, clogged breather (if any), and so on. I suspect (like many things installed in industry) that it was told to do something it was not designed to do.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[tbuelna]

Ideally, there is little difference in the approach to how worm gearsets are designed versus other gearsets. The materials and heat treatments chosen for the worm and gear are usually selected to provide an equivalent fatigue life between the two gears. Since the worm flanks usually experience a far greater number of load cycles than the gear flanks, the worm is usually made from a more durable material like hardened steel, and the gear is made from a less durable material like bronze. The relative difference in hardness between the worm and gear teeth also reduces the susceptibility for galling to occur.

 

[desertfox]

Hi

Well it must be my eyes but I still cannot see very much although on your last image I can see all the teeth are gone in a small section.
Does the worm wheel only operate on a segment of the gear or through 360 degrees and how long as the gear unit been in service?
I have included this site so you can compare your failure with some gear failure images.

http://www.tribology.co.uk/services/investigate/gear.htm

 

[sokenyou]

Thanks everyone for your input. We are now investigating if someone had jammed something in the conveyor that caused this overload (if it is true), but also the last time this gearbox was replaced remains a big question mark. The Maintenance Mgr claimed that the gearbox was in there for over 15 years but the company we work for was acquired back in 2010 so records prior to that are nowhere to be found, so we don't have proof. I got my number from asking the distributor but it's not 100% accuarate.

Anyway, if I have more updates, I will post them. As far as I am concerned, I will put in a PM task to check this gearbox for play on a semi-annual basis. Check for excessive heat, check chain/sprockets on a monthly basis.


Thanks!

 

[limey1]

I was wondering if this conveyor starts and stops, and if so, at the same position. That would add probability toward Tbuelna's theory.

 

[chicopee]

If the gear box horsepower rating matches the motor HP and exceeds the starting torque requirement of the conveyor than overload may not be an issued. I doubt misalignment of the gear train unless the bearings are shot. Jamming the conveyor may be, however, there should be some visible damage to the conveyor drive train external to the gearbox.

 

[romke]

was it a sudden failure or a failure that occurred after a longer time of non-standard operation (higher temperature, more noise)? looking at the lubricant it looks that the temperature has been quite high, and also it looks like moisture may have been present. did you check with a simple crackle test whether there was some moisture in the lubricant?

when there was, that may have been a cause. when a moisture droplet gets trapped between the mating gears and heats up, it may vaporize - and the lubricant film collapses locally giving rise to wear. once when wear has started conditions will get worse and the loading on that particular tooth will be higher and also somewhat so on the neighbouring teeth - from there it will go on worse.

in the end you will most likely see scuffing on the bronze gear and wear particles circulating will speed up the wear process still more, leading to failure.

one of the bearings also indicates moisture has been present, leading to corrosion.

 

[tbuelna]

Looking at the pictures, what seems strange is that there is a large sector of the gear's teeth completely sheared away down to the root diameter. But in the same picture the mating worm teeth seem relatively intact. Based on this picture, the only type of failure I can imagine is an overload condition at the gear, which resulted in the worm gradually shearing away the gear teeth.

 

[berkshire]

This is anecdotal,
I have seen the same kind of worm drive gearbox failure on a belt conveyor some 50 years ago.
The conveyor would be stopped at the end of the working day with product on it. For some unknown reason ,it always stopped at the same position.
At start up the hypoid oil had drained to the bottom of the gearbox, so for about half a revolution the gear was almost dry. This caused excess wear on the same teeth in the box, until one failed, the worm drive spanned fewer teeth at higher load, until another failed, then a cascading failure took out the rest of the teeth in that sector of the box until the conveyor would no longer start. As long as there were sufficient teeth in the sector, the conveyor ran, It would just not restart.
The solution at the time was to replace the box with the same make and model, and make sure the conveyor was run to empty at the end of the working day reducing the startup load.
B.E.

You are judged not by what you know, but by what you can do.

 

[sokenyou]

Thanks everyone for the great feedback.

The Conveyor system should be running continuously non-stop except every Monday on shutdown PM days BUT due to the deficiencies in our sortation systems, the conveyor starts and stops upteem number of times in a day. The guys here including the Maintenance Supervisor think it's a sudden failure because there were no symptoms apparent to us prior to the complete failure. There are noises coming out from the entire system so it's hard to judge by the noise and if there was, it was loud. Heat wasn't checked but it's warm now we have replaced with new. I kinda of agree with you Berkshire as it started with one tooth failure and a few other came aboard until the slot for the mating is big enough for the helical gear to sit in like a slot. The gearbox still ran until there is insufficient teeth to start.

It's hard to run the conveyor to empty because it's starts and stops. I think looking for a beefier drive might be a better choice?

Thanks,
K

 

[3DDave]

I can't tell definitively from the pictures - but it doesn't look like simple overload failure. Simple overload causes teeth to be broken off. Insufficient lubrication will cause the teeth to wear until they are too thin and the remains get sheared off.

This sort of wear (due to poor lubrication) was the cause of failure of a drive nut in the Alaska Air crash a few years ago. See

http://www.ntsb.gov/investigations/summary/AAR0201.html,

figure 20. There was no warning to the flight crew though backlash checks (which were skipped) would have certainly detected the problem.

Since it is continuously driven, it's possible the typical load is too high for any lubricant to matter, which would cause accelerated wear. This might cause fatigue to damage the the worm face, leading to a high-abrasion rate. Is the worm in good shape( smooth) or is the worm tooth face rough? If it went 15 years, that's not likely.

For the service it is getting, I'd suggest oil rather than grease. In cases where there is a wiping action, grease can be ineffective. Worm drives are all wiping action connections. If the oil-film can't reform fast enough, it's like not lubricating it at all. Also, I agree with what Tmoose said.

For a quick check, get a power meter on the motor to see that there isn't a periodic high load from the conveyor. It is certainly tempting to think something like a bent shaft causes a large increase in load that hits the same segment of the worm gear each turn, causing just that segment to wear faster than the rest. Motor inertia would push through the bind and have most of the turn to get back to speed. There wouldn't be a noteworthy increase in noise.

I agree on the PM plan; make a measurement of the gearbox backlash by first driving the input to take all the clearance, then locking the input and turning the output in the same direction. This should avoid including slop in worm support bearings in the backlash, and will also make it repeatable for later. Repeated at intervals it should show a linear increase in backlash as the worm and worm gear wear. The rate of wear will increase dramatically if surface roughness increases or lubrication fails. The trick is figuring out a backlash limit. Making a measurement on the existing box (with the remaining teeth) could provide a guide.

For the longer term, especially if using an oil bath, monitoring the power into the motor and the temperature of the oil vs ambient will also indicate a change in operating condition. The value of this is adding predictability to the down time.

Alternatively, put in a P.O. for a new gearbox, delivery date in 14 years, 6 months.