# Belt Drives and design

### Help Support Homebuilt Aircraft & Kit Plane Forum:

#### plncraze

##### Well-Known Member
Supporting Member
I have been reading the discussion on drive design here and have gotten some of the books and found articles to read but have not found anything besides some brief discussion on Epi-Eng website about belt drives. I understand that a Polychain belt is a stiffness that attaches two other parts with inertias and frequency but cannot find how to connect the two. The articles I have found about geared reduction drives reduce their two different rotating parts to what they call an "equivalent system" where they calculate the speed ratios to create a model of their system. Is there something that explains design with a belt instead of gears?

#### plncraze

##### Well-Known Member
Supporting Member
I have the design guide for my belt. But thanks for the link because Gates spreads their info over multiple publications and you never know what you will find in a new one. I am trying to figure out how to calculate how two different shafts with sprockets will interact with each other when connected with a stiff belt like a Poly chain.

#### dog

##### Well-Known Member
The Gates site states that the new poly chain belt has an equivilant capacity to the same width roller chain. Thats surprising.

Works in longer lengths.

Includes an static conductive version.

Claimed efficiency improvements and a version just for bicycles.

The french buried engine belt driven dual wing mounted prop aircraft in another thead can get a big tech update.

Last edited by a moderator:

#### plncraze

##### Well-Known Member
Supporting Member
Cool! It is easier to find chains in the old text books than kevlar belts LOL

#### plncraze

##### Well-Known Member
Supporting Member
Looking at this problem after cruising through books and online it looks like I could treat the Polychain as a stiffness, like another shaft, calculating it in radians as the other shafts were rather than treating it like a gear.

##### Well-Known Member
Looking at this problem after cruising through books and online it looks like I could treat the Polychain as a stiffness, like another shaft, calculating it in radians as the other shafts were rather than treating it like a gear.
That sounds logical.

#### plncraze

##### Well-Known Member
Supporting Member
Modeling this stuff gets interesting. Bifilar pendulum, calculating shaft stuff, stir, cook, repeat.....

#### rv6ejguy

##### Well-Known Member
Supporting Member
Modeling this stuff gets interesting. Bifilar pendulum, calculating shaft stuff, stir, cook, repeat.....

Yup, did this on my setup. Took me a several weeks to compile all the data and do the calcs.

#### plncraze

##### Well-Known Member
Supporting Member
You are one of the folks I watch here for that. I also read the conversation you had with Dan Horton about torsionals on VAF.

#### rv6ejguy

##### Well-Known Member
Supporting Member
Yeah, Dan is the man on TV and he helped me a lot with my analysis. I couldn't have done it without him.

#### TLAR

##### Banned
I hope this thread goes on for 72+ pages.

How efficient can a belt drive be?

I need a very light weight reduction drive, belt driven, that would facilitate an in flight adjustable propeller.

I stayed @ a Holiday Inn last night

#### plncraze

##### Well-Known Member
Supporting Member
We'll see how long it goes. Cog belts can be very efficient with approximately 1 percent losses but weight depends, in my ignorant opinion, dependent on your engine and is vibration. Its a challenge to even mess with this much less make the darn thing.

#### wsimpso1

##### Super Moderator
Staff member
First stop on this topic is:

The problem in any of these systems is identifying what the spring rates and MMOI are for all of the parts, so that you can run an Eigen solver. But rest assured that there is a chain of masses and springs. I have no experience with these belt drives, but I do know that with some diligence, you can model all of the pieces and get a pretty good idea of the MMOI and spring rates of crankshaft, flywheel, PSRU shafts and pulleys. Other methods are out there such as bifilar pendulum for obtaining MMOI, and will almost have to be used for the prop.

In airplane systems, the biggest single inertia is usually the prop, followed by the engine and any flywheel. You may be able to model most of this in SolidWorks - I have done such things for a long shaft direct drive system on a Jabiru powered system. You have multiple parts in a belt drive that are all pretty darned low inertia compared to the prop and engine, and these same parts are usually tightly coupled to either the prop or engine so maybe they can be safely joined. The springiness under torsion of each of the pieces has to be estimated too.

