Fuselage tubing size - shoulder harness analysis

[skip]

Does anyone know the tubing wall thickness of the 7/8" rear-spar carry-through tube just behind the seat?
The fuselage drawing seems to have its Taylorcraft part number as A-1815 at station 4 on the drawing, but the parts list here on the web site only lists completed fuselage frames as parts and doesn't go down to the individual tubes welded together. I can't find the specification anywhere.

My purpose is that I am adding shoulder harnesses. I really want shoulder harnesses. I have a nice new set of FAA/PMA'ed shoulder harness that match my seat belts.

I know that ACE-00-23-561-01 lets us retrofit add on shoulder harness as long as the harness is PMA'ed (mine is), the plane is old enough (CAR4-1938 certainly is), and there is no drilling or welding or cutting on the frame (Thus, a bolt-on split-tube system). This just requires an A&P sign-off instead of a 337 based on the argument that the seat belts alone meet the certification requirements, so add-on shoulder harnesses are minor mods. The ACE says that 337/field-approvals are preferred, but simple sign offs should not be denied.

So it is clear that I can wrap the belts around the tube and be legal, my goal is not just to be legal but to actually keep my face off the instrument panal in a realistic set of potential crashes. Now, I know that some folks out there have just wrapped the harness around the tube, and are a lot safer than with nothing at all. Good for them! I applaud you all for adding whatever you can in the way of shoulder straps. I believe that a good number have escaped bad injury because they added something. But the anal-retentive engineer part of me wonders what the failure modes and strengths really are and what wll really keep my face from the panel in the widest range of incidents. So with some help from a good mechanical engineer, we have been doing some calculations.

I have the 337 here on the web site for the split 1" X .058 tubing that goes over the rear-spar carry through, and am exploring using a similar system but not exactly the same. For example, I want to have a fitting for the harness attach so that the harness can rotate and not have the potential for side loads on the harness when it wraps around the tube.

I have AC21-34 that lists the strength requirements for all the different certification basies out there. CAR 4-1938 only requires 1000 lbs 45 degrees up. That is ridiculously low! it will keep my face out of the panal on a 25 mph ground loop, but most any more substantial crash will exceed that load easily. Also, it wasn't until CAR3 in 1946 that they started requiring G's rather than just load. A much more realistic case.

I am using a pilot load of 200 lbs rather than the FAA standard 170 to be real, and I have assumed that the tube is probably 7/8 X .058 because that's about right for a spar carry-through tube. But to trust the calculations, I really need to know the wall thickness. 7/8 X .049 gives much lower numbers. Thus this request.

Running the numbers with the harness wrapped around the tube just outside the short diagonals shows the tube failing at a fairly miserable 6 Gs or so. That is a fairly low-speed ground loop running into a fence post or something. I have helped my situation in a lot of crashes, but anything serious will still fail the tube and I may eat the instrument panel. So what can I do to make it better?

I looked at a 5/32 swaged cable from the rear spar carry through back to the next station to spread the load back, but there is little room back there and it is really hard to crawl back into the fuselage unless I really need to. I reall don't want to cut and patch fabric for this.

Adding a middle piece of split tube at the center of the spar carry-through tube and a set of 1" X .090 flanges welded to the front of the split-tube bolt-on fittings angled down toward the pilot's shoulders at 45 degrees stiffens the whole thing a lot and also gives a good place to run an AN5 bolt to attach the harness fitting. I don't want to add stuff at the back because of the aileron cables running by there and I don't want to poke up toward the fabric. The flanges down to the front just touch theheadliner in the cabin. The calcs seem to get this up to about 16Gs. That covers most accidents short of flying into a shear cliff at cruise.

Anyway, as with a lot of my previous posts, I have dumped out a lot of my thinking here. Any corrections or additions are welcome.
So does anyone know the wall thickness of the tube?

[Hank Jarrett]

Re: Fuselage tubing size - shoulder harness analysis

I don't know about the other guys, but the engineer in me LOVES it. I just looked at the tube in bending with the loads applied the distance in from a fixed end. I don't even remember the numbers now, but you DO need to remember that you will NOT be applying your full weight to the shoulder straps! Your CG is a couple of inches above your belt buckle, not at the base of your sternum like most people believe. The huge majority of the load will be taken out by the lap belt with the shoulder harness only taking the load applied at your collar bone from you bending at the waist. I think the tube was good for much more than 6G with those assumptions since the force it would actually see was much less.
I wish I had the papers from when we did the GA crash worthiness tests at NASA but they are all boxed up in the hangar. If you are anywhere close I would be happy to go out with you and dig through the boxes. LOTS of interesting stuff from the AGATE project. I would love to see your numbers, but if you think they may be too close to the margin, why not just put a split tube around the spreader to reenforce it? Simple, easy and will open up as it fails, but if you are getting that much force, it will never have to be loaded twice.

