[szicree]

I think you "incorrectly assume" that both feet are applying the force to generate the pressure... The last time I checked on RV brake design, the right and left sides are independent systems.

Noah, thanks for the analysis; it makes a lot of sense to this 31-years-in-the-business EE.

I'm making no statement about line pressure, only that 250 lbs with each foot equals a total force of 500 lbs.

Go to the gym, get in a leg press and, with knees bent, press 500 lbs using only the balls of your feet. Good luck.

 

[Noah]

Great Suggestion

Suggest shelving argument and replacing calipers with pressure gages. Fill with 5606, sit in airplane, stand on brakes until either you or the pedals fail. Then you'll know how much you're getting. Empirical. BTW, deforming doesn't necessarily mean failure does it? I think swelling first. Then after deformation they work harden.

This is a great suggestion. Four or five data points like this might reveal something useful instead of the one data point we presently have from Bob. Unfortunately, my plane is not presently in a state of assembly to be able to do this (and I don't have any brake fluid). It's important to note that the load on the rudder cables is only about 1/3rd the force on the pedals. Take a look at where the cable attaches and you will see what I mean. As a result, I don't think that there is much risk of damaging the rudder. After all, this system is designed for the maximum effort of the pilot, right?

Incidentally, I don't think that burst due to overpressure is a failure mode we need to be concerned about, given the lack of such failures reported in the fleet (by a substantial number of commentators in this very thread, no less). The ultimate strength of 3003-O is much higher than the yield strength. My concern is that yielding the tubing stretches and work hardens it and this makes it more brittle, which could lead to increased likelihood of fatigue failure. That and designing a component which exceeds yield strength in normal operation is totally unorthodox, virtually unheard of in engineering. (One exception to this is designing for crashworthiness, whereby materials are designed to bend, yield, and buckle to absorb energy to protect occupants from impact - which obviously doesn't apply here).

 

[Noah]

Basically, it sounds like brake system is being ripped out because of some paper calculations and playing around with a bathroom scale. Wouldn't prudent engineering require some additional bench testing to verify valid input ranges, and that these inputs realize the calculated outputs, especially since field results do not support the calculated conclusion? I suddenly feel the need to go watch an episode of Myth Busters.

BTW, there have been reported failures of the brake lines, usually near the caliper, attributed to over-tightening or over-bending the tubes, resulting in hardening. These resulting field failures have led to a number of mods that have become popular on the rv brake system, including alternate fluids, bolts, and lines.

Thanks for sharing. It's interesting stuff, but perhaps more scientific method can be applied before concluding that the stock system is unsafe.

That's a fair point Don, and I agree with you. I am not going to rip out my brakelines until a more definitive answer can be obtained. To me the remaining unknown is whether I can apply the same load in the airplane seat as I demonstrated with my bathroom scale. The remaining points of the analysis (pedal mechanical advantage, master cylinder area, and tubing stress as a function of fluid pressure) have pretty close to zero uncertainty.

 

[Mel]

To me the remaining unknown is whether I can apply the same load in the airplane seat as I demonstrated with my bathroom scale.

I guess I'm missing something here. There are many things in my airplane that I can break by force but I'm not going to. Why do I need brake lines that will sustain all the force that I can muster? I simply apply the amount of force needed to stop or hold the airplane.

I sincerely hope that you are not building an "A" model. That nose gear can be broken too.

 

[Noah]

I guess I'm missing something here.

Mel, do you, or do you not believe that your airplane should meet the requirements of FAR 23.395? It seems that there are those that do, and those that do not.

 

[Mel]

I do not think that my airplane should necessarily meet part 23! If I did, I would only fly part 23 airplanes.

There are many places where kit built aircraft don't meet part 23. If you have a problem with that, you probably shouldn't be building a kit-built aircraft. You will probably never be comfortable. And there is nothing wrong with that.
Cessna spends a large sum of money to meet part 23. The problem I see is in expecting part 23 compliance without paying for it.

I'm done now. You may have your thread back.

 

[Noah]

That's fine Mel, like I said, there are those that do, and those that don't. But I hope that when a passenger gets in your plane and asks you about the passenger warning which states "this aircraft... does not comply with the federal safety regulations for standard aircraft" that you aren't one of the people who say with a smile, "don't worry about that, I built this myself, and it exceeds them".

'Nuf said on that.

