# Driveshaft development

### Help Support HomeBuiltAirplanes.com:

#### pictsidhe

##### Banned
Unmanned testing would encourage testing of the more dubious designs. I believe there are already regulations for unmanned craft of any size? Is there any reason not to build with dual paperwork? I am fairly sure that switching from unmanned to manned, or vice versa, would involve considerable paperwork. But it could be worthwhile. Lots of projects start with a scale proof of concept flying model.

#### BBerson

##### Light Plane Philosopher
HBA Supporter
The current unmanned rules are limited to 55 pounds, as far as I know.

#### dsigned

##### Well-Known Member
TFF said:
As someone who has flown RC for a long time, flying a homebuilt as a RPV/ drone opens up a lot of unknowns with the government and equipment. I would not trust any off the shelf RC with a real airplane; and I am one who is solid with my RC stuff. Industrial RC equipment is very expensive. A real airplane covers a lot of ground fast; above average control is needed.
I wouldn't do off the shelf either, and I honestly haven't priced industrial RC stuff. I imagine a lot of it would be cobbled together of higher quality servos from different industries until the market catches up.

TFF said:
I also don't want to mix metaphors if you know what I mean. I want drones on one side and flown aircraft on the other. Mixing leads to government pressure on parts that don't need it on either side because safety is now averaged instead of specific. Also if you are not a pilot, you are thinking like a model flyer instead of a prepared pilot. Be a pilot. Don't inch in for what ifs. Do or not do. Also I want to keep flying vehicles and airplanes/ helicopters separate. Autonomous and flying are not the same thing.
I think whether we want to mix metaphors or not is somewhat irrelevant. They are being mixed. Whether we can profit from this and carve out a place for GA is what's at issue. As far as pilot vs. rc flyer: the US military has trained a ton of drone pilots recently, and there are lots of people who have done both RC and flying. I think that the lines are being blurred, and whether that's a good thing or not will depend a lot on how the implementation is handled. As far as regulation: I don't doubt the FAA's ability to make a mess of things, especially when it comes to general aviation. But I think by and large they have followed what people are actually doing. If there is no pressure from an innovative industry, they will have no reason to make regulations to support one.

As far as autonomous vehicles and manned flight, I would again stress that I believe that the lines are already being blurred. Sending unmanned helicopters to retrieve wounded soldiers isn't so far off, even if the most difficult flying will still be manned. For our purposes, the distinction is one of whether or not I am on the airplane and whether I am flying it. Being able to test an aircraft, in the air, before I ever put my butt in it, not to mention having flight controls made of something besides fancy twine.

#### pictsidhe

##### Banned
The current unmanned rules are limited to 55 pounds, as far as I know.
Bigger things can be flown, with over 55lbs of paperwork

#### dsigned

##### Well-Known Member
Bigger things can be flown, with over 55lbs of paperwork
I think that's about the gist of it.

#### BBerson

##### Light Plane Philosopher
HBA Supporter
Drones up to 9oz need not be registered.
From 9oz to 55 pounds it needs to be registered or the user could be fined. Many don't register.
Any drone operations for commercial activity needs remote pilot certificate.

#### wsimpso1

##### Super Moderator
Staff member
Log Member
Ok, I hear two sorts of things on a regular basis, and feel that I have to give my two red cents on the topics.