Hank

[Bravo 6]

Re: Fuselage tubing size - shoulder harness analysis

Skip,

You certainly have done an admirable amount of homework! Thank you for sharing and asking the question(s). I too just purchased new harnesses. My BC12-D has never had shoulder harnesses and I am about to install them as well. Please share any more info as you work through your process. I am very interested in what you come up with.

[skip]

Re: Fuselage tubing size - shoulder harness analysis

I don't know about the other guys, but the engineer in me LOVES it. I just looked at the tube in bending with the loads applied the distance in from a fixed end. I don't even remember the numbers now, but you DO need to remember that you will NOT be applying your full weight to the shoulder straps! Your CG is a couple of inches above your belt buckle, not at the base of your sternum like most people believe. The huge majority of the load will be taken out by the lap belt with the shoulder harness only taking the load applied at your collar bone from you bending at the waist. I think the tube was good for much more than 6G with those assumptions since the force it would actually see was much less.

Hank,

Thanks for the feedback. I agree that in reality, the shoulder strap and tube probably doesn't take that much. That's one good reason to keep telling everyone we can to put something in even if it just wraps around the tube. But I am kind of thinking of being really foolish and giving my numbers to the FAA on a 337 just to get them out as approved data. The basis that I have been told that they will use is in AC 21-34 that specifies (6.c, page 31) "A forward load distribution 40% to shoulder straps and 60% to lap belt. In addition, lap belt should be able to carry 100%..." Then in the next paragraph 6.d they say that when doing stress analysis on an airframe without original design data, conservative assumptions should be used. So if I go that way, I need to use the 40% load distribution.

I wish I had the papers from when we did the GA crash worthiness tests at NASA but they are all boxed up in the hangar. If you are anywhere close I would be happy to go out with you and dig through the boxes. LOTS of interesting stuff from the AGATE project. I would love to see your numbers, but if you think they may be too close to the margin, why not just put a split tube around the spreader to reenforce it? Simple, easy and will open up as it fails, but if you are getting that much force, it will never have to be loaded twice.

Hank

Unfortunatly, I am out in New Mexico. Hmmm, I have been suggesting to my wife that I should show her the National Gallery and the Air and Space Museum on a vacation...

[M Towsley]

Re: Fuselage tubing size - shoulder harness analysis

What belts/shoulder harnesses did you guys obtain? Thanks.

[skip]

Re: Fuselage tubing size - shoulder harness analysis

What belts/shoulder harnesses did you guys obtain? Thanks.

I have the Aero Fabricators harness sold by Wag Aero and in their catalog (page 52). My installed-in-1991 seat belts are also FAA/PMA Aero Fabricators belts and so the harness matches. Wag's part number is H-702-xxx with xxx for the color. H-701 is the seat belt part number and H-858 is the seat/shoulder combination. The good part is that they are FAA/PMA tagged so they meet one part of the requirement for installation by an A&P as a minor modification as stated in ACE-00-23-561-01.
Funny thing is that Aircraft Spruce resells them and sometimes has a bit better price.

[Hank Jarrett]

Re: Fuselage tubing size - shoulder harness analysis

Before you approach the FAA let's do some Peer Review of the analysis. I really need to find my crash data since it has the latest numbers from actual GA aircraft we drop tested at the Impact Dynamics Facility. at NASA Langley Research Center. I am sure the real crash loads would be better than the FAA "Best Guess" of 60/40.

If you DO get out this way, you need to visit the Military Aviation Museum in Virginia Beach VA. It has the largest privately owned FLYING collection of aircraft from WW-I and WW-II in the world. No Taylorcrafts yet, but I am working on it. ;-)

Hank

[Guest]

Re: Fuselage tubing size - shoulder harness analysis

Hi Skip,

Not trying to push you into anything but want to be sure that you are aware of a no weld solution using a cable loop.

Check out my post on page 2 of this=> http://vb.taylorcraft.org/showthread...lder-harnesses.

