Fairfield Software Question 2

[cfordyce05]

I'm working on a spur gear design and would like to get a rough check of the gear strength so I downloaded the Fairfield software and entered my data. I have a question on the material strength entries. I would like to use 17-4PH in an H900 condition so under Rating Stress for bending I have 77,500psi. From the datasheet the UTS is 240ksi. What am I suppose to put in the surface stress box? The default is 225000psi. Is it actually wanting the UTS there? The directions when you click on the bending stress box are:

"Enter a Rating Stress for Bending. This is the material allowable stress, also known as the Endurance Limit Stress. The default value is 65,000 PSI, which is for carburized and hardened gearing per AGMA Grade 2 requirements."

The directions when you click on the surface stress are:
"Enter a Rating Stress for Surface Contact. This is the material allowable stress, also known as the Endurance Limit Stress. The default value is 225,000 PSI, which is for carburized and hardened gearing per AGMA Grade 2 requirements."

They both say that it is the Endurance Limit Stress but show two different numbers.

Thanks,

Casey

 

[tbuelna]

The tooth bending is based on root tensile stresses. The allowable stress limit you should use will be based on loading conditions (unidirectional or reversing loads), the number of load cycles at the stress level, the level of reliability required (ie. L10 or L2), mechanical working like shot peening, etc. For your 17-4PH H900 material, 77.5ksi root stress sounds high, unless your number of fatigue load cycles is very limited. The tensile endurance limit for this material with unidirectional loads would probably be less than half that number.

As for contact stress, the endurance limit is affected by both the subsurface shear strength and the surface compressive strength of the material. Once again, 225ksi is way too high for 17-4PH H900 material unless the number of fatigue cycles at this load are very low. The surface contact endurance limit for this material is probably far less than half of the 225ksi number.

Hope that helps.
Terry

 

[cfordyce05]

Thanks Terry. Do you know where I can find material strengths as they relate to gear applications? I am trying to determine what face width will give me the strength that I need. The application is the gear train for a small RC type servo. The gears need to be 99.999% reliable as they will be going into an aerospace application. The motor speed and torque, DP (80 or 96), and tooth counts are decided to get me 45oz-in continuous and 85oz-in stalled on the output shaft. The 14 tooth pinion is driving the 35 tooth gear with 33oz-in stalled.

Do you have any recommendations for a material for this size of gears? Is there an alloy steel that is commonly used? I don't think I need a corrosion resistant material because the gears will be sealed in an IP67 housing with grease lubrication. I am not opposed to having the gears hardened. I am trying to make the gear train as light and compact as possible without adding extreme cost. Every thing is a trade-off...light and compact = $$$. I looking at keeping the face width on the final gear .100" to .200". I'm not extremely worried about the other gears because the torque is so low. I also wouldn't mind using a titanium alloy, I just need to know what fatigue numbers to be using and size accordingly. The endurance limit is usually about 40% of UTS correct?

Sorry for the newbie questions. This is my first stab at gear design. I've been reading everything I can get my hands on.

One other thing, if I were to just size the face width using the Lewis Equation, would this be a conservative enough number or would I be setting myself up for failure?

Casey

 

[tbuelna]

Unfortunately, the small size of your gears (80 or 96 DP) makes it impractical to use the conventional gear design equations. A 14T, 96DP pinion has such small teeth that it also becomes very difficult to meet the normal tolerances required. And the teeth are so small that it is not possible to case harden them, so they must be thru hardened.

The max face width for a 14T, 96DP spur pinion would typically be around .15". A safe value to use for endurance limit in unidirectional root tensile bending stress for most steel alloys would be around 20% of UTS.

Good luck.
Terry

 

[israelkk]

cfordyce05

From your questions I get the feeling that you have little knowledge how AGMA rating was constructed and what is the basis for the bending and surface strength rating. I suggest you throughly study AGMA standards for gear rating and material selection before plugging input to Fairfield program (as a wise man said: garbage in - garbage out).

It is specifically mentioned there in the program that it is for experienced users and only actual tests can verify the gear complience for the job. Just for the record, AGMA rating is based on the steel hardness and not the tensile strength (both for bending and surface rating).

One more problem is that from experince the hardness of PH steels can vary by 10RC for the same heat treatment. For example, H900 hardness can vary in the range of 45 to 55 RC for same ultimate tensile strength.

