Gear drives and design

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
172
It seems that the sticking point for amateur built gear drives primarily includes TV "black art" (ignorance) at the theory/development level; and hardening/distortion/post finishing at the practical level.

Let's ignore the first for now to see how feasible some approaches to the second are.

Steel gears for automotive apps are hardened for 2 reasons:
1.) smaller gears with stronger sections & teeth than non-hardened equivalents
2.) wear resistance.

c955 aluminum bronze is as strong (tensile) as a high strength gear steel such as SCM415 or, say 4140. 95Ksi tensile.
It is used sometimes for gears, or more specifially wormwheels for higher stress applications (lots of shear and tooth face sliding pressure) in combination with a hardened steel worm where c954 or other lesser materials fail.

Hardened steel spur or bevel spur gears from automotive apps are widely available.
Strictly from the materials standpoint, what are the limitations & considerations to running a hardened steel spur gear in a 954 or 955 aluminum bronze ring gear for power transmission? (Think Marcotte.) Assume designed for the app in terms of tooth strength & engagement: how will it fail, and how fast compared to 2 hardened steel parts? How much larger (%) or not would the components be? Was the ring gear for Marcotte thermally hardened? As opposed to say, being machined from a prehard (4140 or 4142 prehard, or ETD150, e.g.? ) & perhaps only stress relieved? Nitrided? or?

Alternately, ETD 150 at 150ksi tensile would offer huge strength increase, still machinable, but without further hardening (nitirding?), how would it fare in terms of wear, running with a hardened steel pinion?

There are materials for which cost is the primary inhibition to commodity use for apps in which they *might* be suitable.
So the quesion is, sort of, **** the raw materials costs, are there any likely that could yield as "good" (reliable) a 2 gear mating system in which the driven component could be machined & put in service without further hardening, other than perhaps thermal stress relief? Or is that quixotic on the face of it?
Obviously (i hope) the assumption includes "enclosure, & provision for lubrication typical to the app"

Thanks!
smt
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
4,383
Location
Calgary, Alberta, Canada
Marcotte gears 4340 hardened to 62RC. 450 hours on mine, never been touched internally, same backlash as when new.

Never seen bronze gears used for power transmission in any auto or aircraft gearbox. C955 doesn't have near the tensile, yield or shear strength nor the hardness of 4340.

I'd guess that it's also more expensive.
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
172
Do you know how the gears were heat treated? Induction hardened teeth?
Any info on post heat-treat finishing? Grinding? Polishing?

Good catch on the strengths-
I use 954 & 4140 prehard for smaller parts and sometimes replacement gears for machine tools (obviously low stress apps and lesser form control than necessary for automotive) & have machined c955. Looking up the tensile & yield values they are farther apart than i was thinking. & both less than 4340. CDA955 is some multiples in cost over 4140/4140 prehard. I've not used 4340.

However, if a hardened bronze/hardened steel pair would be "durable" in terms of wear, it could be sized for the loads & the question becomes how much size/weight penalty.

Or the simple solution: :) Any insights why Marcotte is no longer available?

Thanks!
smt
 

WonderousMountain

Well-Known Member
Joined
Apr 10, 2010
Messages
2,296
Location
Clatsop, Or
Copper Alloys, and Nickel ones might make good gears.
There are high strength Bronzes available as Rod/Tube.

However, my suggestion for small gears is tungston.

Copper alloys make great bearings, Nickel alloys take
high temperatures well. Marcotte gears look good, it
has a lot of tooth width and some angle to engage well.

Cutting Tungston probably requires an expensive tool.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,053
Location
Saline Michigan
First off, power transmission gear are made of heat treated steel because you can make the lightest gearsets that will last just about forever out of them. Fatigue strengths are high, surface fatigue resistance is high, gear tooth manufacture and heat treat are well understood - we understand gears made of steel. If you want to build spur gear sets with weight and durability as important issues, either straight spur or helical spur gears, medium carbon alloy steels are it. AGMA has a manual on gear design, and it is the bible for folks who do gears. It also gets into details on other gear materials...

Now to clarify, power transmission gears do not rub. The contact between involute gear teeth is one curved gear tooth rolling on another curved gear tooth. Very similar to rolling element bearings. Wear on gear teeth with rolling contact is not abrasive, but surface fatigue that eventually starts breaking particles out of the surfaces, called pitting. So the deal becomes to make the surface the right shape, texture, and hard, all for fatigue resistance. Now if one were to design hypoid or worm gear systems and the like, that is where one of the gears can sometimes become bronze. You have sliding friction, and the game gets a lot stranger when you have hypoid or worm gear sets. Efficiency usually goes way down too when you get into sliding friction. But if you want it to self lock wherever you cut the power, worms are it.

