Dear All,

How to find Gear teeth life cycle using simulation.

As mentioned below I need to find stress of the gear teeth while engage.

What is the procedure to do it.

Can we do it using Soldiworks Simulation Standard.

Regards,

P.S.R.Kirubakaran

Dear All,

How to find Gear teeth life cycle using simulation.

As mentioned below I need to find stress of the gear teeth while engage.

What is the procedure to do it.

Can we do it using Soldiworks Simulation Standard.

Regards,

P.S.R.Kirubakaran

What version of SolidWorks are you using? 2013? 2014? 2015?

Gears are normally purchased off-the-shelf. The gear vendor should be able to give you this information.

The toothforms created by SolidWorks are simplified representations.

As indicated - there might be some basic information you can get, but not sure how useful. Can you attach your assembly here (only need the two gears)?

Yes you can, you can cut the gears and use only a sector of two teeth that come in contact and set a 2D analysis with same depth as the gears you have ( although you should do some corrections according to Shigley or any gear handbook to take into account that you have a helical gear and the simulation is a spur gear). You need a really refined mesh maybe 20-30 elements per tooth, and set the surfaces that come in contact as no penetration point-to-surface contact. Then with the local max stress value at the root of the tooth you can analyse the fatigue using the S-N diagram of the material and you can find also the P-V limit of the gear using the local stress at the contact point. All these you have to do by hand but the simulation will give you only the stress values of the teeth.

The PV calculations should be similar as in this paper:

The Hertzian contact stress will still be calculated incorrectly and should be hand-calculated using a source such as Roark's.

No, SW does not calculate it correctly even with extremely fine mesh. (This is supposed to be fixed in 2016 according to our VAR after I had submitted the problem last year.) I'm sure there are other FEA codes which can calculate it correctly but there are so many easy hand calculations and web pages devoted to this that it really isn't hard to figure out.

I've never trusted contact stress from FEA. I think Hertz hand calcs make fewer assumptions and are more likely to make sense. In complex situations, your best bet is to come up with a linear model and extract the contact forces, then dump the forces into a hand calc. But gears are not a complex situation. FEA just has nothing to add to gear analysis.

I agree with both of you guys, and I didn't know SW has a bug (another one!) with respect to this problem. My situation is that I calculate stresses for harmonic drives, which are not simple problems, normally harmonic drives have 20-50 teeth sharing same load and in my prototypes the number of teeth loaded varies with the load from 3 to 15 teeth so maybe I could use a spreadsheet with the load distribution from FEA and use the hand calcs for each tooth individually. Thanks for the info.

In my situation this is not a harmonic drive. .

Actually using SImulation we can apply torque in drive unit but gear will engage to only one gear. So the result will highlight the max. stress in that particular gear.

So How can I extract the contact forces for all teeth for doing hand calc.

Give me the procedure for hand calc.

Hi Andrei,

Are you sure that we can find gear teeth contact stress calculation using the above method.

You have mentioned thomsonlinear pdf, How to co-relate this with gear teeth stress.

What stress are you talking about? Contact stress or teeth root stress? Fatigue is calculated using the teeth root stress. PV limit is calculated with the Hertzian stress (contact stress). Both stresses can be calculated with FEA or hand calculation. Normally should be calculated both stresses with both methods. There is no method better than the other, each has its own limitations. Read the books Roark's Formulas for Stress and Strain and Shigley's Mechanical Engineering Design, you can find them both as pdf on google.

The short answer is 'no'; gear teeth like this have bearing-type contact (Hertzian) and SW does not provide realistic answers for this condition.

The longer answer is likely somewhat along the lines of 'it depends'; if you are willing to ignore the contact stress, have a fine enough mesh, etc. you should be able to get some valid numbers for the root of the gear teeth. However, I doubt the geometry will cooperate as it must - exactly - for good mesh quality and the needed contact surfaces with the helical gears shown. You could try with just the gears and constrain the IDs appropriately.