All-metal versus all-steel, what gives?

by | |

As usual, an askJOHN comes about due to a question. But this time it's one of my own devising because I wanted to respond to a pal at my flying field posting about his innovative use of a low profile standard class servo instead of a mini-class. This, in a model (a 69" wingspan Extreme Flight Turbo Raven) in which most guys use a super mini like our DS160CLHV as the good choice, the D205BLHV as the better choice, or maybe the incredible DS355CLHV mini as the best choice from amongst the various mini-class servos, which we produce.

Basically, the harder a guy flies, the more he want lusts for in the way of power form his servos. Speaking of the DS355CLHV, specifically, this is a so-called super-mini because it's capable even in a substantially larger model like an 85" Extreme Flight Extra like this making it the servo of choice amongst the thinkers in the pilot world.

As for why Hinson used the low profile? Dunno but sometimes in life you use what you have on hand even if you're a pal. In this case a set of MKS's magnificent 599HBL low profile servos outputting 320oz-in even though they're ~$160 a pop. Their servo at 320oz-in, 0.09sec/60° (MKS model number 599SL(, which Hinson is using are a great standard-class (40x20mm) low-profile servo.

Me? I've been doodling a low-profile standard with which to compete - but - I can't touch their 53.5g because I use all-steel gears instead of all-metal. So what do I mean about all-metal versus all-steel? Which is better? And why does it matter? Read on!

- Our prototype low-profile is presently outputting 340oz-in but this isn't set in stone

Allow me toss in a bit of EDU. Regarding gears, have you ever seen one of those laser cut educational clock kits where everything is made of wood? Make the gear teeth large enough (to make up for wood being weak) and you can drive impressive loads. No, not for long because wood is still wood (so you'd need to really lubricate them), but real work could be done! Heck, you could use cheese to make gears if the teeth are really large (larger is a proxy for more strength).

Point being, all-metal in servo-speak means pinions are steel (pinions are the small part of the gear and they're almost always steel). Other gears in a transmission, larger ones worked on via smaller pinions, are called bull gears (the large part of a gear in a servo, the two - pinion and bull - are pressed or machined together). So the steel pinion is usually pressed in the bull. The bull can be made of aluminum, brass, titanium, steel . . . cheese. Or the whole thing can be molded of plastic!

Anyway, the bull gear material is typically softer material because it's cheaper (except Ti). This works because the usual gear ratio doesn't exceed ~6:1 (means there's 6X less pressure acting on the whole thing). Really means as engineers we know that for some applications you can get away with using a softer material (reference the wood gear clock kit and sport servos using plastic gears).

And note, this is all much more complicated than I'm making it - but this explanation is close enough for something that's not involving much math. So moving on, these are the gears out of some of our servos. At ProModeler, the smaller gear in our lineup is always steel. So the small gear is pressed into an aluminum, brass, or steel bull gear (and sometimes titanium). In this case the ones on my finger are a brass and a steel bull. And always when it comes to gears, the bigger (can be thicker, same thing), the the better!


- Steel pinion pressed into a brass bull and another steel pinion pressed to a steel bull

Summarizing, aluminum is a good material for low pressure gears and while 7075 isn't as hard as steel, it's harder than 6061 so that's what's used. And while Al is more costly than steel, it's much easier to machine. And because it can have a wear coating, it's actually pretty good material for some applications. Servos are one of these applications. In fact, brass (harder than aluminum) is also a good materials for low pressure gears (witness their use in clocks, too).

Also the typical output, or final gear is 7075 aluminum to save weight (and more weight with Ti). In the case of aluminum, it's usually a T6 temper (for toughness). Also, since effectively, the output gear is a bull gear, this output gear can be softer than the pinion driving it, and live a pretty good while. Proof being so many servos in the hobby market being made of plastic, aluminum, and in some cases titanium!

So summing up all-metal gear trains are usually composed of materials like aluminum, brass, (sometimes Ti), plus steel . . . and all four of these are metals, hence, all-metal terminology. And all-metal is usually stronger than a hybrid gear train where some of the gears are made of plastic (for example what Hitec calls theirs, Karbonite). Added to which, this is true for all brands - us, Futaba, MKS, Hitec, etc. Also, despite other materials being softer than plastic, because if you make them larger enough you could use gears made of wood (or even soft cheddar cheese) and use them to turn the world 'if' they're large enough!

The greater point being material choice is not the sole determining factor because other factors plays a role alongside cost! In engineering, harder, weight, size, durability (duty cycle), plus cost are amongst the factors when you think about with respect to gears. Focusing on weight alone makes you one-dimensional, e.g. not a good engineer.

Unfortunately, the fly in the ointment for aluminum and titanium is they're soft. Means they'll wear quickly, Much more quickly than steel. Steel, like titanium, is also hard to machine (turn and broach). Fortunately, costs less. And note, even 7075 (a lot harder than the 6061 we use to make cases) is still softer than steel by a mile. Ditto Ti, as both it and aluminum are softer (alongside brass). For the lightweight materials, the answer for durability is a wear coating (do do this with brass), and with steel it's another process altogether but again, I digress.

With aluminum hardening the surface means they're hard anodized. Another word for anodizing is oxidizing, like rust on steel, but different. And note; a few years back, there was a fashion for using titanium instead of aluminum for the output gear( but it's falling out of fashion as modelers catch up to industrial practices). Industrial is my way of using a word to stand in for industrial or heavy duty use (modeler's use is intermittent). However, note; even steel isn't indestructible. This next photo is an example of a modeler exceeding the ultimate strength of a steel output shaft.


