Cracks in the Predator’s frame

Today we found two pretty serious cracks in the front of the Predator’s frame.

We found a smaller one on one of the mounts for the front body.

I took pictures but they didn’t show the cracks very well.

The cracks in the front are in the upper curved tube.

I would not expect a crack in this area.

The body mount is at the back of the front body coming off the top tube.

Ed is going to try to get some better pictures tomorrow.

We also found some cracks in the exhaust.

I will not be flying this weekend.

It may put off going to Rotors over the Rockies.

I want it to be ready to fly to San Carlos for The Vertical challenge.

I had already found the crack in the body mount.

I would not have found the two cracks in the frame.

If the tubes cracked all the way through the front landing gear might collapse.

This is a good example of how valuable an annual condition inspection is.

We have both a welder A and a welder B so hopefully she can get welded tomorrow. I may end up having to trailer her to welder C. Smokey is out of town until Monday so he is not one of the plans.

The Predator has 770 hours on her and 130 hours with the IO-320 B1A.

We used a bore scope and the engine appears to be a happy engine.

Leak down was good and the spark plugs look happy.

I have to do some fabric repair on the vertical stabilizer.

I am going to repair the dings in the Propeller.

Ed did a great job of mitigating the rust at the frame joints.

I am going to clean up some wiring now that the body is off.

I found a problem with the connector to the antenna for the transponder. We are going to install a new connector tomorrow.

Thank you, Vance

Vance,

just some general remarks on cracks in welds from an engineering point
of view. Quite often the effects of low amplitude cyclic loads are the
cause for cracks in welds. Any weld considerably weekens the stucture
since the weld cools down very quickly causing the steel to solidify
in an unfavourable microscopic structure. Modern German design codes
assume the admissible cyclic load of a fillet weld to be as low as
27 N/mm^2 ( about 4000 psi) and that this value is independent of the type
of base and weld material. If the weld is stressed a few times very much
higher, near it's yield point, the number of admissible load cycles drops
siginificantly. In picture number #4 it seems as if the crack initited in a spot
where the weld makes a very steep angle with the base material. (At this
point the weld looks a bit as if I had made it .....;-). Such a spot will cause
a so called stress concentration. The stress might be two or three times
higher there than the nominal stress you calculate for the section. From
the above it follows that you have some ways of improving the service life
of your weld:

a) the part may be heat treated after the welding to give the steel a
better microscopic structure.

b) there should be no step or sharp angle between the weld and the base
material. Using the process of Metal Inert Gas (MIG) welding in my
experience gives smooth transition curves for fillet welds. You could
also carefully grind the weld to remove a step (but you don't want to
damage your base material in the process!) and you also want to insure
that the cross sectional area of the weld is not reduced.

c) It has been found that shot peening a weld considerably increases the
service life of the weld due to compression stresses in the outer fibre of
the weld which partly compensate the tension stresses from the load
(tension stresses are more demaging in terms of fatigue than compression
stresses)

Measure a) an b) are equally important and both have a much better effect
than c). Shot peening an untreaded roughly made weld is rubbish.

Hope I'm not carrying coal to Newcastle.

Cheers,

Juergen

No suspension

No suspension

Vance,

Ever thought of putting some kind of small suspension on the front end? Should reduce your stresses on those welds. I believe you had that planned on Mariah Gale...

More coals to Newcastle:

A weld is nothing more than free-form cast steel. Although the weld metal is a mere casting and rather brittle, aircraft tubing is typically so thin that the tube right next to the weld usually cracks or tears out before the weld itself will. Post-heating with a torch is helpful in relieving stresses caused by the shrinkage as the weld cools.

The more conservative EAA types will tell you that only oxyacetylene welding should be used on aircraft-tube structures. In fact, this process burns the adjacent metal more than others (from what I've seen, anyway), but it does heat a larger area to reduce localized shrinkage stress. It also allows you to preheat and favor the larger-thicker piece to spread the heat more evenly than the various arc processes can.