I have no idea where did you take the 240ksi as the 17-4PH ultimate tensile strength. According to MIL-HDBK-5J (MMPDS-01 31 January 2003) which is the bible for aerospace metals it is only 190 ksi (bar per AMS 5643).

As a long time designer and developer of very high strength gearboxes for aerospace, using fine pitch gears (DP of 96,80,72,64,48,32 etc.) including the use of high strength stainless precipitation hardening alloys, I feel you are for a big challenge.

For example, selecting 17-4PH in the H900 heat treatment is risky. Here is a quote from MIL-HDBK-6J "The impact strength of 17-4PH, especially large size bar in the H900 and H925 conditions, may be very low at subzero temperatures; consequently, the use of 17-4PH for critical applications at low temperatures should be avoided." Although your gears are small, from experience, H900 should be avoided.

How are you going to hob the gear in the H900 condition (45 to 55RC)? If you are thinking to machine it in the "Condition A" then it will be difficult to machine and the gear will shrink during heat treatment. I suspect that for the application you mentioned you need precision gears (at least such as AGMA 2000-A88 Q10C) therefore, you probably will not be able to meet the dimensions after the heat treatment without additional grinding which I am not sure that grinding is possible for DP 96.

Another difficulty is how to remove the heat tint after heat treatment before passivation without changing gear dimensions. It is difficult if possible to control the amount of metal removal. Using chemicals will/may create hydrogen embrittlement risk. Relying on hydrogen removal by baking for such high strength steel is risky and unreliable.

Another issue that bothers me is that you are trying to design a miniature high strength gearbox using high strength alloys for fine pitch gears. Fine pitch gears are common for light load gearboxes (such as gears made from PIC, SDP, BERG, etc. gears) at the same time you asking for high reliability and $$ consideration in addition to meeting aerospace standards, which I find to be full of contradictions.

Even mentioning Titanium for such application shows no knowledge of galling issues and is surely not in the $$ reduction direction.

 

[israelkk]

Correction MIL-HDBK-5J.

 

[cfordyce05]

israelkk,

It is true that I have little knowledge of the AGMA rating. This is my first experience in performing gear design so I am trying to go through the process slowly so I don't miss anything. I am becoming familiar with their strength formula and design factors. It is slow going. There is a lot to take in, but I want to do it right. Your response is the kind I was looking for. The comments on heat treat considerations are especially useful. I knew there would be dimensional changes with heat treat, but it looks like there are quite a few more aspects of it that I need to pay attention to.

I am not in love with 17-4PH. It just appeared, as a new designer, to be a logical strength step while still being stainless. I do not need to use a stainless steel. I would love to hear your recommendation for a material/heat treat process for use in these small gears. If I can get by with 96DP that makes my diameters smaller, but I think the biggest I can go is 80DP to maintain my size requirements. At this point in the design process, I am open to all directions for materials. For a budget, my product could handle between $30-40 per gear for low volumes, but I would need to come in well below that for production. I cannot purchase off the shelf gears for this application so I have a blank canvas to define what I need to a local gear manufacturer. Being new to this field, I have been basically looking at other manufacturer's designs and then comparing them to my own. I'm starting to get the feeling that the even the manufacturers that I have been comparing to may not have done the level of design requirements that I'm trying to achieve. It's looking like I basically need to design as best I can to an aerospace application and confirm through testing that the gearing will last the intended life of the product.

I took the 240ksi number from an AKSteel datasheet for the H900 condition. Thank you for your advice. I really appreciate it.

 

[tbuelna]

cfordyce05,

I've attached some (old) AGMA basic design guidelines for root and contact stress limits at various numbers of load cycles. Grade 2 basically means premium aerospace quality (double vacuum melt) materials and heat treatment. If your gear tooth loads are reversing, you should use 70% of the root stress limit shown. The root stress limits also do not take into account mechanical working of the surface using processes like shot peen.

I believe the root stress limits shown for grade 2 materials are based on an L2 rate (98% reliability). With critical aerospace gear drives, it is common practice to design the gears such that the probability of a fracture failure at the root is extremely remote.

I would agree with israelkk's comments about using 17-4PH in an H900 condition for aerospace applications. As a general rule, 17-4PH is not used in an H900 condition due to stress corrosion concerns. If you require a high-strength corrosion resistant steel for your gears, I would recommend Custom 455 H950 or PH13-8Mo H1000. Both of these alloys have good fracture toughness and stress corrosion resistance at strength levels above 200ksi.