- Steel 25T spline twisted like a pretzel - usual solution is bigger!

So as I've explained both Ti and Al are very soft metals. But even the strength of steel can be exceeded. But it's rare, takes high loads. The solution is bigger. And bigger can apply to more than the output spline and to the gear s faces themselves. Made them longer and you reduce the pressure loading (psi again). So back to gears in the typical servo, for some gears, the duty cycle is so extreme the solution is larger faces, larger shafts, hardpoints in the soft materials. This photo shows the series 1 versus series 2 brushless servo gear trains. If I decide to make it, our low profile will involve a series 2 gear train.


- Even with steel, surface treatments exist - note the difference in color!

Fortunately, for gear use they (Ti and Al) can be surface hardened with a coating. BUT, when the wear coating goes (when not if because the laws of physics aren't repealed for MKS or Hitec) then because gears work by sliding the face of one tooth across another to lever the tooth of the other gear (which becomes rotation), and because any two materials sliding against each other eventually wear, then the underlying gear train fails.


- All involute gears (even helical, which are worse) wear due to sliding

So eventually is a factor of time, so it's gonna happen. When is the exact question and the harder the load (measured in psi), the quicker it happens. Duh! Major point being, gears wear because they slide against each other during operation and they are always going to wear.

However, coatings help prolong the process but overall, the process process can't be prevented. What's a good engineer do? He considers the cost, weight, benefit tradeoff and selects a harder material over a softer one. In general, steels is harder than both titanium, brass, or aluminum by a country mile. Tougher too. Bottom line? Steel gears don't break as easily as ones made of other metals. MKS have good engineers. They know this. But there's another factor (this is a blind judgement because NOBODY at MKS or Hitec, of Futaba tell me anything - not even the time of day).

Back to my point, when the gear wear accelerates (meaning after the anodized coating is consumed) then in a blink of an eye the gear itself wears because the underlying material is very soft. I have a video somewhere on my phone where I took a piece of 6061 and lightly hit it with a sanding block. Rounds it right off! Does the exact same with titanium. Rounds the edge beautifully because both are nothing but soft shiny wood in comparison to steel. The steel you won't round off in a similar fashion. Eventually? Yes, of course, everything wears, but steel takes much longer and much more pressure.

So steel is way the heck harder than aluminum or titanium. Better for gears. This by the way brings us to another thing. Titanium is expensive. Lighter than aluminum, and harder, but nowhere near as hard as steel and 10X the cost and harder to machine. Most focus on cost, I focus on harder to machine. Means it takes more time and the tooling costs more. All factored in, of course. But yes, Ti is stupid expensive. Some things Ti's light weight means cost doesn't matter, think SR-71 Blackbird where money was of no concern.

Anyway, a good engineer does the math and NEVER uses titanium or aluminum for gears that'll wear if they also have to last. If they're for a one and done product, especially if weight is a concern (like it is for a servo), then there are perfectly valid reasons why all-metal gear are used instead of all-steel (in fact, we use these with some ProModeler servos). But for the big boys? Nope, we turn to all-steel. And closing the circle, there's no free lunch and steel weighs more.

All this by way of saying, for high load applications (and for lower load but you'd like the gear train to last forever and a week, then ProModeler concludes (our professional engineering opinion) that for big loads and long life, because titanium and aluminum (and brass when it comes to it) are a downright terrible material for gears. It's because all are comparatively softer than steel, and the first two require a wear coating (expensive), and in the case of Ti it's stupid hard to broach and 10X the cost. Pointless, except for the marketing department.

Marketing? Yup, if there's one thing marketing really knows is this; carbon fiber and titanium are words that make a modeler (and men in general) reflexively reach for their wallets. fast, like faster than slipping a few bucks to a pole dancer's lower when it's your turn in a titty bar . . . but I digress. Point being, those words, carbon fiber and titanium, equal sales and all else being equal, more sales are mo betta'.

Anyway, ProModeler doesn't have a marketing department and because if Ti, Al, and brass were better for high performance gears then Ferrari would use it in their money-is-no-object F1 cars. They don't, so it isn't - the 30,000' view. They use steel, we do too.

So a servo with an all-steel gear train means we use steel for pinions (little), bulls (big), and final gear. Bulls and pinions are pressed together final gears are machined in one go because their output is typically a spline or keyway. For consumer's servos, it's usually a spline. Fly in the ointment is steel weighs more. Is what it is.

Meanwhile, I haven't decided whether to even make a low profile, or not because the prototype weighs 69g versus 53.5g for the MKS. Of course, instead of 320oz-in mine is making 415oz-in, so it depends how much you value the 15g saving. Major point I wish I had a low-profile in my line up with which I compete head to head, but I don't because this is just a prototype and until I poop or get off the pot and make it, we don't offer a production low profile. That said, if weight is a concern, then as mentioned previously, folks can look into our DS355 mini, which is more powerful, faster, and lighter, and because it's equipped with all-steel gears, will last longer between refreshes.

Finally, using the low profile standard is an innovative approach and since Hinson flies at my club I'll keep my ear to the ground to see how this works out in the real world. I'd expect it works out very well because they guy's not exactly stupid.

In closing, Hinson, if your deal with MKS ever falls apart, eyeball this servo if I haven't produced the lowDS355CLHV> profile.

This entry was posted in no categories.