Joints that feature a smaller-diameter tube that "T's" into a larger one are especially prone to tearing out the wall of the larger tube under load. The factory-made tailwheel fork of my Kolb Firestar featured this type of arrangement, using .035 wall tubing. I think it lasted about ten hours before tearing out the wall.

Thank you for the input.

Thank you for the input.

Thank you Juergen, you explain it well.

My experience with thin MIG welded mild steel structures is limited. The reason I say it cracked in a place I didn’t expect is because that tube doesn’t seem particularly loaded. I would expect to find the crack next to the weld. I am a TIG welded 4130 enthusiast.

The Predator was built with mild steel because several people around the build process felt that 4130 tended to break and mild steel was more forgiving of the loads a gyroplane puts on a structure. I don’t know why it was MIG welded. In my experience the MIG wire makes a very hard bead and the base material tends to crack at the edge of the bead. I like to use a softer alloy with TIG welding and match the alloy to the application.

I don’t know why that tube is curved.

I collapsed the landing gear at around 200 hours and built the replacement out of 4130. This may have contributed to the problem because the new landing gear is stiffer and has a higher resonant frequency.

There is no telling how long the crack has been there because I have not had the body off for 450 hours of operation and that area is difficult to see. It took a person with different perspective shining a light through a small hole to find the two front cracks.

I don’t land hard but the nose wheel is very small and hard and without suspension even taxiing puts some shock loads into that tube.

My feet are very close to that tube and the chassis feels very smooth there.

I feel fortunate that we found the cracks before the joint failed.

The body mount on the other hand is where I would expect to find a crack. The tube is unsupported and supports the back of a 45 pound piece of fiberglass. It is easy to feel vibration at that point in flight. There are only two body mounts at the back that connect to the frame and the rest connects to the rear body.

Thank you Toby, I feel suspension would take some of the loads off that tube while landing and taxiing.

I was going to suspend the front but once I learned to land her gently I lost the impetus and other projects like more power seemed more pressing.

It is hard to know about such things because the extra weight may have actually made the problem worse.

Thank you for your input Doug, I was just about to post this but I type slowly.

The Predator is MIG welded mild steel because the people around the build felt the loads imposed by a gyroplane were better managed with mild steel. They had experience with 4130 structures failing on a gyroplane. The structures’ tubing is thin wall. The area that failed at the front of the truss has very little load on it compared to some of the other parts of the frame like the rotor tower or the tail boom. These are the first cracks we have found.

It is interesting to me that both ends of the tube cracked indicating some localized stress that I did not anticipate.

Thank you, Vance

Good catch Vance. Glad it was found and being sorted.

Had to smile at your increasingly elegant landings removing pressure to have nose-wheel suspension.

Fortunate to have friends

Fortunate to have friends

I feel fortunate Leigh; I don’t feel I would have caught it. I am fortunate to have friends like Phil around to make up for my weaknesses.

The paint makes it especially hard to see the cracks and they look a lot like dirt.

It is all about priorities Leigh; more power and a faster climb became more important to us than nose wheel suspension.

Mariah Gale will have suspension and a larger nose wheel for off airport landings, a different priority.

Thank you, Vance

Even if you re-weld the cracks closed they will probably re-crack later. If you examine the fuselage trussing there are no diagonal tubes in the front fuselage area to carry the vertical loads from the nosegear, that means the four longeron tubes supporting the nosegear are acting as cantilever beams and subject to bending loads. This is poor structural design practice.

.

Hope you get it all together for Vertical Challenge so that we can see you there. We'll be working to get two 18A's there (a flyable pair together in the same place is increasingly rare these days) as well as my Bell 47.

Dont overthink it... clean it up, get out the electric glue gun, and zip it up. Done deal.
Probably never crack again.

Get her done!

Get her done!

That is an astute observation Alan. When you see Ed’s pictures that have a better shot of the cracks you may change you mind.

The curved tube is tearing out of the wall of the top tube and it might be that if it had less give it would have cracked sooner. It has not cracked in the parelelagram created by the lack of a diaginal. It does apear that the forces are up on the curved tube at the top and it may be pulling everything together.