Then there is the belt. The manual does not seem to provide any information on belt stiffness or spring rate. I strongly suspect that the belt stiffness will require knowledge of the details of the reinforcement. Buy a representative belt and couple pulleys, then figure out a way to load and measure extension of the belt. Once you know this, you can extend the knowledge for your belt width and length between pulleys.

In a system with prop turning slower than the engine, remember that effective inertia of the prop side parts is: Irefi = Ii*SR^2 where Ii is MMOI of part i, SR is speed ratio of part i to engine (with a PSRU, this is less than unity).

Billski

#### TLAR

##### Banned
Maybe the engine of choice should be equipped with a cross plane crankshaft

#### plncraze

##### Well-Known Member
Supporting Member
Thanks Billski! I was hoping you would contribute to this.

##### Well-Known Member
I would try something like

Set up a belt around 2 pulleys.
Fix one pulley to be locked.
Attach a (stiff) lever arm to the other pulley.
Measure the angular deflection when the lever arm is loaded.

This might give you a measure of torsional stiffness.

#### wsimpso1

##### Super Moderator
Staff member
One of the interesting things here is that there are a lot of belt drives out there with no single "soft" element and they seem to work fine. That sort of begs the question of why do all the analysis? Well, in my memory, there were a couple with V belt drives that burned belts and had to be upsized a few times, and some cog belt drives that tended to destroy belts and wear out sprockets. Then there was the original development layshaft arrangement on the BD5 that broke stuff with regularity. After considerable effort, BD came up with a flexible shaft and one-way-clutch that worked fine.

If you size the belt to carry just mean torque from the engine, and firing torques are being transmitted, you can easily overload the parts. In a strictly "stiff" system, like a car engine with its flywheel or a direct drive airplane engine with its prop, the torque capacity of the joint has to be on the order of eight times the mean torque just to carry firing pulses. On the cogbelt schemes, I know it would be safe to design at that level, but is perhaps way overbuilt. There might be some folks who would know empirically what factor above mean torque is needed.

Then there is resonance. I have never run the modeling on a belt drive. I really do not know if they work well because they are so stiff the resonance is off scale high or if the total system between crank flange and prop flange is soft enough to put resonance at or below idle. The best bet for a new design is to come up with a basic scheme, model it, run the Eigen solver, and see where the first torsional mode is and what the loads and stresses look like. Then you can iterate the design while you still have it as electrons in your workstation. The next best solution may be to look at systems that are already known to work well with the engine size and power you want to run, and copy them completely. If you start changing stuff, well all bets are off, and nasty surprises may await in test...

BIllski

#### plncraze

##### Well-Known Member
Supporting Member
Thank you for all the contributions here. I agree with Billski about doing something basic and testing it. I have done the bifilar on the inertias and figuring the spring rates on the shafts. I sized the shaft which comes off the engine for the torsionals that my engine (Geo Metro) might produce so it is very strong. Now I am working on the structure for holding the parts together. I really like the Crosley Mooney reduction drive but we will see. Thanks again!

#### Heliano

##### Well-Known Member
Interesting discussion. I am learning form it. Natural frequencies can be figured out by modal analysis (eigenvalue calculations), but for that to be realistic one MUST know the elasticity of the belt. The method suggested by AdrianS can generate at least rough elasticity data. I do not know if the belt elasticity is linear.
As Billski mentioned one can not base belt sizing on mean torque. Designing a reduction for a piston engine is far more challenging than designing for a turboprop, because piston engines have intermittent torque. The peak torque can easily be two to three times higher than mean torque. Therefore the design must take into account those peak torques. Additionally, intermittent torque means cyclic loads - which produce fatigue. That must be taken into account too. I do not believe that belts are much fatigue-prone, but I would look into it anyway.
Another point related to belt reduction is the temperature effect. If the subject aircraft is to be flown in summer and winter at altitudes up to, say, 15000ft, then it is something the designer must check, to evaluate the need or not of a belt tensioner (hydraulic or mechanical).