 

[Noah]

Several data points that differ:
Hoop Stress Calculations: Thin wall calculations applicable to material with "inner diameter/wall thickness ratio of at least 10 (often cited as 20)". Tubing in question has ~5.14 ratio (see dimensions below) or ~5.8 using your wall thickness of .032. In either case, the formula used is not appropriate (per wiki reference given in earlier post).

Several astute readers have raised this fair point, use of the thinwall hoop stress equation is a first order approximation and we should expect to see some error as the wall thickness to diameter ratio increases. I assumed that this error was small and would not be enough to offset the huge difference between predicted stress and yield stress in this application.

So, to eliminate any uncertainty raised by the fact that the thinwall assumption might not be valid for this tubing, I am posting the results of a Finite Element Analysis of the tubing completed in Abaqus.

In this analysis, the exact dimensions of the tubing are modeled and a 1 psi pressure is applied to the ID. The resulting Von Mises stresses are plotted. It can be seen that the stress in the tube is 3.90* the fluid pressure. This is very close to the same result predicted using the thinwall equation but using the incorrect outer radius instead of the inner radius. This also confirms the yield pressure of 6000/3.9 = 1539 psi.

3003-0 tubing dimensions:
Measured wall thickness as .035 on tubing supplied by Van's. Consistent with stock from Wag-Aero (SKU L-347-375). Your data indicated .032 wall. I was not able to find 3003 tubing on Spruce web site. Unable to verify '795 psi max working pressure' reference.

Here is a link to the ACS tubing, which indicates "Soft aluminum tubing for instrument air and vacuum lines, fuel and oil lines, and primer lines":
http://www.aircraftspruce.com/catalog/mepages/3003versatube.php

Using Pipe Working Pressure Calculator (http://www.engineersedge.com/calculators/pipe_bust_calc.htm) and this data:
Strength 3003-0: Yield 6000 psi (same as your data), Tensile 16000 psi. (http://www.wilkinsonsteel.com/Aluminum/3003.htm)

I compute the following:
Yield pressure = 2300 psi. ((2300-2713)/2300 ~18% over-stress at your max pressure)
Burst pressure = 6200 psi.

Again, this calculator uses the thinwall assumption which is subject to significant error given this tube geometry as pointed out. In this case the FEA must be used. The FEA shows that fluid pressure at tube yield is 6000/3.9=1538 psi.

 

[Ironflight]

Noah,

I have no doubt that you are much better at analysis than I am - school was just too long ago, and I don't do that kind of detailed engineering work anymore. But I do understand the process, and analysis is only the first step. Next comes testing, to see if the analysis holds true in the real world. That is why we test fly - to prove that the airplane performs like the design analysis says it should.

Several folks have said that they tested pressures in the brake system with actual pressure gages in a completed system. That's testing. They got different results than you did by analysis. Hence, a conundrum. Most likely, the geometry of the real world is different than your assumptions. Or their tests were not precise. the only way you can be sure of your analysis is to set up a test, and measure the results. That's good TOTAL engineering. I, for one, being curious, am always interested in a coherent analysis/test suite to see if things correlate. Just as we shouldn't discount your work before you get a chance to prove it by tests, you probably shouldn't discount other folks test data until you have your own to show. Fair?

(Hope your basement turned out OK - I've lived through a few floods here on the Gulf coast - cleaning up is no fun!)

Paul

 

[szicree]

I guess I'm missing something here. There are many things in my airplane that I can break by force but I'm not going to. Why do I need brake lines that will sustain all the force that I can muster? I simply apply the amount of force needed to stop or hold the airplane.

I sincerely hope that you are not building an "A" model. That nose gear can be broken too.

I guess I'm with Mel. I don't care what the FARs say about it, you can't build a good airplane and still expect it to be unbreakable by the pilot. I could easily break/bend the control stick, quadrant, rudder pedals, flap lever, etc. if I give it my all. Who does this? I can remember the first time I got in a small plane thinking how flimsy everything looked compared to a typical car. A moment's thought revealed that the reason is that cars don't have to fly. I have to assume that when the regs say "maximum effort of the pilot" they mean in flying the plane, not in testing ones physical strength.

I think the number crunching has reached its limit, and now we should just go find a plane, put a gauge on the master cylinder and have at it. Oh wait, Bob already did that.

 

[gasman]

When you start off with numbers that don't fit the problem, the end results can send you in the wrong direction.
Van's pedal ratio is more like 2 to 1. And your pressure on your bathroom wall is not what your pedals would receive.

Matco says its brake systems are designed to use 450 psi. And Grove says that their system is set up for 600 psi.
KITPLANES May 2010 issue.