Good luck to you.
Terry

 

[gearcutter]

If you intend on making this type of work your carrier path; then I would highly recommend that, at the very least, you purchase the following standard from the AGMA's website;

AGMA 917-B97 - Design Manual for Parallel Shaft Fine-Pitch Gearing



Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[gearcutter]

Woops...........that should have read:
If you intend on making this type of work your *career* path.

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[gearcutter]

I would also recommend that you get hold of an excellent paper, available from the ASME's website, by T. Walker titled 'A Rating Formula for Fine Pitch Boundary Lubricated Gears'.

Abstract,
For the Instrument Engineer involved in the design of mechanisms that transmit power under boundary lubricated conditions, little information is available on which to base fine pitch gear load capacity and life. This paper discusses a gear test program and the development of a rating formula for the surface loading of these gears, and in particular those made from stainless steel and aluminum and stocked by precision gear manufacturers.


Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[cfordyce05]

I've now read the majority of AGMA 917-B97. That filled in quite a few holes. I need to check out the AGMA standard for getting my J geometry factor. I am also going to need to pay attention to my addendum modifiers. It looks like I'll have to add a bit to my pinions with my current tooth counts. Did I read it correctly that if I want to keep a standard center distance, then if I add to the pinion I just need to subtract from the gear? Is it better to just leave the gear "stock" and modify the center distance so I don't loose any strength on the gear teeth?

I'll purchase the ASME paper as well and start pouring through it.

 

[gearcutter]

In answer to most of your questions; it all depends.
Is your set going to be speed increasing or decreasing?

For speed decreasing; you're always better off to add to the pinion.
In reality there is no such thing as a standard centre distance.
Remember that addendum modifications are not only about strength; there are many, many more factors to consider.

The single largest issue that I've encountered with fine pitched gearing is lubrication. You'll most probably find that a simple 'splash' type system will not be adequate.

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[tbuelna]

cfordyce05-

Your 14t/35t combination would benefit from some addendum modification for several reasons.

First, ideally you want to balance the tooth bending strength between the pinion and gear to provide equal fatigue life. Besides using addendum mods, you can also make the face width of the (weaker) pinion wider, which helps a bit. Second, with a 14t pinion, you will need to use some profile shift if you want to reduce/eliminate undercutting of the tooth flanks. Third, judicious application of profile mods on both the pinion and gear will improve efficiency by balancing contact sliding. I believe the AGMA document recommended by gearcutter provides analytical methods for optimizing your gear designs for bending, undercutting, sliding, etc. But while the methods described will give accurate results, they also require a fair amount of effort to work through.

Good luck to you.
Terry

 

[cfordyce05]

Do making those profile mods require a custom cutting tool? The addendum mods are just changing how "deep" the cutter goes into the blank and not the profile shape, right?

 

[gearcutter]

Do making those profile mods require a custom cutting tool?

It depends. Which mods are you referring to?

The addendum mods are just changing how "deep" the cutter goes into the blank and not the profile shape, right?

For addendum corrections; no special cutter is required.
Note that it is the tip diameter that changes, not the whole tooth depth.

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[tbuelna]

Do making those profile mods require a custom cutting tool?

The requirements listed in your earlier posts would appear to demand fairly high quality and well optimized gear geometries. As gearcutter noted, modest addendum corrections do not typically require custom tooling. But with your 14T/35T combination, the gears will likely require some tip relief.

In reality, since your gear design requires high-strength through hardened materials and uses an extremely small gear size (80 to 96 DP), they likely must be form ground using CBN, or hard shaped/skived using carbide. So in your case there is no such thing as "standard tooling".

Look at the bright side. These gears will end up costing a few bucks to produce, but there are lots of gear houses that just love tackling these types of challenging jobs. So you're sure to make some gear cutter very happy with this order.

Best of luck to you.
Terry

 

[israelkk]

As I recall DP finer than DP48 can not be grinded.

 

[tbuelna]

israelkk-

You are probably correct. I took a look at the capabilities of some existing machines that use CBN coated mandrels to form grind spur gears, and I did not see any that had published capabilities to form grind a 80-96DP tooth.

I think the 80-96DP spur tooth could still be hard shaped or skived though.

 

[mfgenggear]

It can be form grind from solid. it would be cheaper to hob.