Billski
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
If you can choose the right gears,mount them at exactly the right distance and keep everything in plane
and then verifiy the work is right,then hey anybody who works to .0001" can do what they want,right?
I might make a 90° 1 to 1 drive but
thats it.
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
4,383
Location
Calgary, Alberta, Canada
Guy Marcotte is in his '70s now and semi retired but still building a few parts and drives for customers on 6 month wait times.

A fellow from California with CNC capability was interested in perhaps making these gearboxes if Guy would sell the rights. Not sure if that discussion is progressing or not but I'll drop an email and see.

I don't have insight into the heat treating process used but guess it would not be nitriding on gear sets.

Here are a couple photos of the M300 gearsets:

mar1.jpg mar2.jpg

The ring gear is not an easy piece to machine and keep accurate after heat treating. Whoever makes these for Marcotte knows a thing or two about this stuff.

There a few things which could be improved as far as sealing goes. I've made some external mods to mine as has Russell Sherwood who has over 800 hours on his but the gear and bearing setup has proven pretty sound at the 150-230 hp level. The prop shaft is supported by 2 huge Timken tapered roller bearings.

This is really the best gear layout IMO for this application. It's quiet, strong with excellent gear contact, retains lubrication really well and won't easily seize if something goes wrong as the ring gear will grow and increase clearances if it gets hot. The gear offset is also helpful in lowering the engine for more cowling clearance up top. This is often an issue with planetary gearboxes.
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
172
Thanks, Billski - I was specifically hoping you would chime in.
It sounds like there is no hope without full heat treat process on appropriate steel parts.

:(

RV6- Thanks much for the extra info - those are impressive pieces!
My interest is considerably lower HP - in the 50 - 90HP range. For some reason i though Marcotte made sets for smaller engines like the 4cyl Subaru (also not my interest, but in that range)?

I don't have insight into the heat treating process used but guess it would not be nitriding on gear sets.
This is somewhat interesting, in that he is using 4340? (for which nitriding could be a typical hardening process?)
As opposed to, say one of the lower carbon gear steels & carburising, or one of the higher carbon and induction hardening?

Probably wrong, I've become superannuated; but cnc won't be much use for making gears themselves unless things have really changed.
Someone is running a gear shaper for those parts.
Or happy to be educated further. :)
This is interesting.
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
4,383
Location
Calgary, Alberta, Canada
Marcotte used to make a 150 model which I've seen and they are really tiny plus a 200 model which is much smaller than my 300. He may still be able to make the 200, not sure. I don't think the 150 has been produced for at least 15 years.

The 4340 is much tougher than low carbon steel and maybe that billet material is more consistent as well being commonly used for things like con rods and cranks.

The output gear would be hard to induction harden without a custom rig I'd think but furnace heat treating might result in a lot of distortion. I'd be interested to see how these are actually produced! I think a 4 or 5 axis mill could cut the output gear profile these days but a shaper would be faster and require less lapping. If I recall, these gearsets were lapped before final assembly.

It would be fascinating to talk to Guy about how these were made.
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
172
4340 is much tougher than low carbon steel
This is not quite true, but there are too many factors to consider (level of heat treat/hardness, and the process of heat treating to get there) vs carburizing the lower carbon steels, how deep, where, and quenching. Level and type of temper for each. It is a deep field. But just as an example, 8620 or 4320 might be used for gears and parts specifically for toughness while 20 points carbon would not easily/practically harden. So it is carburized and case hardened. 4340 might be nitrided. It could be quenched, though not to 60's Rc. 4350 starts to have significant carbon, and might be induction hardened.

Depending on application, there are various ways to do most steps in the process to attain specific final results.
I heat treat simple common tool steels like S7, A2, CPM M4 for blades, sometimes O1 for quick cutters or parts, so have studied it off & on "as necessary to the task at hand" and can rapidly get in well out of my depth. :) I am unfamiliar personally with the types of steel usually used for gears. It is too late in the day for me to explore that field other than out of curiosity. Though there is a good local industrial heat treater I have used for other things. Several of the local industries used to make gears and harden them in house. Niles Crane, and Bendix come to mind as auctions where multiples of tooled up Fellers went for scrap. Like most processes, it gets ever more specialized. Both technology and economics force consolidation
 
Last edited:

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
279
Do you know how the gears were heat treated? Induction hardened teeth?
Any info on post heat-treat finishing? Grinding? Polishing?