I will be at the airfield Thursday and can measure tube thickness that you are looking for.

Dave

[Bravo 6]

Re: Fuselage tubing size - shoulder harness analysis

Marty,

I just bought the same belt sets from Wag Aero that Skip listed.

[skip]

Re: Fuselage tubing size - shoulder harness analysis

Hi Skip,

Not trying to push you into anything but want to be sure that you are aware of a no weld solution using a cable loop.

Check out my post on page 2 of this=> http://vb.taylorcraft.org/showthread...lder-harnesses.

I will be at the airfield Thursday and can measure tube thickness that you are looking for.

Dave

Dave,

Thanks for the followup. Yup, I had seen your hookup in that thread and it is a good one. It also closely matches one method diagrammed in AC 43.13-2A, Acceptable Methods for Aircraft Alterations (not to be confused with 43.13b Acceptable Methods for Inspection and Repair).
Chapter 9 is all about shoulder harness installations and is a good read for anyone wanting to play in this area. Figure 9.16 on page 75 shows a swaged cable loop around tubes much like yours.

I have looked at using a 5/32 swaged cable around the tubes like you did to re-enforce the fitting on the rear spar carry-through tube as described above. It is really a fairly neat solution as it adds plenty of strength in the horizontal plane. The cable does not need to be taught, e.g. with a turnbuckle or anything. I looked at a swaged loop with an AN-111 and an AN-4 bolt to a dual .063 bracked welded to the back of the split-tube fitting. As the tube fails forward, it takes up a minimal bit of slack and doesn't fail any more!

The big problem is that It is really hard to get back to that next station crawling back behind the seat to install a loop of cable. So I prefer trying to trade a little more weight and complexity for not having to crawl back into the fuselage. If I were doing this on an uncovered fuselage, that would be a really good way to do it. But I'm trying to come up with a retrofit that can work on a completed covered airframe. Thus I'm trying to add more split fittings along the spar tube with the dual .090 flanges to add the strength. Talk about my laziness and lack of flexability leading to engineering design tradeoffs... :-)

I'll appreciate it if you manage to get out and find the wall thickness of that 7/8" tube. Just remember that when you hacksaw through the tube so that you can measure it for me, do a good butt-weld and then grind it down nice and clean so the inspector doesn't ever know...

[M Towsley]

Re: Fuselage tubing size - shoulder harness analysis

Thanks guys.

[Guest]

Re: Fuselage tubing size - shoulder harness analysis

Dave,

Thanks for the followup. Yup, I had seen your hookup in that thread and it is a good one. It also closely matches one method diagrammed in AC 43.13-2A, Acceptable Methods for Aircraft Alterations (not to be confused with 43.13b Acceptable Methods for Inspection and Repair).
Chapter 9 is all about shoulder harness installations and is a good read for anyone wanting to play in this area. Figure 9.16 on page 75 shows a swaged cable loop around tubes much like yours.

I have looked at using a 5/32 swaged cable around the tubes like you did to re-enforce the fitting on the rear spar carry-through tube as described above. It is really a fairly neat solution as it adds plenty of strength in the horizontal plane. The cable does not need to be taught, e.g. with a turnbuckle or anything. I looked at a swaged loop with an AN-111 and an AN-4 bolt to a dual .063 bracked welded to the back of the split-tube fitting. As the tube fails forward, it takes up a minimal bit of slack and doesn't fail any more!

The big problem is that It is really hard to get back to that next station crawling back behind the seat to install a loop of cable. So I prefer trying to trade a little more weight and complexity for not having to crawl back into the fuselage. If I were doing this on an uncovered fuselage, that would be a really good way to do it. But I'm trying to come up with a retrofit that can work on a completed covered airframe. Thus I'm trying to add more split fittings along the spar tube with the dual .090 flanges to add the strength. Talk about my laziness and lack of flexability leading to engineering design tradeoffs... :-)

I'll appreciate it if you manage to get out and find the wall thickness of that 7/8" tube. Just remember that when you hacksaw through the tube so that you can measure it for me, do a good butt-weld and then grind it down nice and clean so the inspector doesn't ever know...

Hey Skip,

I went to fly today so it was a good time to get an answer to your question. Yes I had planned to cut the cross tube right in the middle. That is where the stringers hold the fabric the farthest away from the tube so I could have room to patch it back up.

