Belt Drives and design

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
Belts will continue to improve on the already very impresive streangth.
There has been progress in making artificial spider silk fibres recently.Spider silk is some of the strongest materials on the planet.
What isnt generaly known is that certain spider
silks have the highest heat conductivity of any
material in existance.
This is only relivent to this discusion in the sense that anyone designing a belt drive now,
can look forward to working in a field that is advancing in capabilities and presumably greater aceptance and use.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,766
Location
US
What isnt generaly known is that certain spiders ilks have the highest heat conductivity of any material in existance.
Source, please?
I found one study by one researcher on silk from one species of spider that reported surprisingly high levels of thermal conductivity: much higher than other organic materials, about as good as copper (which is very high), and not as good as silver, graphite, etc. The spider silk is about 25% as thermally conductive as diamond. But, yeah, if this guy's measurements are right, spider silk might be a very good conductor of heat. Magazine piece on the study.
There will certainly be improvements in belts and other materials in time. The available products will be bounded by economics (i.e. what the market demands and will pay for, not what is technically possible) just like everything else. Maybe spiders will work pretty cheap, we can just keep supplying juicy flies.
 
Last edited:

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
Played with Holzer this morning. If you read what both Billski and DanH have written here (and VAF for DanH) then play with the same numbers it is very educational. My drive has a belt that is the most flexible piece in the system. With a MMOI like a prop opposite the 3 cylinder engine it is in conflict with the wishes of the rest of the system. DanH had a belt come apart in flight while doing touch and goes which ran his drive through the worst part of his frequency range multiple times. Ross has talked about his efforts damp out his "worst part" with the flywheel weights. It would take a ton (almost literally, LOL) of weight to get the numbers acceptable by tuning the inertias. The prop dominates the process. Ed Lesher had this happen even with a Lanchester damper. As soon as his friend Molt Taylor told him about the Dodge coupling he went and got one. I think it is time to play with minimizing amplification.
 

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
Here is a question for the group: I am using the friction part of a factory Chevy Metro clutch. This has springs in it. Do I calculate these in like a stiffness? Also does anyone have a figure for Shor A stiffness? My Lovejoy spider talks about angular limits but does not provide any torsional stiffness numbers.
I am still working on figuring driven inertia as well.
 
Last edited:

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
I did new calculations for the attached Holzer where I replaced the prop and prop shaft with the equivalent inertia. My drive ratio is 3.21 and I am using 0.196 lb. ft. for prop MMOI. My lowest omega comes out to 494.5 HZ. I am using a Powerfin example from a Rotax paper to get the prop MMOI.
 

Attachments

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
Source, please?
I found one study by one researcher on silk from one species of spider that reported surprisingly high levels of thermal conductivity: much higher than other organic materials, about as good as copper (which is very high), and not as good as silver, graphite, etc. The spider silk is about 25% as thermally conductive as diamond. But, yeah, if this guy's measurements are right, spider silk might be a very good conductor of heat. Magazine piece on the study.
There will certainly be improvements in belts and other materials in time. The available products will be bounded by economics (i.e. what the market demands and will pay for, not what is technically possible) just like everything else. Maybe spiders will work pretty cheap, we can just keep supplying juicy flies.
Was working from memory,and will see if I can backtrack to the published article and data if its not paywalled.
The recent picture I remember was of little filiment winder and they were discussing there ability to
dial in specific micro crystaline structures in
the "spider web"
The reference to heat conductivity is from two years ago,and again I will see if I can link it.
In any case heat and weight are the enemies
and this seems to address both,however incrementaly and at a remove.
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
Didnt find a reference to the exact claim I mentioned about spider silk,but found something else which is real interesting,which is that when spider silk is stretched,the rebound is dampened because some of the
mechanical energy of stretching is released as heat as it returns to its original length.
Reference below.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,766
Location
US
Didnt find a reference to the exact claim I mentioned about spider silk,but found something else which is real interesting,which is that when spider silk is stretched,the rebound is dampened because some of the
mechanical energy of stretching is released as heat as it returns to its original length.
Reference below.
Just like a rubber band heats up when stretched.
The stretch of spider silk would seem to make it less useful as a belt fiber than existing fibers. For a belt, we do not want a fiber that elongates a lot under load, adding heat to the belt as it expands and contracts.
To get back to the subject at hand: The existing aramid belt fibers are demonstrating that they are up to the task, if loads are kept reasonable. If loads aren't reasonable, then a "super belt" won't help if resonance, TV, etc loads are simply transferred to another susceptible part
 
Last edited:

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
Just like a rubber band heats up when stretched.
The stretch of spider silk would seem to make it less useful as a belt fiber than existing fibers. For a belt, we do not want a fiber that elongates a lot under load, adding heat to the belt as it expands and contracts.
To get back to the subject at hand: The existing aramid belt fibers are demonstrating that they are up to the task, if loads are kept reasonable. If loads aren't reasonable, then a "super belt" won't help if resonance, TV, etc loads are simply transferred to another susceptible part
Ok.
l will ask for a slight extension on this diversion.
The numbers posted about the tensile and ultimate streagth in the article linked suggest that the syretching that does occur,happens at
loads beyond other materials currently in use.
Yes ? no? and that the implication is the thermal properties and behaviors of spider silk ,differ from those currently in use.
This is all usefull as a proxy for learning about
the belt drive I can build now.
Got me thinking that perhaps I should consider
not only cooling of the belt drive,but also thinking about filtering ALL of the air entering
the engine ,drive, and cooling systems,and plan ahead for thermal management based on
which ever component is the most vulnerable to over heating and unessesary thermal cycling.
 