First is folks talk about how we ought to be able to do torsional vibration measurement dirt cheap. That would be nice. It might piss off Rotec Munich and a couple other folks, but when technology marches on... Trouble is that it has not happened yet. The best we have now is to put either a magnetic or optical target on each of the rotating shafts we think will be interesting, point the appropriate mag pickup, laser tach, or whatever at the target, and record passage time for every high-low or lo-high event on each wheel for the duration of the run. The clock has to have a min clock speed of 10 GHz or more FOR EACH CHANNEL you are recording. This is not everyday computer stuff. Maybe later, but right now you still gotta spend about $70k to the boxes in your desk. Once the data is captured, current laptops can run the rest if you understand what you are doing. Then comes knowing which questions to ask it (which analyses to run, how to display it, etc, so you can figure out if it means anything... And of course, if you grab the wrong shaft/wheels, mount the sensors to the wrong "fixed" points, etc, you may have nothing but gobbledegook. This sort of thing still requires smarts going in and going out to find out if your product is OK or not. Second is this notion that affordable CAE is upon us. Last I checked, the Dynamics package on SolidWorks was a pricey option. Oh sure, the Student version they are letting us EAA members use has it. For one part made of one material, with limited loadings, limited constraints. If you want to build a system with more than one material, more than one part, realistic attachments and constraints, well, you gotta pony up... Then, if you do all of the modeling correctly, mesh it correctly, connect the pieces correctly, let it move in the degrees of freedom it is supposed to, and ask for the right analysis, you might get the Eigen modes and frequencies of the system, and you might be able to figure out what is going on and figure out how to adjust the design to separate the modes, isolate the engine, and in general make it run without tearing itself apart. Once again, it requires smarts going in and coming out to ask the right questions and then make sense of what comes out. The last part is of course to adjust the design so it will work... Do I see this all becoming the realm of laymen using freeware and getting experimental data and modeling that line up and facilitate elegant designs? Let me tell you, with talented, well trained, and well motivated engineers, we only get those results some times... Billski #### BBerson ##### Light Plane Philosopher HBA Supporter That court case was overruled and the FAA reinstated the registration. #### rv6ejguy ##### Well-Known Member wsimpso1 said it before but I'm not sure many people took it in. You COULD be lucky with a TLAR redrive or shaft design but your chances dramatically worsen as you go down in the number of cylinders and get something more flexible between the engine and prop (like a long shaft). You can always build it and try it on the ground running it across the whole rev range and see if it breaks after many hours. That may be good enough to know and lots of folks have done that and been lucky. Lots of others haven't been lucky. We can look at something like the Rotax 9 series engines and see they have a rather severe TV period around 800 to 1200 rpm (depending on prop MMOI) and they simply tell you you MUST idle well above that rpm range. They only have a partial solution in the gearbox with some ramp/ dog/ spring arrangement but it certainly doesn't get rid of this period. During our EFI testing on a steel test stand running 2 different prop types and a club, it was scary how the engine TV coupled into the stand and flexed it down in those rpm ranges (over an inch of deflection!). I wish I still had the video to show in this thread. I've dealt with many customer geared auto conversions over the years in my business and most I've seen have had readily detectable TV issues below 1500 rpm. Some are not too severe while others would certainly break things pretty quickly if you operated there for very long. The human body can't readily detect vibration above about 80Hz so you won't be able to "feel" a potentially destructive vibration caused by firing order excitation within the flight operation rpm range of most auto engines. You need instrumentation to detect this. I didn't have the resources to instrument my engine and drive so I ran a math model after measuring and calculating the MMOI and stiffness of all the components for each element. This is a very time consuming exercise and the results are only as accurate as your initial calcs are. Below you can see some of the parts of my system for reference: Mass moment of inertias of my elements (some measured, some calculated) Redrive element stiffness (some measured, some calculated) My original lightweight flywheel had a measured MMOI of .0386 at 7.875 pounds. The revised one .0752 at 16.06 pounds. Note that the propeller has 11 times the MMOI of even the heavier flywheel and 22 times the MMOI of the original light flywheel. The propeller has by far, the most inertia of any element in the system which is why it's a big deal and the maximum MMOI is often specified on engines by the manufacturer. At least 2 rapid gearbox failures I know of were clearly a result of ignoring this limit and fitting heavy props where no gearbox failures surfaced with the proper, lighter props fitted. The MMOI of an aluminum C/S prop may be triple that of some composite models. Lesson- change the MMOI or stiffness of even one element and you can move from many hundreds of hours of trouble free operation to rapid failure. To get some idea of the forces involved when things hit resonance, my friend's 3 cylinder belt drive peaked at something like 3800 lb/ft. and yes, it disintegrated in flight. Last edited: #### pictsidhe ##### Banned Ok, I hear two sorts of things on a regular basis, and feel that I have to give my two red cents on the topics. First is folks talk about how we ought to be able to do torsional vibration measurement dirt cheap. That would be nice. It might piss off Rotec Munich and a couple other folks, but when technology marches on... Trouble is that it has not happened yet. The best we have now is to put either a magnetic or optical target on each of the rotating shafts we think will be interesting, point the appropriate mag pickup, laser tach, or whatever at the target, and record passage time for every high-low or lo-high event on each wheel for the duration of the run. The clock has to have a min clock speed of 10 GHz or more FOR EACH CHANNEL you are recording. This is not everyday computer stuff. Maybe later, but right now you still gotta spend about$70k to the boxes in your desk. Once the data is captured, current laptops can run the rest if you understand what you are doing. Then comes knowing which questions to ask it (which analyses to run, how to display it, etc, so you can figure out if it means anything... And of course, if you grab the wrong shaft/wheels, mount the sensors to the wrong "fixed" points, etc, you may have nothing but gobbledegook. This sort of thing still requires smarts going in and going out to find out if your product is OK or not.