Mark designed her with front suspension but it didn’t work out. I suspect the ridged front end and the small hard nose wheel is ultimatly the problem.

Mariah Gale will have a full length Pratt truss of greater section and front suspension.


Hello JR,

I will be there with or without the Predator. Flying up there with the Predator is a priority for me. She is not airworthy now.


I am not far of that thought Michael.

The plan is to use little finger doublers that tie the joint together that we sort of form as we go.

We are trying not to make it too stiff.

I will post some when I get Ed’s pictures so you can see what we are up against.

Thank you, Vance

WaspAir said:

Hope you get it all together for Vertical Challenge so that we can see you there. We'll be working to get two 18A's there (a flyable pair together in the same place is increasingly rare these days) as well as my Bell 47.

Click to expand...

WaspAir. Sometimes what you do in a few spare hours getting ready for Vertical Challenge simply floors me.

Now I have a dilemma trying to finish off my comment. Maybe it is sufficient to let you know I always pick myself off the floor to get ready for the next thing you are doing .

You can't keep me down Wasp Admiration does that sometimes. Thanks.

Arnie

Vance said:

The Predator was built with mild steel because several people around the build process felt that 4130 tended to break and mild steel was more forgiving of the loads a gyroplane puts on a structure. I don’t know why it was MIG welded. In my experience the MIG wire makes a very hard bead and the base material tends to crack at the edge of the bead. I like to use a softer alloy with TIG welding and match the alloy to the application. Thank you, Vance

Click to expand...

Vance I am a non professional welder who has done a lot of TIG and MIG welding . Because I am not rated I always listen very carefully to what the engineers and metallurgists tell me. They always say exactly the same thing you said above.

I learned by breaking things.

I learned by breaking things.

Arnie Madsen said:

Vance I am a non professional welder who has done a lot of TIG and MIG welding . Because I am not rated I always listen very carefully to what the engineers and metallurgists tell me. They always say exactly the same thing you said above.

Click to expand...

Hello Arnie,

You always seem to have something supportive to say.

It is welcome at the end of a long day.

Engineers and metallurgists learned these things from books; I learned these things by building things that broke.

I once imagined that high carbon steel from England was the same as 4130 from the USA. I discovered it was not on the back straight at Riverside Raceway while leading the sidecar race.

My passenger angrily waved the handhold that had just broken off of our new English racing chassis. He settled down when I waved the right handlebar that had just broken off back at him. We still won the race but it was more difficult and with considerable trepidation about what would break next.

I had always wondered why they brazed it instead of welded it. The mystery was solved.

I have Ed’s pictures and some words about today. I will post them as soon as I write the words.

Thank you, Vance

Ed’s pictures!

Ed’s pictures!

The first picture is how it all fits together. There is a large tube that fits into a socket that goes straight and two smaller tubes that angle back to tabs that also have powder coat that is showing signs of stress.

The cracks are in the vertical tube where it is joined by the curved tube that goes between the two frame sides. The cracks are at the bottom of the joint in the vertical tube that is the nose of the truss.

The second picture is the left side of the upper curved tube.

The third picture is the right side of the upper curved tube.

The forth picture is the right body mount.

It has been a challenging week with today being sort of a capper.

The progress on the empennage for Mariah Gale is gratifying.

The things we found on the annual condition inspection have been daunting.

Most is just taking her further apart to have a better look at things.

Each issue needs a fix and not all fixes come easy.

I was not able to find a fabricator that would fix the cracks in the Predator’s frame on Thursday.

I started to try to find a fabricator to repair the Predator’s frame at 6:30am Friday and I finally had a face to face with someone who may have the skills to do the job at 3:30. We are going to work on it on Monday and may have it done sometime Tuesday. Most of the welders don’t want to touch anything aviation related. We found some more cracks as we planned the repair.

Smokey won’t be back until Monday and he is jammed with work so he is probably not viable.

Jim Belland is 6 hours away and busy with Mariah Gale.