Good catch on the strengths-
I use 954 & 4140 prehard for smaller parts and sometimes replacement gears for machine tools (obviously low stress apps and lesser form control than necessary for automotive) & have machined c955. Looking up the tensile & yield values they are farther apart than i was thinking. & both less than 4340. CDA955 is some multiples in cost over 4140/4140 prehard. I've not used 4340.

However, if a hardened bronze/hardened steel pair would be "durable" in terms of wear, it could be sized for the loads & the question becomes how much size/weight penalty.

Or the simple solution: :) Any insights why Marcotte is no longer available?

Thanks!
smt
I emailed Marcotte a few months ago and he was just getting a batch ready for when the castings came back. Has something changed?
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
The dimensional changes I have seen on 4130
plain bushings when hardening are .007+.
Each alloy is going to change diffently to each
heat treat process,and that change is large enough that it must be accounted for.
My Grandfather worked the tempering ovens at
the Alegany National Forge.Big stuff.And the temps were kept to a degree or two when operating over 1500°.And the charts they published were fake,you had to know the conversion factor.
I think that the whole subject falls under the heading of " if you have to ask,you wont understand"
Pure black magic
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
4,383
Location
Calgary, Alberta, Canada
For highly stressed engine parts, low carbon steel is almost never used. 4340 is the steel of choice here in North America. If you machine or cut 4340, I think you'll agree it's a lot tougher than low carbon steels. I'd be using the best material available if I was going to the trouble of machining something like the Marcotte output gear. The material is well proven here to be durable. It's obviously possible to get 4340 to 62C through some hardening process.

As far as Marcotte making parts and GBs now, I believe there has been renewed interest in his units over the last 3-4 years due in part to the success I've had and Russell Sherwood in particular racing in SARL. He wasn't making much 5-8 years ago and told me he was scaling back on production. I know he has produced some custom parts for Russell and two other people I know so he still has interest which is great. I hope he continues and/or transfers the design to someone who can produce them in a quicker fashion. It would be a shame to see these disappear when Guy fully retires.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,053
Location
Saline Michigan
Big things to know about alloying steel:
  • Austinize (soak above upper critical temperature) then quench (fast enough cooling) makes primary martensite, which is glass hard and almost a brittle, and is about the same no matter what the carbon and other alloying element content;
  • Tempering primary martensite at temperatures of 250F and above convert it into tempered martensite. At 250 F, that barely reduces the strength but increases toughness, higher temps increase ductility but you trade that for strength;
  • Raising carbon from 0.1 to 0.4% improves strength and toughness in annealed and in quenched and tempered steels;
  • Going from plain carbon steel to low and then mid and then high alloy steel does several things
    • Alloy raises the annealed strength a lot;
    • Alloy allows strengthening/hardening to go much deeper into the material, and;
    • Alloy reduces severity of quench required;
  • Resistance to abrasive wear is obtained with medium to high carbon content, plenty of alloy content and austinizing just above lower critical temperature, then quench and temper. This forms carbides of iron, chrome, molybdenum, etc in a matrix of tempered martensite. This is done only rarely in gears,and then usually in "dirty" environments.
Things to know about hardening and its effects upon gears:
  • All hardening can change shape of part from before hardening;
  • The depth of the hard material on a gear tooth surface dictates it fatigue strength in both bending (against tooth breakage) and against tooth pitting;
  • Hardening of teeth can be done a bunch of ways, each with its pros and cons:
  • Quench and temper:
    • In alloy steels this will harden and strengthen the entire gear - You may or may not want this;
    • In plain carbon steels, this will have limited depth - you may or may not want this;
  • Surface treatments like carburizing and nitriding add material and volume to a generally thin layer of the steel - which results in small but real dimensional changes;
Each has its uses. Just like all the variables in airframe structural design, gearboxes have their whole range and uses. This is a really deep and convoluted rabbit hole.

Historically, gears are overdesigned hoping that mild processing will be suitable. This let's every set their packaging, and design all the rest of the parts. If anything does not work out, the design and manufacturing guys get into higher level materials, more and higher level processes, and then into increasing acts of desperation to meet goals as they build and test and find shortfalls. If you see really high level material, processes, etc, they probably started out undersized on critical dimensions.

But if you start out with low to medium level material and processes, design to be suitable with those, accept a little more weight and volume (and it does not add much) and then find a life issue, you can upgrade material, heat treat, surface engineering, etc and get to durability and noise targets.

If your timeline allows it, but you really need to get weight to absolute minimums (you sending this thing to Mercury?), you can design smallest and the next couple steps up, then do the programs in parallel, and pick the lightest one that meets other goals. This does come with the downside of delaying release of the packaging for everyone else in the program...