 

[LarryT]

With all due respect, there is an old engineering expression "one test is worth a thousand opinions." RocketBob has done the test with a calibrated pressure gauge. I have previously posted two "no-brainer" solutions. IMHO, this thread is beating a dead horse. I will be skipping this thread from now on.

 

[9GT]

With all due respect, there is an old engineering expression "one test is worth a thousand opinions." RocketBob has done the test with a calibrated pressure gauge. I have previously posted two "no-brainer" solutions. IMHO, this thread is beating a dead horse. I will be skipping this thread from now on.

I agree. Years ago I was in a build partnership building two Cozy MKIV planes. One was mine and the other was a two man partnership. We met at my shop once or twice a week and put in full days working on both planes. We never even assigned a plane to anyone,,,just going to toss a coin near the end to see who's was who's when the need came. Well the one partner began to question the integrity of everything that was in the plans. Thinking WAY too much about non existing "problems" as he called them and interfering with progress of the build. It got to the point that the partnership had to be dissolved so knowing I wanted a -10, I sold my fuselage to the one partner and kicked the other one out of the shop,,,with his plane! Well the one we finished has been flying two years pretty much squawk free and the other one has been dissected to the point it will probably never fly.

 

[Noah]

Noah,

I have no doubt that you are much better at analysis than I am - school was just too long ago, and I don't do that kind of detailed engineering work anymore. But I do understand the process, and analysis is only the first step. Next comes testing, to see if the analysis holds true in the real world. That is why we test fly - to prove that the airplane performs like the design analysis says it should.

Several folks have said that they tested pressures in the brake system with actual pressure gages in a completed system. That's testing. They got different results than you did by analysis. Hence, a conundrum. Most likely, the geometry of the real world is different than your assumptions. Or their tests were not precise. the only way you can be sure of your analysis is to set up a test, and measure the results. That's good TOTAL engineering. I, for one, being curious, am always interested in a coherent analysis/test suite to see if things correlate. Just as we shouldn't discount your work before you get a chance to prove it by tests, you probably shouldn't discount other folks test data until you have your own to show. Fair?

(Hope your basement turned out OK - I've lived through a few floods here on the Gulf coast - cleaning up is no fun!)

Paul

The basement is finally drying out Paul, so thanks for your and others'
concern about that - the RV is nowhere near the basement so no concern about damage to any of that, thankfully!

I think we're in agreement, Paul. I'm all for more testing on this as I have said a few posts back. It should be noted that the pressure at which the tubing reaches yield stress cannot be determined by test; it must be determined by analysis or closed form equations - unless you can glue a strain gauge inside the tubing which would be pretty difficult to do! And there is no way to determine the appropriate input load except by test.

>Several folks have said that they tested pressures in the brake system with actual pressure gages in a completed system.

I am aware of one post in this thread where somebody offered a single value of measured hydraulic pressure with NO FORCE INPUT given. Was there another? Perhaps I missed it? The only thing we can conclude from this data point, is that it is one data point. Inconclusive. May not represent the input load that I am able to achieve.

 

[Noah]

When you start off with numbers that don't fit the problem, the end results can send you in the wrong direction.
Van's pedal ratio is more like 2 to 1. And your pressure on your bathroom wall is not what your pedals would receive.

Matco says its brake systems are designed to use 450 psi. And Grove says that their system is set up for 600 psi.
KITPLANES May 2010 issue.

Not sure which model plane you have but on mine the perpendicular distance from the brake pedal axis of rotation to the top of the pedal is 5 inches, and the perpendicular distance from the axis of rotation to the master cylinder is 1.5 inches. This yields a mechanical advantage of 5/1.5 = 3.33.

 

[LifeofReiley]

Oh my... I quess we'll quit using the brakes, wait... they don't get used much anyway.

 

[Alan Carroll]

Not sure which model plane you have but on mine the perpendicular distance from the brake pedal axis of rotation to the top of the pedal is 5 inches, and the perpendicular distance from the axis of rotation to the master cylinder is 1.5 inches. This yields a mechanical advantage of 5/1.5 = 3.33.

Noah,

Certainly an interesting discussion in the spirit of "education and recreation"! (even it turns out to be academic).

Maybe worth noting that you don't typically put the balls of your feet on the top edge of the brake pedals (I'm not sure if this is even possible with the RV-7 pedals, without your toes interfering with overlying structure?). If you line up your toes with the top edge, and the balls of your feet are 2 inches below, then then the MA is more like 2:1.