Second is this notion that affordable CAE is upon us. Last I checked, the Dynamics package on SolidWorks was a pricey option. Oh sure, the Student version they are letting us EAA members use has it. For one part made of one material, with limited loadings, limited constraints. If you want to build a system with more than one material, more than one part, realistic attachments and constraints, well, you gotta pony up... Then, if you do all of the modeling correctly, mesh it correctly, connect the pieces correctly, let it move in the degrees of freedom it is supposed to, and ask for the right analysis, you might get the Eigen modes and frequencies of the system, and you might be able to figure out what is going on and figure out how to adjust the design to separate the modes, isolate the engine, and in general make it run without tearing itself apart. Once again, it requires smarts going in and coming out to ask the right questions and then make sense of what comes out. The last part is of course to adjust the design so it will work...

Do I see this all becoming the realm of laymen using freeware and getting experimental data and modeling that line up and facilitate elegant designs? Let me tell you, with talented, well trained, and well motivated engineers, we only get those results some times...

Billski
Help me with my math?
If I happen to have 45 tooth cam drive gear wheel and the engine is spinning at 6000rpm, the tooth signal is 4.5KHz. 10 bits of time resolution should be overkill? That's 4.5Mhz sampling. Ok, that 45 tooth wheel may not be adequate for high harmonics in a stiff system, but it is surely overkill in a soft system? Yes, i could build a system to do that. But there's an easier way:

Wright did not have 10GHz sampling when they found and fixed the 9th harmonic in their Cyclone. They had an analog mechanical system. I believe they had to make it themselves, though.
If I modernised the Wright system, I'd have a tangential accelerometer rigidly mounted to the shaft, along with an FM transmitter and a small battery. An analog radio receiver (cheap constant latency) would pipe that to a PC soundcard. Yes, I'd need at least two systems and a soundcard channel for each.
Recorded waveforms don't need crunching in real time. Modern electronics are wonderful! The 50th harmonic of 6000rpm would generate a 5kHz signal. A cheapo 44kHz sampling soundcard would not be fazed at all by that. A 192kHz commonly available card should do it easily. Some butchery may be needed to extend the low frequency end for startup and shutdown.

What I will say is that if you need to ask how to do measure and fix TV, you probably won't be able to...

As an aside, while my redrive is still in the doodle ideas and see how the numbers look stage, I have just ordered a bunch of data aquisition stuff for measuring IMEP and pressures in an engine intake system you'd likely laugh at.
Cheap USB soundcards. I will convert these to DC input and I have lots of cheap 22kHz analog inputs. I'm a bit dubious about the latency, but that isn't actually a big problem for my upcoming testing as phasing is not critical.

Last edited:

#### markaeric

##### Well-Known Member
While I won't say there's no purpose in having that kind of sampling rate for tv instrumentation, I certainly don't know what it is. We're talking about a period of .1ns, or put another way, during that time light has traveled just a little bit more than an inch! Most of the greatest feats of engineering in the modern age were accomplished without the aid of instrumentation even close to having that kind of performance. If a non-expert can't design a redrive or driveshaft system for modest engines using well established engineering principals, available information and reasonably buildable instruments 120 years after a 2000HP turbine ship took to the waters, or 50 years after man stepped foot on the moon, then I'm embarrassed by how out of whack my expectations are. It'll likely take a lot longer than what someone with significant experience and access to whiz-bang analytical equipment could do, so if time is of the essence for some commercial project then you hire the experts. The rest of us will chug along, and hopefully succeed with our projects sooner rather than later.

pictsidhe, personally I find optical encoder disks and counting the duration between transitions to be the easiest method to wrap my head around, but they can be tricky to align and mount, and the higher the desired resolution, the more critical that becomes. Your idea seems pretty interesting. Any thoughts on what you'd use as an accelerometer? First thing that crossed my mind was an electret microphone element, but then it dawned on me that it would by its nature, likely drown out what you're interested in with literal noise. Then I recalled this blog post: http://unreasonablerocket.blogspot.com/2015/07/gps-on-several-fronts.html where the author mentions the effects acceleration has on the frequency of a crystal oscillator. I have no idea how difficult it would be to characterize such a sensor, though. A bunch of strange ideas keep whizzing through my head, but then I keep coming back to thinking how an optical system is probably overall the easiest to implement and get pretty good results out of.

