Benefits of a two stage transmission?

[ian.newson]

Hi, I've been thinking about this for a while, and my googling has been ineffective, but what are the benefits of a two stage power system, as we see in goblins?

Intuitively it seems there must be a loss of efficiency. I also expect they can handle higher power loads, but clearly a single stage is strong enough.

Anyone got any pointers?

[Raz0rSh4rp]

In case of speed you get more surface area for power transfer but most importantly it allows you to make the helicopter much slimmer. This doesnt work with Goblins due to servo layout.

[AntDX316]

Originally Posted by Raz0rSh4rp

In case of speed you get more surface area for power transfer but most importantly it allows you to make the helicopter much slimmer. This doesnt work with Goblins due to servo layout.

Have to engineer a custom frame. Be the first to 3D print a working design then cnc the gearing.

[Xrayted]

Originally Posted by Raz0rSh4rp

In case of speed you get more surface area for power transfer but most importantly it allows you to make the helicopter much slimmer. This doesnt work with Goblins due to servo layout.

This is also my understanding as well. It opens up the door for a larger variety of gearing choices without having to make a main gear as big as a saucer plate hanging out the sides of the heli to achieve the same ratios.

My Goblin has an "effective" main gear size of 204T which would be a pretty big gear, but as stated the wide servo layout forces the canopy to be blown out wider to clear them.

[ian.newson]

I'm not convinced by the power transfer argument since surely you can just increase the surface available in a single stage, but the thinner layout argument makes perfect sense, and I guess is the main reason goblins have the look they do.

Thank you!

[ian.newson]

Originally Posted by Xrayted

This is also my understanding as well. It opens up the door for a larger variety of gearing choices without having to make a main gear as big as a saucer plate hanging out the sides of the heli to achieve the same ratios.

My Goblin has an "effective" main gear size of 204T which would be a pretty big gear, but as stated the wide servo layout forces the canopy to be blown out wider to clear them.

The gearing argument makes sense too. Thank you.

[extrapilot]

Ian

Mostly, this is a function of power vs deformation at the main gear.

At high power levels, a conventional system sees high tangential and radial forces (and axial, if a helical pinion). Just a random data point- running a 10hp peak @ 20KRPM on a 0.7” pinion (20deg pressure angle, 30deg helical angle), you get about 16lbf separating force, and 26lbf axial- both of which can/will deform the main gear. That screws up the mesh, so you end up with loss of efficiency, heating, gear damage, etc. Belts make all that easier, and setting mesh with a pinion change is trivial also. Doesn’t mean it cant work- the X7 is a good example- but the 425 was a disaster at higher power, so maybe Gaui has opted to just scale the X5 up and deal with some limitations on peak power, or at least with gear longevity at high power.

I think it is fair to say these machines are designed by artists and not in wind tunnels. None of these guys seem to understand what impact a change in canopy width has for a given mode of flight. If they did, they would not do some of the very stupid things they do. So, maybe they hope this helps a bit- but that is probably the extent of that argument-

Cheers

[ian.newson]

Originally Posted by extrapilot

Ian

Mostly, this is a function of power vs deformation at the main gear.

At high power levels, a conventional system sees high tangential and radial forces (and axial, if a helical pinion). Just a random data point- running a 10hp peak @ 20KRPM on a 0.7” pinion (20deg pressure angle, 30deg helical angle), you get about 16lbf separating force, and 26lbf axial- both of which can/will deform the main gear. That screws up the mesh, so you end up with loss of efficiency, heating, gear damage, etc. Belts make all that easier, and setting mesh with a pinion change is trivial also. Doesn’t mean it cant work- the X7 is a good example- but the 425 was a disaster at higher power, so maybe Gaui has opted to just scale the X5 up and deal with some limitations on peak power, or at least with gear longevity at high power.

I think it is fair to say these machines are designed by artists and not in wind tunnels. None of these guys seem to understand what impact a change in canopy width has for a given mode of flight. If they did, they would not do some of the very stupid things they do. So, maybe they hope this helps a bit- but that is probably the extent of that argument-

Cheers

Thanks, I hadn't considered gear deformation.

It seems there must be an inherent loss of efficiency with a two stage power system simply because there's more components. I wonder how that compares with the loss of efficiency with a deformed main gear. I'm guessing the latter must eventually cause more loss otherwise you wouldn't have said it.