FWIW, below is a link to the Matco technical manual. In it they say that a mechanical advantage of 2.5:1 is required to be able to "easily" reach the required pressure (450 psi) to achieve maximum rated torque of the brake calipers. Of course, if you do this in a taildragger with any forward speed and you'll get a nice close-up view of the runway.

http://static1.veracart.com/matco/item_pdfs/2420/document1.pdf

The manual also describes how to assemble the cylinders, concluding with pressure testing at 800-1000 psi.

 

[rvmills]

Noah,

No engineer here (required caveat), but I've had a self-induced brake failure that I thought I'd describe and then relate it back to your query with a question. I'll apologize in advance, 'cause I'll likely use very non-engineering words, like "pushed way hard"

My failure was at the brake cylinder fitting, shortly after swapping out the original plastic fittings for brass fittings (due to seepage at around the 550 hour TT mark). A day or two after finishing that job, I flew to get my prop dynamically balanced (so you probably see where this is going!). In run 1 of the DB at 2300 RPM on an IO-540, with the tail tied down, and brakes pushed way hard (I warned ya), the right brake fitting gave way and I shut her down. Some of that day was then lost to a reinstall and re-bleed, and the failure was my fault due to poor technique on the tightening of the fitting.

My point in relating that is that with my friend the prop guy standing in front of the plane, I was standing on the brakes, far harder than I do during runups at 1800 RPM, and far, far harder than I do during any real-world braking, and as far as I know, the weak link was the fitting. Now that may not be so for a properly tightened fitting, and can't say what pressure I was generating for the period of time prior to the failure, nor can I say whether my AL lines flexed before the fitting failed, but here's my non-engineer (but an Aero Ops major ) question:

If I read right, you've said that you're more concerned about fatigue failures than burst failures, in that perhaps the standard tubing will flex and work harden under high pressure spikes during normal usage. I've seen various numbers through the thread, but I'll ask, how much pressure would be needed to flex the standard lines enough to cause work-hardening (to any degree), and how many cycles at that pressure would be needed to cause a fatigue failure?

It may be that while the pressure needed could feasibly be reached with the brake pedals, perhaps the number of cycles to failure is astronomically (or just very) high, so as to make it a non-issue, perhaps as born out through the lack of reported failures.

Doesn't mean there's no value in evaluating this, and if you are able to determine a pressure and cycle combination that leads to failure and is achievable in a normal, or even a harsh or abusive, lifetime of braking, then it would be of great value to the fleet. Then again, if the pressure/cycle combo is so large as to be impossible to achieve, then you wouldn't have to yank out all the lines and replace them, and would have a good measure of confidence in the system. Hoping that's the case!

Cheers,
Bob

 

[pierre smith]

Oh me....

....by the time the entire braking system has been exhaustively tested, it'll be time to test the spars, front and rear, to see if they're strong enough.....then the bending forces on the stick, to see if it won't bend....and on...and on...

Let's get 'em in the air....Van has done his job!!

Regards,

 

[jthocker]

....by the time the entire braking system has been exhaustively tested, it'll be time to test the spars, front and rear, to see if they're strong enough.....then the bending forces on the stick, to see if it won't bend....and on...and on...

Let's get 'em in the air....Van has done his job!!

Regards,

Right on Pierre!

Except that I'm not so sure anymore if getting in the air is the goal, endless re/over- engineering and internet discourse seems to be some peoples goals!
Sorry, had to say it! Wish I could be as diplomatic as some others.

 

[F1Boss]

+1

Right on Pierre!

Except that I'm not so sure anymore if getting in the air is the goal, endless re/over- engineering and internet discourse seems to be some peoples goals!
Sorry, had to say it! Wish I could be as diplomatic as some others.

Mutha knows best, and Pierre 'gets it'.

GET BACK TO WORK!
Mark

 

[Akutan Bandit]

...5052-0 and 0.250 diam wih 0.035 wall - printed right on the tubes....

Thx for that. I was wondering what exactly was on the Grumman gear legs and now that you posted this I've been saved more detective work. My lines are old and I'm afraid that they have work-hardened and flexed too much. Easy enough to replace though.

Regarding the comments here on the Grumman mixture cable being from a tractor or lawn mower, don't forget that there is a recurring AD on these now. They weren't Part 23 but the standards have improved on the spam cans somewhat over time.

And yes, I fly a Grumman so you guys can start looking down on me now (literally too since my service ceiling is 11,000').