The Predator is more apart than together and I still have to manage the wire from the antenna to the transponder and at the end of the day I broke the transponder antenna off the bottom of the body loading it into the truck so it will have to wait until Monday when I can pick up an antenna. The antenna is nearly impossible to install with the front body on.

I am hoping to leave for Utah on Wednesday but I won’t if the Predator is not back together and airworthy because I really want to fly to the Vertical Challenge and the Predator needs to be ready for my biennial flight review on the 25th.

Phil is ok with signing her annual condition inspection off after he inspects the repair and spends some “quality time” with the rotor system. It frustrates him that there aren’t any detailed inspection instructions for the blades and rotor head. He has inspected everything else. I still have a little work to do on the vertical stabilizer and finish up the repairs to the propeller.

We hope to make the sign off happen on Tuesday.

The time he spends on the Predator takes time away from working on Mariah Gale’s empennage.

Ed says I look thrashed and exhausted.

Thank you, Vance

I once imagined that high carbon steel from England was the same as 4130 from the USA. I discovered it was not on the back straight at Riverside Raceway while leading the sidecar race.

My passenger angrily waved the handhold that had just broken off of our new English racing chassis. He settled down when I waved the right handlebar that had just broken off back at him. We still won the race but it was more difficult and with considerable trepidation about what would break next.

Click to expand...

I just blew soda out my nose and all over everything! :lol:
I laughed so hard my sides hurt! Vance can really paint a vivid picture with the written word.
And folks, don't forget, according to Jim Mayfield, his sense of humor has been surgicaly removed! :first:

Alan_Cheatham said:

Even if you re-weld the cracks closed they will probably re-crack later. If you examine the fuselage trussing there are no diagonal tubes in the front fuselage area to carry the vertical loads from the nosegear, that means the four longeron tubes supporting the nosegear are acting as cantilever beams and subject to bending loads. This is poor structural design practice.

.

Click to expand...

Hello Alan,

I have been studying Ed’s pictures and you may be correct about the cause.

The curved corner that broke is part of parallelogram without the brace.

As the parallelogram deforms the bending goes to the corner. The tube held it ridged so it flexed and broke at bottom of the junction.

I stand corrected.

I am not clear on just where the loads are coming from and in which direction but I feel it is a design weakness.

Thank you, Vance

Vance - Ed says I look thrashed and exhausted.

Click to expand...

With that tight schedule....I would think so! Cut yourself some slack Vance, or this will cease to be fun.....not to mention health effects. "Don't worry, be happy" .

Sure glad you found those nasty cracks.....

Vance - I suspect the ridged front end and the small hard nose wheel is ultimatly the problem

Click to expand...

Alan Cheatham - If you examine the fuselage trussing there are no diagonal tubes in the front fuselage area to carry the vertical loads from the nosegear, that means the four longeron tubes supporting the nosegear are acting as cantilever beams and subject to bending loads

Click to expand...

These are two sides of the problem. Any type of spring (that includes a softer nose wheel) would lower the load on the upper cross beam. Yet I'd second Alans very consice engineering description of the problem.
As a temporary measure you could add a wire rope bracing with two turnbuckles to the front frame after you have welded it. The turnbuckles should not be welded to the frame, use wire wrapping or sheet metal strips around the frame. If you then tighten the turnbuckles you take part of the load off the cross beam, tighten until you have somewhat preloaded the frame the other way. In the lower right corner I have tried to depict how to sling the wire around the
T joint of the two tubes.

Cheers,

Juergen

PS: I was not sure whether the lower crossbeam is connected to the strut but that does not matter in this design
PPS: My turn buckles are not to scale

WaspAir said:

Hope you get it all together for Vertical Challenge so that we can see you there. We'll be working to get two 18A's there (a flyable pair together in the same place is increasingly rare these days) as well as my Bell 47.

Click to expand...

I will be there with my SparrowHawk and Bell 47 also. We should make sure to park the gyros together. 3 (hopefully 4) gyros will be a record for VC .

Marc