[ian.newson]

Originally Posted by extrapilot

Ian

Mostly, this is a function of power vs deformation at the main gear.

At high power levels, a conventional system sees high tangential and radial forces (and axial, if a helical pinion). Just a random data point- running a 10hp peak @ 20KRPM on a 0.7” pinion (20deg pressure angle, 30deg helical angle), you get about 16lbf separating force, and 26lbf axial- both of which can/will deform the main gear. That screws up the mesh, so you end up with loss of efficiency, heating, gear damage, etc. Belts make all that easier, and setting mesh with a pinion change is trivial also. Doesn’t mean it cant work- the X7 is a good example- but the 425 was a disaster at higher power, so maybe Gaui has opted to just scale the X5 up and deal with some limitations on peak power, or at least with gear longevity at high power.

I think it is fair to say these machines are designed by artists and not in wind tunnels. None of these guys seem to understand what impact a change in canopy width has for a given mode of flight. If they did, they would not do some of the very stupid things they do. So, maybe they hope this helps a bit- but that is probably the extent of that argument-

Cheers

Just occurred to me, it seems like you've answered a different question?

I was asking about single versus two stage power transmissions systems, where you seem to be talking about belt vs non belt power transmission systems.

It seems like the majority of two stage systems use a belt, but belts aren't exclusive to two stage, e.g. the Protos.

I don't know of any but I can't see any reason why there couldn't be a two stage power system that doesn't involve belts.

[extrapilot]

Hi Ian

Well, there was at least one 2-stage geared system in the past 5 years- the Gaui 425- and it was a disaster at higher power. The only fix was to move to carbon frames and metal cross-bracing to prevent the mesh errors- which could melt the driven intermediate gear in seconds at higher power levels absent these mods.

The R5 is a similar layout, except with the first interface replaced with a belt. I don’t expect you will see a repeat of the geared design by any major manufacturer.

From an efficiency perspective, for a properly-sized setup, a belt has similar efficiency to a gear, but it has much higher tolerance to things like frame deformation. And because you can alter the belt length, you have much more flexibility with gear ratios- where you don’t need to alter the driven pulley diameter etc.

People lose more to inefficient blades and higher IR packs. When you start looking at 5-10% variance in motor efficiency and the huge inefficiency of the rotor design in general, it is pretty high up on the tree of low hanging fruit.

Regards

[ian.newson]

Thanks. It seems like efficiency isn't very high on any airframe designers list of concerns.

[desertstalker]

Originally Posted by ian.newson

Thanks. It seems like efficiency isn't very high on any airframe designers list of concerns.

because a heli is really inefficient anyway, you might be able to save an extra 1% in the drive-train, but who will notice?

[ian.newson]

Originally Posted by desertstalker

because a heli is really inefficient anyway, you might be able to save an extra 1% in the drive-train, but who will notice?

Yes!

[TucsonFlyer]

I think the real benefit of the 2 stage design really lies in allowing for a narrow frame and canopy.

I own an original Gaui X4 with the two stage drive train in the CF frame and honestly it worked OK. It was never running big power and it sounded like a old farm truck due to the straight cut gears. I never managed to melt a gear, but I did have issues with OWB's dying since they had to be undersized to fit. The canopy tho was rather narrow and that made it pretty darn slippery. It's a bit hard to quantify but despite not having much power it would really get up to speed in a hurry.

I've also got a Diabolo 550 which is also a 2 stage geared design. The quality and design are far better than the X4. The frame is almost exactly as wide as a 480XX but without a wagon wheel of a gear in the middle. This means that the canopy sits maybe a quarter inch off the frame on the sides. The 480XX canopy is a good bit wider to cover that gear. The 550 has the same slippery quality as the X4 did except more so.

The Diabolo S's I believe use a metal intermediate gear for really high power setups to prevent melting.

Extra, just curious, what sort of aerodynamic impacts does canopy width have aside from the obvious frontal area term?

[extrapilot]

Hi Tuscon!

Unfortunately, no easy answer. We operate in a weird Reynolds range, which is right in the general boundary (pun intended) of laminar and turbulent flow. So, even very expensive tools have trouble determining what happens with the flow, made all the worse with the rotor throwing choppy air around, etc.

