Spur gear excessive wear in harsh environment 4

[strokersix]

I'm having a problem with excessive wear on an agricultural machine design I'm responsible for and would like advice.

A pair of spur gears, 5/8 inch face width, 22 tooth meshing with a 36 tooth, 6 diametral pitch, 20 degree pressure angle run at approximately 200 RPM and 100 lb-ft torque are showing excessive wear. They are used on a piece of agricultural equipment and are run dry, thinking being grease will attract dirt and promote abrasive wear. Current design is 1045 steel, hardness 45-53 Rc, .060 inch minimum case depth for both gears. Life goal is 1000 hours. Wear of approximately .020 inch is evident at 200 hours. This wear has been observed on field durability machines and also on a lab test stand in a clean environment so I don't think dirt is a significant cause. I have begun testing some samples of 8620 material with Rc harness of 60 but they have not shown much improvement to date. I've also tested the 22 tooth 8620 and 36 tooth 1045 combination.

Can anyone give advice on my problem? I'll probably try greasing the gears if the 8620 doesn't solve the wear issue. I would rather leave the gears dry if at all possible.

Thank you, Mike

 

[diamondjim]

You might try using a 21 tooth instead of the
22 tooth unless you need that particular ratio.
Mike,
Cut a 21 tooth on the same blank as the 22.
You basically want to try a long addendum design
on the pinion thus the 21 on a 22 tooth blank
and same center distance. If you can change the
center distance and go with a 50 percent long
addendum on the pinion, that would be ok as well.
I would use the 8620 material as you suggested
on the pinion or use 4340 material and probably
get a little greater surface hardness on the pinion.
Can you better define the .020 wear? .010 on each
part? What do the addendum and dedendum of the parts
look like with the .020 wear? I like the idea of
no grease. Have heard that others have come to
the same conclusion in very dirty atmosphere
conditions. Interesting.

 

[plasgears]

Don't fool yourself; you need lubrication in metal gears. Prelube is good sense in PM gears, plastic gears, and any gears. In one application PM gear prelube, in the form of an oil bath collector off the sintering line, made the difference between pass or fail. In our plastic gearbox lines we prelubed all gears, even when we added grease to the gearbox.

 

[strokersix]

Unfortunately the 22/36 ratio is fixed. A deviation from this must be minor, say 1 percent or so. The gears are swapped with each other for high/low range and deviating from this ratio will be a problem. Centerline is flexible but would desire to keep the same.

Lubrication would be great but for a couple issues. One, we can't count on proper maintenance of the machine in the field so requiring lubrication carries a risk of not getting it. Two, these gears are exposed and will spray lubricant and make a heck of a mess as well as being messy to change the high/low range. If proven that lubricant is the only reasonable way to get the needed life then we may have to deal with the mess.

Fortunately, only about 10 percent of the applications have the highest power requirement. The lower load of the remaining 90 percent of the applications still produce wear on the gears but they survive for 1000 hours.

I'll post a link here to a photo of worn gears.

Thanks in advance for all comments and suggestions!

http://users.qconline.com/~friestad/spurgears.jpg

 

[diamondjim]

We had a case similar to yours and they
greased the gears. The dirt and grease
combination ate up the gears at a quicker rate.

You could keep the same center distance and gear
ratio and go with 25 percent long and short
addendums to see if that would be an improvement.
It would require different gear blanks.

What dedendum ratio are you using? The clearance
looks excessive. I was surprised that you
have hardened gears without having full fillet
radii.

 

[automatic2]

I agree that the clearance looks excessive. Is your alignment too flexible. Unless you can adequately house, forget the lub. A gear belt may be the most practicle.

 

[strokersix]

The excessive clearance appearance is due to the wear. when new the clearance was much less. The shaft centerlines were checked and within about .020 inch of nominal design specification. Whether we can duplicate this shaft centerline in production is another issue entirely! I would love to run a gear belt but rotation reversal is required, plus it's going to be tough to beat the cost of these two spur gears.

These gears are supported on 7/8 inch hex shafts which run in good ground race ball bearings right next to the gears. There is a bit of clearance between the hex shaft and hex bores but I don't think that's a major factor in the excessive wear.

diamondjim: I must admit my limited gear design expertise and that I copied the geometry from an existing part and did not design this gearset from scratch. Will long and short addendums require special hobs or is that a standard profile?