#### rv6ejguy

##### Well-Known Member
Strain gauges and optical encoders are some of the most common methods to instrument for TV and are not too expensive these days. Wireless accelerometers are readily available too. If you have a good electronics background you could certainly build your own test equipment for a fraction of the cost of commercial systems. What you also need is the background and understanding to process and interpret that information and IMO, few people without the proper engineering background can do that.

A few years back, I looked at some new stuff to measure TV and there is equipment that could do the job in this application for around $7-12K if I recall. I think they were only 1 and 2 channel though which is not nearly as nice as having 3-4 as you have to repeat tests and synch the data somehow. Frequency response of some types of sensors could be a concern in some cases but engine vibration is moving at a glacial pace compared to even 20 year old small processors. I don't see the latter as being any issue these days. Last edited: #### BBerson ##### Light Plane Philosopher HBA Supporter I think strain gauges is needed. Accelerometers would be overwhelmed by normal vibration. Another option is strain paint. Paint the crank and run it two hours and disassemble and check for strain. If any exists, then repeat to find the rpm... or give up #### dsigned ##### Well-Known Member Accelerometers would be overwhelmed by normal vibration. I wouldn't rule this out, actually. You might get a lot of junk data, but if the accelerometer takes data often enough to get the right frequencies, I wouldn't be surprised if you could filter the data (before it's logged) to make something meaningful out of it. Hmmmm...now I'm curious... #### dsigned ##### Well-Known Member If a non-expert can't design a redrive or driveshaft system for modest engines using well established engineering principals Firstly, I think whether your expectations are out of whack or not is going to be dramatically affected by what you mean by "non-expert"? A mechanical engineer with no graduate work or field experience working with torsional vibration? A non-engineer who can do differential equations? A team of people who can punch numbers into excel? Secondly, I don't know that I would characterize it in binary terms. The success of a project is going to be limited by money, time and ability. Given an unlimited budget, a non-expert could conceivably spend$10 million to contract out the development. But budgets aren't unlimited, and neither is time, so then we look to where our threshold is. For me, I like using senior design projects as my line in the sand. If a team of reasonably talented senior engineering students couldn't accomplish it as a senior design project, then it's probably going to be tough for the average joe to tackle it. The reason I like this as a metric is that senior design projects are generally constrained by time, money and what they can outsource (almost nothing), so (with some notable exceptions) you're not comparing yourself to a multibillion dollar company that can throw fifteen teams of engineers at a problem, has access to all kinds of resources that your average home gamer absolutely does not, and can hire a team of experts for a problem they can't be arsed to learn about.

I'm not sure if I think a driveshaft like this falls above or below that line. I suspect that a senior design team could build something that would function, but maybe not that I would want to trust in the air.

#### Autodidact

##### Well-Known Member
I didn't have the resources to instrument my engine and drive so I ran a math model after measuring and calculating the MMOI and stiffness of all the components for each element. This is a very time consuming exercise and the results are only as accurate as your initial calcs are.
Hi, Ross. As I remember from a thread you started a year or two ago, you had actually hired an engineer who did understand TV, to run that math model for you, as well as try to explain the concept behind the math to you; what was his or her name again?

Bret

#### rv6ejguy

##### Well-Known Member
Dan Horton is not an engineer but he has a better understanding of TV than most engineers do, who are not trained in that area. He instrumented his drive after the first failure and learned that the whole thing was dangerous. He pointed me in the right direction for the Holzer model spreadsheet put out by Tom Irvine and answered my dumb questions along the way. I did all all my own stiffness calcs and MMOI calcs by hand and measured the MMOI of the engine, flywheel and prop parts myself. After I was done Dan said he had spreadsheets for all that! He wanted me to learn it the hard way so it would stick.

Later he did send me a spreadsheet to speed up evaluating different fixes. I could not have done any of this without Dan's guidance.