My plan was JB Weld tough so as to avoid welding and heat on the fabric. I have heard so many good things about JB Weld. I figured that I could just make golf ball size glob of it and then stick each end of the cut tube into it and smooth the outside then paint to match the rest of the cross tube.

I think that making it into a ball and then sticking the tube end into it is best. That is because it would fill the inside of the tube with JB Weld adding support and preventing collapse at the tube end. This could prevent a buckling failure (especially since it is at the middle). The smoothing and painting would make it not show so much to future buyers and inspectors.

I ran into a problem though. The local hardware store had no JB Weld for me. There is a guy in town restoring a Stearman. Its a big plane with two wings so I think he must be using up the local supply.

I could have waited until they get their order in on Tuesday. Rather than make you wait for that I decided to use my ultrasonic thickness machine. At first I was concerned that introduction of the high frequency sound into the tube would harden it resulting in fatigue cracks later on. Also was not sure if the high freq. sound could cause the greasy couplant to start on fire.

Well the airplane was outside of the hangar so I look a chance and used ultrasonic thickness machine anyhow.

First I calibrated it on the .250" standard that came with it. Then I measured a new unused piece of 7/8" x .049" from the shop. It measured .046". I am not sure why. But I re-calibrated on the known sample tube to .049". Then I measured my rear cross tube. The aft headliner is out but the forward headliner extends about 8" behind the tube so you have to reach up and past the end of it. Then you have to feel with your fingers and place the transducer. It not really very difficult. I did it 3 times each with new couplant. The readings were in the .036" to .038" range. Each reading was taken when the machine indicated that the transducer was coupled. I suspect the fact the I had to feel what I was doing rather than see and that the tube is painted with 3 coats of paint probably account for the variation. Wicks lists 7/8" tube in .028", .035", .049". I suspect that mine is .035".

Dave

p.s. hopefully the audience can tell where fiction ended and reality started in this post

[Hank Jarrett]

Re: Fuselage tubing size - shoulder harness analysis

Fiction?!?!?!?

Hank

[skip]

Re: Fuselage tubing size - shoulder harness analysis

Hey Skip,

...

Well the airplane was outside of the hangar so I look a chance and used ultrasonic thickness machine anyhow.

First I calibrated it on the .250" standard that came with it. Then I measured a new unused piece of 7/8" x .049" from the shop. It measured .046". I am not sure why. But I re-calibrated on the known sample tube to .049". Then I measured my rear cross tube. The aft headliner is out but the forward headliner extends about 8" behind the tube so you have to reach up and past the end of it. Then you have to feel with your fingers and place the transducer. It not really very difficult. I did it 3 times each with new couplant. The readings were in the .036" to .038" range. Each reading was taken when the machine indicated that the transducer was coupled. I suspect the fact the I had to feel what I was doing rather than see and that the tube is painted with 3 coats of paint probably account for the variation. Wicks lists 7/8" tube in .028", .035", .049". I suspect that mine is .035".

Dave

p.s. hopefully the audience can tell where fiction ended and reality started in this post

Dave,

Thanks Loads!! That is wonderfully helpful!

I remember when the infamous strut AD came out. I dug up our really good old propeller wizard who was certified doing ultrasound inspections for propellers and took him some new sample streamlined 4130 pieces to get his machine calibrated. I remember being impressed at both how good he had to get consistant readings and how flakey the technique could be in the presence of things like paint or crud. Given that data point, I'm really impressed with your data range. You must be good!

I am a little surprised at the .035 instead of .049, but then I know C.G Taylor was really good doing the calculations. Given the geometry of the struts and the fues I can accept that he probably put less force through the rear spar than one might think. Also, that he probably used the little 6" diagonals on that tube to increase the buckling resistance and allow for the .035 which saves a bit of weight and money. He was good at that. My engineer friend and I did the calulations based on .058 tube hoping... but I had really expected .049. We will look more closely at the numbers for an .035 tube.

And by the way, it is a really good feeling to be on an internet forum where one can make comments about grinding welds and gluing with JB-weld without some jerk busting in with a flame thrower screaming about too few smiley faces. I really like you guys!

Dave, I owe you a tube of JB-Weld and a beer next time we run into each other!

[Hank Jarrett]

Re: Fuselage tubing size - shoulder harness analysis

Don't mix the JB-Weld and Beer. Tried it, tastes terrible and REALLY blocks you up!
Hank