The ideal solution is a material selection that lasts longer. I considered powdered metal but the tooling cost was prohibitive. Perhaps just tool up the smaller gear in powdered metal and oil impregnate it?

 

[diamondjim]

Strokersiz,
With a 25 percent long and short addendums
the od of the 22 tooth gear would be 4.0833
and the tooth thickness .2901/.2861
measurement over .28 dia. pins 4.102/4.090

the od of the 36 tooth gear would be 6.2500
and the tooth thickness .2295/.2255
measurement over .28 dia. pins 6.290/6.279

The same hob rack cutter could be used to
cut these parts by withhold the cutter on
the 22 tooth and sinking the cutter deeper
on the 36 tooth gear to get the tooth
thicknesses shown above.

I would be surprised that the pm parts would
last any longer than the quench and tempered
parts. Are the gears nitrided or induction
hardened to get the .060 deep case?
If the parts are plated, are the gears surfaces
not plated?

If you are getting this type of wear without
lube on a test stand, something will have to
change. What kind of gear quality are we talking
about? You might want to stub or trim the larger
gear to reduce the approach action and see if
this helps. I am still puzzled why the dedendums
are so great and yet do not have full fillet
radii. What kind of hobs are you using?
The flank wear seems to indicate that the gears
go in one direction only? I am confused on the
number or hours on the stand and the number
of hours on the parts. Does this represent
hours from two different sets of gears?

What kind of temperature is the gear being
subjected to at 200 rpm?

 

[strokersix]

diamondjim,

I have answers to some of your questions.

The 1045 gears are induction hardened, quenched and tempered with a minimum case depth of .060 inch. The 8620 gears are carburized, quenched and tempered with a case depth of approximately .030-.040 inch. Plating is on the entire 1045 parts including the teeth, no plating on the 8620, simply to epedite part sourcing for test. If 8620 is adopted for production it will be plated.

No AGMA gear quality is specified on the print. Tool marks on the root diamter indicate the hob was advanced .100 inch per gear revolution. Surface finish on the involute looks pretty decent for this intended application. Other than that, I don't know what kind of hob is used. Correct, the gears always drive in the same direction. I could turn them over to gain more life, perhaps. The photo shows hours on stand and hours on the second set of gears. The first set was 1045/1045, the second 1045/8620, and the third will be 8620/8620, the fourth most likely 1045/1045 with grease and some dirt thrown into the mix.

Actual environment could be as hot as 100F plus bright sunshine. Test environment is 70F and inside. I called the test engineer to find out the gear temperature and actual speed/torque numbers. I'll post that information soon.

As far as mounting centerline goes, I agree that it is important to maximize gear life but I don't think it is a major contributor to the observed wear. I say that because we have observed this wear on field machines and test stand on multiple occasions, even when there has been minimal backlash when new. There will be some sort of assembly tool to center the shafts in production.

I just found out that another factory in my company has very similar gearing with altered geometry. Sounded like the long/short addendum modification. I have a call in to the engineer in charge of those parts to learn more.

Thanks for all the comments to date!

 

[strokersix]

I just got the drawings of another gear set. 6 diametral pitch, 20 pressure angle, 18 and 44 tooth. The teeth have about .100 inch taller overall height, some of which is in the full root radius. Actual increase of the involute is probably about .060 - .080 inch. The teeth are specified as cut with a topping hob, which in this case I think is a special design intended to produce quieter running gears. Don't know that for sure though.

If the tooth is taller then the contact ratio will increase, reducing load per tooth but increasing sliding velocity at entry and exit, correct? Help or hurt my situation?

The gears on the test stand are running 120-130 degrees F in 75 degree ambient. Speed of 36 tooth is 200 rpm, torque is 75 lb-ft.

An additional test I'm contemplating is swapping in a brass 36 tooth running on a steel 22 tooth. I suspect that brass gear will be destroyed but worth a try?????