The simple approach to this is just wetted area- where less skin friction is good, so less canopy area is good. But it gets very ugly if you look at what is actually happening with the flow- where for example in FFF the machine may have a deck angle relative to the flow of 20-50deg-ish. So what may look streamlined (which is rarely a good indication of performance) can look very different from a different flow angle. Same goes for sideways flight, FRF, etc.

With respect to many designers/manufacturers out there, I would wager none has an aerodynamicist on staff (much less a rotary aerodynamicist), or has access to the supercomputers/CFD software required to evaluate this sort of question. It is not difficult to try different canopies and measure speed vs power for a fixed RPM, but it is difficult to determine why design X has less drag- and it may have nothing to do with width, but rather delayed separation of flow, etc. And absent that, you are simply guessing.

[p.weyts]

What do you think off the Henselite TDR and TDR 2 or banchee.

[extrapilot]

Not sure if you are asking for my opinion, or if this is a general question on the thread.

I don’t have an opinion. I don’t know the designers’ background. I don’t know if either team have an aerodynamicist on staff or consulting. I don’t know the designs have ever been in a real wind tunnel.

This is a very complicated problem- and much of the drag we see is not from the fuselage. It is VERY expensive to do this type of design/testing. It can take several tens to hundred of hours on a supercomputer to solve for drag on a helicopter design at a single operating point for a single setup (a particular blade set/pitch/RPM, machine angle in flight, TR RPM, etc). Then, you try 10 other combinations, take what you learn, revise a few things, and go again. It is a huge effort.

You can easily spend hundreds of thousands of dollars designing a fuselage that works fairly well. Unless these people are independently wealthy, or they work at Eurocopter and have friends who run simulations for them, chances are that they just used whatever they learned from car aero etc, and hoped that it would help on the heli. Otherwise, they just could not stay in business making 50-500 machines of a type, where they have many other expenses in developing internal machine layout, setting up tooling and machining for custom parts, in testing, etc.

Cheers

[helicek]

Simple 2 stage for trex 700. My aim was to obtain low rpm(1,250) on main rotor and have a not too low throttle curve. Motor is 570rpm/v from the combo deal.
Ahmet

[Cozz]

I've stumbled across this, after looking for something else, and thought I'd throw in my 2 cents... I queried this a while ago, in regards to the chase 360 being released.

Firstly, flexing under load should not be an issue. The loads will be present, sure, but they should be accounted for in frame stiffness. Take the Gaui X3 for example, the main gear is supported at 3 points, above and below the gear, preventing shaft displacement. The same has been accounted for on the bevel gear (tail), using a really neat design/packaging solution.

In the context of an RC heli, the design of a gear train is (somewhat, there are always other things to consider) constrained to:

Packaging (what can fit within frame/canopy)
Speed (motor-to-blades)
Torque (motor-to-blades)
Weight
Losses (friction & inertia)
Usability (maintenance & ease of building, etc)
Cost (MOST important)

These same considerations apply to tail-drives, and the argument between torque-tube and belt drive.

Never underestimate the effects of marketing; what looks attractive, sells! And this trumps all else in a market such as RC hell's, which is inherently complex, yet pseudoscience usually prevails... Whenever you see an acronym, or technical language, be skeptical...

At the end of the day, "simplicity" wins in ALL cases regarding design. The simple solution is ALWAYS best... and always ask "how much do I need?", rather than thinking more is always best. (this is highly prevalent in the RPM & plastic vs metal parts argument)

[extrapilot]

Uh, no.

Mate, you don’t get to proclaim something as a good design because it ‘should be’ a good design. The 425 should have been a good design- except for a small problem. It wasn’t. It was a terrible transmission/frame design. Doesn’t matter if the X3 has 20 bearings on the main shaft- running helical gears puts axial load on the periphery of the gear itself, and that DOES deform it. When it deforms, it alters the mesh. The question is- how much? And so you end up with heavier gears, and a frame that has to absorb axial loads.

Simple is generally heavy. Simple is generally bulky. There is a reason the pinnacle machines in any category are not simple- from a modern inverter-based TIG welder to an F1 car.

Cost is NOT most important in heli design, and never has been for anything like a competitive acro/3D, endurance, or speed machine. It would be dramatically less expensive and less complex to manufacture frames from stamped steel and blades from wood- but the performance hit is so horrible that it isn’t even plausible to consider.