DanH

Well-Known Member
Joined
May 20, 2019
Messages
52
I plan to mount the engine and drivetrain like on a BD-5....
Is there an airframe-specific reason to do it that way? What sort of aircraft is this anyway?

Another highly recommended book (or books): Daniel Raymer's Aircraft Design; A Conceptual Approach, or its lighter brother written for homebuilders.

Here's the thing. If weight and package size are not limiting factors, it's easy to build a redrive. However, the goal here is a good drive with a small footprint and low weight. The only reason to build one with driveshafts and separate mounts for engine and drive is that you have no other choice; the airframe configuration requires it, as is the case with a BD-5.

The only number that I used directly from your most recent list was the stiffness of the belt.
You should not use it. It's just 20 year old data on my hard drive, and it may have been an early guess. Determine the stiffness of the belt you have on hand, or a suitable replacement. That means a test rig in your shop, which will probably take the form of angle vs load, just like any other stiffness, or manufacturer data. They have it.

My hope is that the friction clutch against the flywheel will give me the same result as Jeron had with his dry friction surface in the upper shaft.
Your friction clutch setup is nothing like Jeron's, which was located near the first node, in parallel with the soft element, and purposely made of a material which minimized the delta between static and dynamic friction.

I am using the friction part of a factory Chevy Metro clutch. This has springs in it. Do I calculate these in like a stiffness?
Yes...if you intend to keep it.

Also does anyone have a figure for Shor A stiffness?
The Shore scale is not a measure of torsional stiffness

My Lovejoy spider talks about angular limits but does not provide any torsional stiffness numbers.
Probably because it was designed to be a coupler, not a torsional soft element. It has a torsional stiffness value of course, but nobody intended it as an engineered stiffness product. Lovejoy makes a variety of products that are, so maybe it's time to look at them.
 

DanH

Well-Known Member
Joined
May 20, 2019
Messages
52
Played with Holzer this morning. If you read what both Billski and DanH have written here (and VAF for DanH) then play with the same numbers it is very educational.
That's the idea. Keep moving those stiffness and inertia values to different locations in the system. Then explore some different values.

DanH had a belt come apart in flight while doing touch and goes which ran his drive through the worst part of his frequency range multiple times.
A two-plate style drive purchased from Reductions Inc, in Canada. Circa 1996, advertised as "fully tested", complete with video of test stand running...which in retrospect, didn't include any footage while running at the resonant RPM. All in all, it was a torsional and mechanical disaster. Long gone I think. Back then auto conversions were the wild, wild West.

Silver lining? It did provide significant motivation to learn and design. Watching all the belt teeth sail past the cockpit will do that.

I think it is time to play with minimizing amplification.
Focus. Conceptual design first, then explore frequency while re-arranging stiffness and inertia values which fit the conceptual profile. Get the resonant frequencies out of the operating range, and amplification becomes a non-issue.
 

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
Bede was an interesting example because he tried to create a drive system without understanding the issues while trying to keep planes flying in order to sell kits. That is why I am playing with design before putting an airframe around it. When you mess with the Holzer chart you get a lot of answers that create more questions. Great fun, sometimes. LOL
 

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
I forgot in my last response to DanH to answer a question he has asked twice: What kind of airplane? It will be an original design.
My next question for the group: What is the formula for belt modulus when you have the belt stretch, belt tension and the initial length of the belt being stretched. I tried calling Gates but got a wrong extension and gave my contact info to the person who answered. the Rev. Ron Vanderhart told Gates he was building a sorghum mill. LOL
 

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
My plan, once I have a formula, is use Epi-Eng's info which conveniently gives a good chunk of data and is the same as my belt to give me a better number than the one I was using.
 

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
More Holzer!! I took the time to reread parts of this topic today and want to apologize to all who have had questions that have not been responded to very well. My free time comes on weekends usually. I am still working on calculating how to get belt stiffness in foot-pounds per radian. I get that the sprocket is a lever but I am having a hard time getting from there to belt stretch to belt stiffness. I called Gates but have still not heard back. My cover story is "off road vehicle" LOL. Dan asked me to move the flexible element and I am going to that next but what I just finished was running Holzers with varying crank/flywheel MMOIs
and different prop MMOIs. I did a bifilar on my crank/flywheel combo with both a single flywheel and two flywheels. I found the prop info from a discussion on another website about using Rotax's service letter for MMOI. With a single flywheel and a light prop (Powerfin?) the second omega was 7996 rpm. Two flywheels with a light prop-6908 rpm. One flywheel with heavy prop 6108 rpm and 2 flywheels with a heavy prop 4860 rpm. I was not watching the direction of the vectors for this.
 

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
2,095
Yes it really did only take a few minutes to do this. Thanks Tom Irvine!!
DanH had said to move the "softer" element from one side of the drive to the other. Just for entertainment I just rounded the shafts and belt down one at a time to see what would happen. By the time this becomes intuitive you will be half nuts LOL!
Softening up the piece between the flywheel and the first inertia pushed the second omega to 8504 rpm. The redline on this engine from the factory is 5900. Softening the belt (second shaft) drops the second omega down to 7380 rpm and softening the last shaft from between the big sprocket and the prop pushes the second omega back up to 7992 rpm.
 
Top