 

[CDGear]

I would suggest trying to reduce the Specific sliding (Slip) ratio of the gearset. I am assuming since you do not state the center distance that the gear and pinion operate at 4.833" centers (Standard Centers) this would gives slip ratio on the pinion of 3.184 and 1.609 on the gear.

The slip ratio can be reduced by increasing the pressure angle to 25° and utillizing long and short addendums. From my calculations it would appear that 0.13 long on the pinion and 0.13 short on the gear would be the most effective in reducing the slip ratio.

The slip ratio is an important measure of the frictional characteristics of the gear an pinion mesh.

Hope this has helped

 

[diamondjim]

You mention that the whole depth increased on the
other parts and not the face width of the gear.
If a topping hob was used, it probably also cut a
small radius at the tips of the gears which might
help.

I would not go to 25 degree pressure angle design
but the long and short addendum is still appropriate
for reducing the sliding velocity. I am assuming the
small gear drives the larger gear. If this is not
so then make the larger gear have the long addendum.

Using brass on the other part may change the friction
and sliding dynamics, but will be surprised if it
makes any difference. Whichever is the driver, I would
expect it to be the harder material. Keep us posted!

 

[strokersix]

chrisdale,

Can you please elaborate on what slip ratio means is practical terms and how it is calculated? How will 25 degree PA improve slip ratio?

diamondjim,

The gear pair is swapped one for the other for high/low range so sometimes the pinion drives the gear and sometimes the gear drives the pinion.

One thing I'm going to do for sure is add a second cross hole in each hub so the gears can be flipped over to make use of both sides of the teeth. I'm also going to try grease/dirt just for curiosity, even though it's an ugly solution from the customer's standpoint. I'm still thinking about the oil impregnated powdered metal idea. We have had good experience with PM roller chain sprockets and other parts.

I'm using Darle W Dudley "Handbook of Practical Gear Design" as my primary reference and have a copy of AGMA standards. Does anyone have recommendation for other reference material online or hard copy that an inexperienced gear designer would find helpful?

Thank you.

 

[diamondjim]

Stokersix,
If both gears can be drivers, then
do not incorporate any long and short
addendums. You could incorparate double
long addendums if the SAP dia. is nearly
the same as the Base dia.

Earle Buckingham's book "Analytical Mechanics
of Gears" is another must in gear books.
He likes to use radii rather than diameters
in many of his calculations which helps sometimes
to understand the equations along the line of
action to visualize the triangles formed by the
two base diameters of the gears. It sometimes
helps to see these two triangles to understand
the formulas. I also like the Machineries
Handbook section on gearing. Ray Drago and
Errichello also have books of interest on gear
design. Drago does cover slip ratio rather
thoroughly. I do think with high revolutions,
this might be something to be concerned about.
You can generally look at the wear patterns
early on and see the scoring or minor sliding
marks at the first start of wear. Look also
at arc or approach and recess action to help
understand much of the wear patterns. Errichello
teaches a course on failure and wear of gear teeth
with many helpful examples and pictures to illustrate
gear wear and its causes. Tip relief may be something
else you might want to consider for your design.
You do lose contact ratio with tip relief or
stubbing teeth but you seem to have plenty of
contact ratio in this design. If the gears are seldom
back driven, you still might want to consider
long and short addendums.

 

[carburized]

Stokersix

First thing you need to check, if 5/8" face width is enough for carrying the load. Probably effective face width is

1/2". And, make sure you do no have any misalignment which effectively will reduce the face width. Next you want a long addendum and short addendum gearing. You can enlarge the addendum on the pinion to the extent where contact length on the gear and the pinion is about same, but keeping contact ratio at least 1.25. We in the gear industry
design worms most of the time with long addendum. Material could be 8620 stl with case carburized or 4340 stl with nitriding. I would think 1045 would work as it has enough carbon content for wear resistance. I have worked with some agriculture gearing and they go through lot of abuse. You need to put some extra factor of safety. Usually 2:1. In other words design these gears for 200 ft lbs. of torque rather than 100 ft. lbs. Steel on steel with no lubrication are notoriously bad and gall rightaway.

YOU MIGHT want to try Meehanite SH100 grade ductile iron gears. They can be driven dry at low rpm. You need to cut in soft state and then Induction heraden. That will solve your problem.