Billski
 

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
172
My question has been answered:
There is no practical substitute for steel, nor would a steel such as ETD150 work, that would not need post-machining heat treat, that should practically be considered for a ring gear for IC engine power transmission in a critical app.

AFA heat treatment and machining of steels, there is some misunderstanding about the concepts of hardness, toughness, and wear resistance (which Billski has addressed nicely). Hardness is simply a proxy for tensile strength. Alloys are developed/chosen for wear resistance in a given application/environment. Toughness varies with hardness and with the actual heat treat process. Toughness is measured by the energy value of an impact test (Charpy, Izod), the test sample values can be modified by the notch geometry depending what the application or designers interest might be. (You can't blanket say "x steel is tougher than Y steel")

Blades/cutters & gear teeth are crudely analogous: you pick a hardness where an edge of the blade (gear tooth) with an acuteness angle (tooth DP) appropriate to the substrate (loads) does not bend in application. Then try to modify (through the heat treatment process) to eliminate chippiness by studying the chart charpy values at given quench & temper temperatures and methods (the "recipe" for how to cook it). Gears are more likely to include heat treat processes that yield a softer core for toughness with a suitable hardness gradient to the surface, which surface is very hard for impact and deformation resistance. Gears are also more likely to use medium carbon steels (lower carbon to edge toolmakers :) ) and may sometimes be provided surface hardness through surface treatment including carburizing (add carbon only where needed, for hardness at the surface) , nitriding, and combinations of same. Wear resistance is largely an attribute of additives (alloying elements) though in blades, some comes from the intended formation of carbides & their size & distribution in the matrix, Billski mentioned. In many products, other surface treatments can enhance wear resistance where the source of the wear is understood. For blades, failure in many substrates is through chemical erosion, not abrasive wear.

That is enough for here.
My specific question about material options has been answered.
I've done enough small scale heat treatment, including high speed steels, & have made items like "unobtanium" low speed ball bearing races, to know where to look for answers in that area. Further, i am unlikely to apply it to gears for critical apps. If necessary, there's a very good local heat treater who has processed commerical products for me.

Disappointed there is no expensive lunch (material) that would give a free ride past the heat treatment issue.
But thanks!
:)

smt
 
Last edited:

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
172
I emailed Marcotte a few months ago and he was just getting a batch ready for when the castings came back. Has something changed?
I initially searched online, possibly 6 months ago, and could not find any current information or contact.
Various notes on various posts on this forum indicated others could not reach him, including some (I think it was this forum but may have been one of the searches) who had not received parts nor communication over an extended period. If he is only working for select individuals, that is as good as being unavailable as far as most of us are concerned.

If there is contact information or a website, the information would be welcome.

Thanks!
smt
 

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
279
I initially searched online, possibly 6 months ago, and could not find any current information or contact.
Various notes on various posts on this forum indicated others could not reach him, including some (I think it was this forum but may have been one of the searches) who had not received parts nor communication over an extended period. If he is only working for select individuals, that is as good as being unavailable as far as most of us are concerned.

If there is contact information or a website, the information would be welcome.

Thanks!
smt
It appears his website has been taken down? But the information is still available here-

His email to me indicated he was making a run ofM-300 Mazda pattern gearboxes beginning in May 2021.
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
4,383
Location
Calgary, Alberta, Canada
When it comes to complex helical gear production as on the Marcotte drive, the machining and heat treating processes are really critical. Heat treating almost always involves some post distortion which has to be dealt with in some way. At 62C, mostly likely you are talking finish grinding, which is really specialized for these shapes. Perhaps if distortion is really low, lapping will suffice. I believe Guy was using a US vendor in the east to do the gears. Anyway, they clearly know how to do it successfully. It's certainly beyond what I can do around here.

As far as contact, Guy has always been a bit hard to contact and I've been dealing with him for over 20 years. He had a sales partner early on and then later Ray Fiset took care of that side but sadly passed quite a number of years ago. This part makes it frustrating sometimes and with Guy's age, future long term support could be a concern. Guy has always been honest with me and treated me well. Deliveries have always been slow but he'll generally give you a pretty accurate wait time. There has been no website for at least 10 years to my knowledge unfortunately.
 
Last edited:

Aviacs

Well-Known Member
Joined
Oct 21, 2019
Messages
172
I believe Guy was using a US vendor in the east to do the gears
You probably couldn't realize how valuable that tidbit is.
If it's not who i think it might be, (based partly on geography, partly on skill set) they would know.

Thanks!
smt
 
Last edited:
Top