 

[MattRR]

strokersix,
In 1996 I had to design some gears that ran hot and dry. We had two solutions. If the speed is very very slow <1000 ft/min, and contact stress under about 80 ksi, then you can use tungsten carbide flame sprayed on the gear teeth. It the compressive stress is over 80 ksi, and/or the pitch line velocity is greater than 1000 ft/min, then non-involute gears work well. We had an application where the involute gears destroyed themselves in a few hours, but the sprayed gears and noninvolute gears both worked well. We went with noninvolute gears as they were lower cost. We had previously tried various means, including using full recess action gears, and nothing worked. Note that all the old clocks used cycloidal gears, my understanding is that there is no way involute gears survive unlubricated unless they have a good coating.
If you would like more details, email me at mhawkns@aol.com.

 

[JRCD]

JRCD

For good evaluation of gears teeth surfaces deterioration will be desired photos of teeth surfaces, in addition to gears lateral view.

Within limits of available data:

1- Teeth surfaces are very worn in one side, but the opposed side is in good conditions, so gears life will be doubled if they are inverted after half life;
2- Generally the dedendum wear is more accentuated than the addendum wear, mainly if long and short addendum is not used, but photo dedendum wear seems too excessive. Maybe tooth surface is overloaded or deteriorated by the tip edge. I suggest rounded tooth tip edge or tip relief on both gears;
3- If deteriorate surfaces are not in the tooth working side, than gears may be overloaded by impacts or great inertial loads;
4- Gears working at dirt environment generally run with big backlash, but photo backlash seems excessive. Is the photo gears center distance exact?
5- Gears most have close fit over hexagon shaft;
6- Gears design can be optimized by long and short addendum, without necessity of any change of center distance, gear ratio and tooling, but in this present case the selection of the profile displacement must be take in consideration that gears can be swapped between each other, for high and low range;
7- Good lubrication improve all gearing, but dirt environment lubrication is problematic. You may try solid lubrication like with Molybdenum Disulphite;
8- It’s recommended that the small gear has to be harder that the wheel. AISI 8620, casehardened to 58 – 60 Rc and case depth of 1,0 to 1,2 mm probably will be a good choice for the pinion.

 

[JRCD]

Gears are running at very low speed for significant abrasive wear. I believe more at an overload or tip edge cutting wear, which will be aggravated by eventual hydrogen embrittlement caused by the gears plating. After plating gears must be backed to mend this. Hydrogen embrittlement deleterious action is more pronounced at stressed parts from the heat-treatment, like the gears we are talking about. It will be worse if those gears are manufactured from steel plate, and not from rolled bars.
Running at correct center distance is necessary to avoid interference or excessive backlash. Because of this we need close fit between gear and shaft. In this circumstance the misalignment probably will not be excessive because of the small ratio between tooth face width and the tooth size, but the probable small length of gear hub may be a problem.

 

[strokersix]

All,

Thank you very much for the shared thoughts on my gear wear issue.

The hex bore measures about 0.883 inches across the flats while the hex shafts come in at about 0.874 inches. I would love to have a tighter fit but assembly issues force this clearance. I don't believe this shaft clearance is a major factor. When torque is applied the hex tends to self-center and eliminate looseness in the bore. Hub length is 2 inches. Hydrogen embrittlement is something I had not considered.

This is the direction I have chosen to explore: 100-70-03 ductile iron cast blanks, 1.25 inch face width, cast in retention feature on both ends of hub to allow wear on both sides of teeth, teeth machined to same profile as current design. I could use some advice on heat treat specification. Also any other comments on this chosen direction will be most welcome.

Thank you.

 

[PaulJuhnke]

Ductile cast iron is a great choice! It’s often the best choice for open kiln drive gears, which are hot and dirty.

Hard, thin coatings are available via Eutectic and others, which will probably further reduce wear rates.

Sprayed graphite or Molybdenum Disulphite lubricants will obviously help. If applied dry, or sprayed and allowed to dry, they will cling on the surface and in the pores of the ductile iron, and greatly reduce wear.

Is there any way to dribble or spray a dry powder lubricant via a small automated delivery device, onto the gears every 30 minutes or so during operation?